Submarine optical cable burial depth detection equipment

文档序号：612593 发布日期：2021-05-07 浏览：21次中文

阅读说明：本技术 海底光缆埋深探测设备 (Submarine optical cable burial depth detection equipment ) 是由左名久于 2020-11-23 设计创作，主要内容包括：本发明提供了一种海底光缆埋深探测设备,包括发信机、海底光缆、探头和信号处理与显示器；发信机与海底光缆与电连接,发信机向海底光缆输出低频正弦波；所述探头包括第一线圈、第二线圈和第三线圈,其中第一线圈和第三线圈对称设置于探头的两端,第二线圈位于探头的中心处,三者的中心在同一条直线上；第二线圈与探头同轴设置,第一线圈和第三线圈的轴线与探头的轴线相垂直；探头线在待测的海底光缆所在区域内移动；第一线圈、第二线圈和第三线圈分别将测量到的感应电信号反馈至信号处理与显示器；信号处理与显示器根据上述三路感应电信号确定海底光缆的路由和探头与海底光缆之间的距离。本发明准确确定海底光缆的埋设深度。(The invention provides submarine optical cable buried depth detection equipment which comprises a transmitter, a submarine optical cable, a probe and a signal processing and display, wherein the transmitter is connected with the submarine optical cable; the transmitter is electrically connected with the submarine optical cable and outputs low-frequency sine waves to the submarine optical cable; the probe comprises a first coil, a second coil and a third coil, wherein the first coil and the third coil are symmetrically arranged at two ends of the probe, the second coil is positioned at the center of the probe, and the centers of the first coil, the second coil and the third coil are on the same straight line; the second coil and the probe are coaxially arranged, and the axes of the first coil and the third coil are vertical to the axis of the probe; the probe line moves in the area where the submarine optical cable to be detected is located; the first coil, the second coil and the third coil respectively feed back the measured induction electric signals to the signal processing and displaying device; and the signal processing and display determines the route of the submarine optical cable and the distance between the probe and the submarine optical cable according to the three induced electric signals. The invention accurately determines the burying depth of the submarine optical cable.)

1. An submarine optical cable burial depth detection device is characterized by comprising a transmitter, a submarine optical cable, a probe and a signal processing and display; the transmitter is electrically connected with the submarine optical cable and outputs low-frequency sine waves to the submarine optical cable; the probe comprises a first coil, a second coil and a third coil, wherein the first coil and the third coil are symmetrically arranged at two ends of the probe, the second coil is positioned at the center of the probe, and the centers of the first coil, the second coil and the third coil are on the same straight line; the second coil and the probe are coaxially arranged, and the axes of the first coil and the third coil are vertical to the axis of the probe; the probe line moves in the area where the submarine optical cable to be detected is located; the first coil, the second coil and the third coil respectively feed back the measured induction electric signals to the signal processing and displaying device; and the signal processing and display determines the route of the submarine optical cable and the distance between the probe and the submarine optical cable according to the three induced electric signals.

2. The submarine optical cable buried depth detection equipment according to claim 1, wherein when the probe is horizontally placed in the area above the submarine optical cable to be detected and moves, induced electrical signals measured by the first coil and the third coil sequentially show a dead-point change rule; in the process, when the induced electrical signal measured by the second coil is in a peak value, the signal processing and display device determines that the submarine cable to be measured is positioned under the second coil.

3. The submarine optical cable burial depth detection device according to claim 2, wherein after the second coil of the probe is positioned right above the submarine cable to be detected, the angle of the probe is adjusted so that the second coil is vertically placed right above the submarine cable to be detected and the axial line of the probe passes through the submarine optical cable, and the signal processing and display device calculates the distance from the submarine optical cable to the bottom end of the probe through the induced electrical signals measured by the first coil and the third coil.

4. The submarine optical cable burial depth detection device according to claim 3, wherein when the probe is vertically placed right above the submarine cable to be detected and the axial line of the probe passes through the submarine optical cable, the first coil is located below the third coil, and the signal processing and display device calculates the distance from the bottom end of the probe to the submarine optical cable by using the following formula:

where r1 is the distance from the first coil, i.e. the bottom of the probe, to the submarine cable, ξ_L1Induced electromotive force, ξ, measured for the first coil_L3The induced electromotive force measured for the third coil, and l is the distance between the first coil and the third coil.

5. The submarine optical cable burial depth detection device according to claim 4, wherein in the process that the probe is positioned above the submarine optical cable to be detected and the probe is rotated to be vertically placed, the signal processing and display calculates the distance from the probe to the submarine optical cable in real time according to the induced electrical signals measured by the first coil and the third coil; and when the calculated distance is the minimum value, the distance from the bottom end of the probe to the submarine optical cable is the distance from the bottom end of the probe to the submarine optical cable, and the submarine optical cable is in the axial lead direction of the probe.

6. The submarine optical cable burial depth detection device according to claim 5, wherein the transmitter is used for converting alternating current in commercial power into AC/DC/AC and outputting low-frequency sine waves with adjustable frequency and voltage; the transmitter comprises an SPWM modulation circuit, a drive circuit, a protection circuit and a filter circuit; after alternating current is input, a low-frequency sine wave is output through an SPWM modulation circuit and a filter circuit, wherein a driving circuit generates 4 paths of SPWM waves, and after optical coupling isolation, two IR2110 are used for driving four MOS (metal oxide semiconductor) tubes of the SPWM modulation circuit; the driving circuit adopts a double-loop control strategy that an outer loop is a voltage effective value loop and an inner loop is a voltage instantaneous value loop; the protection circuit uses a pulse absorption circuit to protect the MOS tube; the filter circuit adopts a boosting isolation transformer with leakage inductance; the driving circuit adopts the output voltage of the filter circuit as the effective voltage value and the instantaneous voltage value as the feedback voltage.

7. The submarine optical cable buried depth detection device according to claim 6, wherein the first coil, the second coil and the third coil respectively output induced electrical signals to the one-way data processing module, the two-way data processing module and the three-way data processing module for data processing, and then the induced electrical signals are sent to the display terminal through the RS485 chip in an uplink manner; the three data processing modules are communicated with each other by synchronous signals.

8. The submarine optical cable burial depth detection device according to claim 7, wherein the one-way data processing module, the two-way data processing module and the three-way data processing module are arranged identically, and a hardware system of the device comprises a low noise amplifier, a filter, a sampling module, a DSP and a memory; the induction electric signal is transmitted to a low noise amplifier through matching with a proper capacitor, the amplified electric signal is filtered by a filter to remove a middle-high frequency part, and the rest low-frequency signal is transmitted to a sampling module; the sampling module quantizes the analog signal into a digital signal at a certain fixed sampling frequency, and transmits the data to the DSP at intervals of a short time; the DSP stores the sampled data in a memory, and performs discrete Fourier transform on the data in a short period of time at intervals of a fixed time to obtain the amplitude and the phase of a certain frequency; the DSP controls the gain of the low noise amplifier through the serial port and sends the gain to the display terminal through the RS485 communication module.

9. The submarine optical cable buried depth detection device according to claim 8, wherein the DSP initializes the sampling module, the low noise amplifier, the memory and the RS485 communication module, the DSP obtains quantized data from the sampling module, the data is stored in the data storage area through the AD data processing unit, when FFT calculation is ready, the data in the data storage area is copied to the data calculation area, then a series of FFT calculation is carried out, required amplitude and phase data are transferred to the result storage area and are given to the communication protocol part to interact with the upper computer, and the whole process is finished; the data calculation area generates control signals of a low noise amplifier and a synchronization signal, and data in the data storage area is preprocessed before being copied to the calculation area.

Technical Field

The invention relates to the technical field of submarine optical cable communication, in particular to submarine optical cable buried depth detection equipment.

Background

With the continuous development of communication technology, the wide use of submarine optical cables in the global scope has become a main means of cross-sea communication. Currently, there are millions of kilometers of undersea optical fiber cables throughout the world. In recent years, a large number of submarine optical cable communication lines are built in China coastal areas and used for communication of defense islands, and since submarine optical cables are applied in large quantities, the submarine optical cable communication lines have many safety problems due to various reasons including human and natural factors. On the one hand, natural disasters such as earthquakes can directly cause the interruption of a plurality of submarine optical cables, causing great loss in communication and internet. On the other hand, offshore economic activities are increasingly frequent, and the safety of the submarine optical cable is often endangered when anchoring, fishing operations and various offshore engineering projects are constructed near the submarine optical cable route. In order to protect the submarine optical cable from being damaged by human, the submarine optical cable is buried in a depth of 1-3 meters below the seabed in a buried mode in an area allowed by geological conditions of a geological shallow sea, and the measure effectively reduces the damage frequency of the submarine optical cable.

The laying construction of submarine cables also brings about problems:

firstly, the embedding depth of the submarine optical cable is an important index for determining the engineering quality during the line construction, and due to the lack of effective means, some existing submarine optical cable engineering construction can only be obtained according to the state of the embedded equipment of a construction party, and the fact proves that the submarine optical cable is inaccurate and lacks objectivity, so that construction quality construction units are difficult to master.

Secondly, as submarine pipelines in coastal sea areas are more and more dense, the interference between the newly built pipelines and the existing lines in other construction processes is more and more, and the burying depth of the existing lines needs to be accurately detected in order to protect the existing lines and to smoothly implement the newly built lines.

Furthermore, the depth of burial of the installation line varies due to the variation of the submarine geology, and therefore, the depth of burial needs to be detected even in the maintenance work for burying the submarine optical cable. Once the submarine cable is interrupted, the detection of submarine cable routes (positions) and fault point positions on the vast in the area is a precondition and key to the emergency repair work. This requires accurate detection of the submarine cable route and the point of failure. Meanwhile, the method brings convenience for subsequent maintenance and repair work.

Disclosure of Invention

The invention aims to provide submarine optical cable buried depth detection equipment aiming at the defects of the prior art, and the submarine optical cable buried depth detection equipment can accurately determine the buried depth of a submarine optical cable.

The technical scheme adopted by the invention is as follows: an submarine optical cable burial depth detection device is characterized by comprising a transmitter, a submarine optical cable, a probe and a signal processing and display; the transmitter is electrically connected with the submarine optical cable and outputs low-frequency sine waves to the submarine optical cable; the probe comprises a first coil, a second coil and a third coil, wherein the first coil and the third coil are symmetrically arranged at two ends of the probe, the second coil is positioned at the center of the probe, and the centers of the first coil, the second coil and the third coil are on the same straight line; the second coil and the probe are coaxially arranged, and the axes of the first coil and the third coil are vertical to the axis of the probe; the probe line moves in the area where the submarine optical cable to be detected is located; the first coil, the second coil and the third coil respectively feed back the measured induction electric signals to the signal processing and displaying device; and the signal processing and display determines the route of the submarine optical cable and the distance between the probe and the submarine optical cable according to the three induced electric signals.

In the technical scheme, when the probe is horizontally placed in the area above the submarine optical cable to be detected and moves, the induced electrical signals measured by the first coil and the third coil sequentially show a dead point change rule; in the process, when the induced electrical signal measured by the second coil is in a peak value, the signal processing and display device determines that the submarine cable to be measured is positioned under the second coil.

According to the technical scheme, after the second coil of the probe is determined to be positioned right above the submarine cable to be detected, the angle of the probe is adjusted to enable the second coil to be vertically placed right above the submarine cable to be detected, the axial lead of the probe penetrates through the submarine optical cable, and the distance from the submarine optical cable to the bottom end of the probe is calculated through the induced electrical signals measured by the first coil and the third coil through the signal processing and displaying device.

In the technical scheme, when the probe is vertically placed right above the submarine cable to be tested and the axial lead of the probe penetrates through the submarine cable, the first coil is positioned below the third coil, and the signal processing and display adopts the following formula to calculate the distance from the bottom end of the probe to the submarine cable:

where r1 is the distance from the first coil, i.e. the bottom of the probe, to the submarine cable, ξ L₁Sense of measurement for first coilIn response to electromotive force, ξ L₃The induced electromotive force measured for the third coil, and l is the distance between the first coil and the third coil.

In the technical scheme, in the process that the probe is positioned above the submarine optical cable to be measured and is rotated to be vertically placed, the signal processing and displaying device calculates the distance from the probe to the submarine optical cable in real time according to the induced electrical signals measured by the first coil and the third coil; and when the calculated distance is the minimum value, the distance from the bottom end of the probe to the submarine optical cable is the distance from the bottom end of the probe to the submarine optical cable, and the submarine optical cable is in the axial lead direction of the probe.

In the above technical solution, the transmitter converts the alternating current in the utility power into AC/DC/AC, and outputs a low-frequency sine wave with adjustable frequency and voltage; the transmitter comprises an SPWM modulation circuit, a drive circuit, a protection circuit and a filter circuit; after alternating current is input, a low-frequency sine wave is output through an SPWM modulation circuit and a filter circuit, wherein a driving circuit generates 4 paths of SPWM waves, and after optical coupling isolation, two IR2110 are used for driving four MOS (metal oxide semiconductor) tubes of the SPWM modulation circuit; the driving circuit adopts a double-loop control strategy that an outer loop is a voltage effective value loop and an inner loop is a voltage instantaneous value loop; the protection circuit uses a pulse absorption circuit to protect the MOS tube; the filter circuit adopts a boosting isolation transformer with leakage inductance; the driving circuit adopts the output voltage of the filter circuit as the effective voltage value and the instantaneous voltage value as the feedback voltage.

In the technical scheme, the first coil, the second coil and the third coil respectively output induction electric signals to the one-way data processing module, the two-way data processing module and the three-way data processing module for data processing, and then the induction electric signals are sent to the display terminal in an uplink mode through the RS485 chip; the three data processing modules are communicated with each other by synchronous signals.

In the technical scheme, a single-path data processing module, a two-path data processing module and a three-path data processing module are arranged identically, and a hardware system of the data processing module comprises a low noise amplifier, a filter, a sampling module, a DSP and a memory; the induction electric signal is transmitted to a low noise amplifier through matching with a proper capacitor, the amplified electric signal is filtered by a filter to remove a middle-high frequency part, and the rest low-frequency signal is transmitted to a sampling module; the sampling module quantizes the analog signal into a digital signal at a certain fixed sampling frequency, and transmits the data to the DSP at intervals of a short time; the DSP stores the sampled data in a memory, and performs discrete Fourier transform on the data in a short period of time at intervals of a fixed time to obtain the amplitude and the phase of a certain frequency; the DSP controls the gain of the low noise amplifier through the serial port and sends the gain to the display terminal through the RS485 communication module.

In the technical scheme, a sampling module, a low noise amplifier, a memory and an RS485 communication module are initialized by a DSP, quantized data are obtained from the sampling module by the DSP, the data are stored in a data storage area through an AD data processing unit, when FFT calculation is ready, the data in the data storage area are copied to a data calculation area, then a series of FFT calculation is carried out, required amplitude and phase data are transferred to a result storage area and are given to a communication protocol part to interact data with an upper computer, and the whole process is finished; the data calculation area generates control signals of a low noise amplifier and a synchronization signal, and data in the data storage area is preprocessed before being copied to the calculation area.

The invention comprises three coil probes and three electronic measuring and signal processing units, and finally displays the magnitude of three signals and the distance between the probes and the submarine optical cable and other related information on a terminal through a series of operations. In the aspect of routing, the advantages of the dead spot method and the peak value method are combined, so that the position of the submarine optical cable can be judged more clearly, and the accuracy is greatly improved. In the aspect of depth measurement, the posture of the three-rod probe is accurately adjusted to be right above the submarine optical cable, and the axial lead of the probe penetrates through the submarine optical cable to obtain accurate depth data. The transmitter is characterized by stable frequency, stable voltage, zero internal resistance and pure sine wave (without distortion) voltage waveform. The transmitter is very close to an ideal alternating current power supply, has the characteristics of strong load adaptability, good output waveform quality, simple and convenient operation, small volume, light weight and the like, and has the protection functions of short circuit, overcurrent, overload, overheating and the like so as to ensure the reliable operation of the power supply. The data processing module comprises three paths, so that the microvolt-level electric signals are measured and processed accurately in a targeted mode, the original signals need to be conditioned according to the characteristics of the original signals, the original signals can be collected and processed properly and accurately, and the reason for noise among electronic elements and the reason for random noise generated by a semiconductor are avoided. The noise level of the signal loop is analyzed to further suppress or eliminate the adverse noise and achieve good results through proper device selection and circuit design.

Drawings

FIG. 1 is a schematic structural view of the present invention;

FIG. 2 is a schematic illustration of the placement of the present invention;

FIG. 3 is a schematic view of arbitrary angular probe placement;

fig. 4 is a schematic diagram of a transmitter;

FIG. 5 is a waveform diagram of a bipolar PWM control scheme;

FIG. 6 is a waveform diagram of a unipolar PWM control scheme;

FIG. 7 is a SPWM modulation circuit diagram;

FIG. 8 is a drive signal generation block diagram;

FIG. 9 is a diagram of a driving circuit and a protection circuit;

FIG. 10 is a schematic diagram of the overall design of a multiple signal system;

FIG. 11 is a block diagram of the hardware circuitry of the present invention;

FIG. 12 is a schematic diagram of an amplifier noise model;

FIG. 13 is a cascade system block diagram;

FIG. 14 is an amplifier cascade model;

FIG. 15 is a graph of simulation results of sampling theorem and anti-aliasing filtering;

FIG. 16 is a receiver software flow diagram;

FIG. 17 is a front and back table system implementation schematic;

FIG. 18 is a receiver data flow diagram;

FIG. 19 is a system software hierarchy diagram;

FIG. 20 is a system interface hierarchy diagram;

fig. 21 is an AD data sampling flowchart;

FIG. 22 is a rectangular window frequency domain waveform diagram;

FIG. 23 is a spectral leakage waveform diagram;

FIG. 24 is a graph of improved spectral leakage waveforms;

FIG. 25 is an amplitude frequency response curve of a 4Hz detection signal;

FIG. 26 is a plot of the amplitude-frequency response of the system to a 4Hz detection signal;

FIG. 27 is a plot of the amplitude-frequency response of a 25Hz detection signal;

FIG. 28 is a plot of the amplitude-frequency response of the system to a 25Hz detection signal;

FIG. 29 is a filtered simulation of the 133Hz probe signal;

FIG. 30 is a plot of the amplitude-frequency response of the system to a 133Hz detection signal;

FIG. 31 is a basic experimental schematic;

fig. 32 is a detection coil position schematic.

Detailed Description

The invention will be further described in detail with reference to the following drawings and specific examples, which are not intended to limit the invention, but are for clear understanding.

The current or moving charge generates an electric field in space and also a magnetic field. If the current density in the current field is a function of spatial position only, and not time, in a region, then this region is referred to as a constant current field. And the correspondingly generated electric field and magnetic field are respectively a constant electric field and a constant magnetic field. When a constant current is applied across a thin conductor of uniform cross-section, the electric field distribution will be directly related to the spacing of the conductor and the length of the two ends. At constant current, the electric field profile is also constant.

The magnetic field generated by the constant current is also constant and is therefore also referred to as a static magnetic field. The magnetic induction intensity generated by the infinite vacuum space micro cut-off element Idl at a certain point in the space can be known from the Biao-Sawa law, and the mathematical expression of the magnetic induction intensity is

In the above formula, r is the distance from the line element dl to the point to be investigated, a_rIs the unit vector with the line element pointing in the direction of the point under study. Mu.s₀The permeability of the medium in vacuum is 4 pi x 10^-7H/m。

By using the principle of superposition, the magnetic induction intensity generated at a certain point in space for any closed loop is

The principle is applied to the long straight wire, the line element on the long straight wire which is electrified with the current I generates the magnetic induction intensity dB,

it can be seen that the magnitude of the magnetic strength B is inversely proportional to the distance from the infinitely long straight conductor by the first power. The direction is perpendicular to the plane of the conducting wire.

The relationship between the magnetic induction intensity generated by the long straight conductor and the distance first power is directly related to the depth sounding researched by the text and the position of the submarine optical cable. However, since the magnetic field generated by the constant current is constant, if this relationship is measured with a magnetometer, the position of the submarine optical cable is determined. The method is greatly interfered by environmental factors, and is not suitable for being used as a more ideal measuring method.

Time-varying electromagnetic fields, i.e. electromagnetic fields whose field magnitude varies with time. The electric field generated by the constant current is different from the magnetic field. It is not a conservative field.

As known from the law of electromagnetic induction, magnetic flux is interlinked with a closed conductor loop, and when the magnetic flux changes, an electromotive force is induced in the loop in proportion to the rate of change of the magnetic flux.

The induced electromotive force within the conductor is defined as follows:

where the integration path l is the direction of the induced current when the path is conductive. If the total magnetic flux enclosed by the closed circuit is

The direction of the vector is determined by the closed loop c and the right hand rule. If the surfaces are considered to be fixed in space, the time derivative of the equation applies only to the time-varying magnetic field B. The above formula can now be expressed as:

the above equation is an integral form defined by faraday's law with the stationary loop in a time varying magnetic field. The induced electric field is a non-conservative field because the linear integral of the induced electric field along the closed path is equal to the induced electromotive force.

If a closed loop is fixed, the electromotive force generated when the magnetic induction strength B of the time-varying magnetic field varies with time is called induced electromotive force. Since the magnetic induction B is a function of the position vector r in addition to the time t. When a closed loop moves in a magnetic field at a velocity v, the resulting flux change is:

the electromotive force generated in the ring is referred to as motional electromotive force. If it is usedTo show, we can obtain:

when this closed loop moves within a time-varying magnetic field, the total induced electromotive force is:

from the above equation, a closed loop moves within a time-varying magnetic field. The induced electromotive force generated by the device consists of two parts, namely a motional electromotive force caused by relative motion, and an induced electromotive force caused by the change of magnetic induction intensity. The motional electromotive force and the induced electromotive force are not relative physical quantities, and they are not absolute physical quantities. If the coil is under different reference systems, the induced electromotive force and the motional electromotive force of the coil are different, and the influence of the motional electromotive force on the measurement is ignored. That is, when the wound coil is used as a probe to induce a change in magnetic field intensity, the probe has a small moving speed and generates a small motional electromotive force. The following analysis was performed.

The induced electromotive force is equal to the induced electromotive force generated:

the above equation is derived for a closed loop that is considered essentially stationary. In this case, the magnetic induction B is a function of not only time but also the bit vector r. The location of the magnetic field source can be determined from the method of measuring induced electromotive force in the above equation.

From the expression of the induced electromotive force, the induced electromotive force is directly related to the change of the magnetic field with time. It can be seen that the constant magnetic field distribution does not allow the spatial relationship of the probe and the magnetic field source to be obtained using this simple method of measuring induced electromotive force. Advantageously, in the spatial distribution of the electromagnetic field of the time-varying magnetic field, the magnitude of the induced electromotive force of the closed coil can be conveniently measured electrically for determining the spatial distribution relationship of the closed coil and the magnetic field source. The magnetic field distribution expression for a long straight wire fed with a constant current magnetic field is also applicable to a time-varying magnetic field. The current flowing in the long straight conductor is now no longer constant but a function of time t. And the magnetic induction B is also a function of time t. As is known from the equation, a time-varying magnetic field is generated by applying a time-varying current to a long straight wire. And at point P relative to the spatial position of the long straight conductor. If a closed coil is placed at this point, the induced electromotive force of the closed coil can be obtained as

Wherein the content of the first and second substances,is the perpendicular distance from a point inside the closed coil to the straight conductor.Is the direction vector of the magnetic field strength at a point within the closed coil. If the closed loop area is small relative to the distance to the long straight conductor, negligible unclocking, the closed loop can be abstracted as a particle P.

At this time, ρ andcan be considered as an integral independent value, can be obtained

Wherein the infinitesimal dS is related to the closed coil, complying with the right hand rule. Then the formula can be changed into

Where η is the field strength vectorThe angle to the bin vector dS.

According to the formula, the induced electromotive force xi is in direct proportion to the sectional area S of the closed coil, the change rate of the current in the long straight wire and the cosine value of the included angle between the field intensity vector and the surface element vector. And is provided with

Obviously, the closing coil induced electromotive force is inversely proportional to its distance from the long straight wire to the first power. This is also a physical quantity directly related to the physical location in space. cos η is also a physical quantity related to the orientation of the closed coil relative to the long straight wire. The position determination of the submarine optical cable is to indirectly determine the magnitude of rho by measuring the magnitude of induced electromotive force in the closed coil. The purposes of measuring routing and sounding are further achieved by selecting a certain measuring scheme.

For a certain point P, the distance ρ, the cross-sectional area S, and the angle η are determined, whereas the induced electromotive force ξ is not a determined value, or a regular value. But directly withThe associated value. For this purpose, a suitable law of the change of the I current is selected. The measurement and the judgment of the induced electromotive force are facilitated.

If the time-varying current I in the induced electromotive force of the long straight wire is subjected to Fourier series decomposition when I belongs to an X inner product space, the characteristics of the time-varying current can be effectively known. If I is selected as I_maxsin ω t, then I is also decomposed into a Fourier series in the trigonometric function basis, and with ω selected appropriately, only one term in the Fourier series is obtained as non-zero. If this characteristic is applied to the induced potential, it can be obtained

For c_k＝<I(t)，e^-ikt>The value of the inner product, if selectedThen there is

The above formula can be further simplified into

The induction electromotive force amplitude is obtained by arranging and simplifying the induction electromotive force amplitude into a relational expression that xi changes along time according to cosine

From the above equation, if the current I is a time-varying current with a fixed frequency that varies sinusoidally or cosinusoidally with time, the maximum value of the induced electromotive force sensed by the closed coil is a time-independent quantity. The magnitude of the induction voltage with the required fixed frequency can be directly obtained by a certain technical means and a digital signal processing method. Meanwhile, in the derivation process of the above formula, it can be found that for external interference signals with other fixed frequencies, if the external interference signals are induced by a closed loop to form corresponding voltages, the induced electromotive force under the correct frequency can be obtained by using a Fourier decomposition method. This will be explained in more detail in the implementation section of the following section.

In addition, because the induced electromotive force which can be induced by a single closed coil is very small, if the coil is densely wound by N turns, the obtained induced electromotive force is the sum of all single closed loops. Then, assuming that a current with a frequency of ω and changing according to sine or cosine law is passed through the long straight conductor at this time, the voltage expression of the induced electromotive force under the induction of the N-turn coil is obtained as

It can be seen that if other fixed quantities are ignored, the amplitude of the induced electromotive force is only related to the distance between the probe and the perpendicular of the long straight conductor (i.e. the submarine cable), and the angle between the orientation of the cross section of the probe and the direction of the magnetic field intensity.

The invention relates to a buried depth detection design of a submarine optical cable, which is a comprehensive measurement scheme integrating routing and depth measurement. The measuring scheme comprises three coil probes and three electronic measuring and signal processing units. And finally, displaying the size of the three-way signal, the distance between the probe and the submarine optical cable and other related information on the terminal through a series of operations. In the aspect of routing, the advantages of the dead spot method and the peak value method are combined, so that the position of the submarine optical cable can be judged more clearly, and the accuracy is greatly improved.

As shown in fig. 1, the present invention provides a submarine optical cable buried depth detection device, which is characterized in that the device comprises a transmitter, a submarine optical cable, a probe and a signal processing and display; the transmitter is electrically connected with the submarine optical cable and outputs low-frequency sine waves to the submarine optical cable; the probe comprises a first coil L1 (shown in the figure, the probe 1), a second coil L2 (shown in the figure, the probe 2) and a third coil L3 (shown in the figure, the probe 3), wherein the first coil L1 and the third coil L3 are symmetrically arranged at two ends of the probe, the second coil L2 is positioned at the center of the probe, and the centers of the first coil L1, the second coil L3 and the third coil L2 are on the same straight line; the second coil and the probe are coaxially arranged, and the axes of the first coil and the third coil are vertical to the axis of the probe; the probe line moves in the area where the submarine optical cable to be detected is located; the first coil, the second coil and the third coil respectively feed back the measured induction electric signals to the signal processing and displaying device; and the signal processing and display determines the route of the submarine optical cable and the distance between the probe and the submarine optical cable according to the three induced electric signals.

As shown in fig. 2, when the probe is horizontally placed in the area above the submarine optical cable to be measured and moves, i.e. from the position P1 to the position P3, the induced electrical signals measured by the first coil L1 and the third coil L3 sequentially show a dead-point change rule; in the above process, when the induced electrical signal measured by the second coil L2 is at the peak value, the signal processing and display device determines that the submarine cable to be measured is located right below the second coil.

In the technical scheme, after the second coil L2 of the probe is determined to be positioned right above the submarine cable to be tested, the angle of the probe is adjusted to enable the second coil L2 to be vertically placed right above the submarine cable to be tested and the axial lead of the probe to penetrate through the submarine cable, namely the position P2. And the signal processing and display calculates the distance from the submarine optical cable to the bottom end of the probe through the induction electric signals measured by the first coil and the third coil.

At this time, the axis line passes through the submarine optical cable, and the cross section of the second coil L2 is perpendicular to the direction of magnetic induction at the position, so theoretically, no electromotive force will be induced in the second coil L2 (in fact, very small electric signals will appear). The first coil L1 and the third coil L3 are both disposed in a horizontal orientation, which is the same as the magnetic field strength. From the previous section on the analysis of the induced electromotive force in the horizontal direction, the first coil L1 and the third coil L3 all obey the following expression

At this time, for the first coil L1, the third coil L3 has x equal to 0 and r equal to ρ equal to H. Due to the first coil L1, the third coil L3 must in practice have one near the submarine cable and the other far from the submarine cable, where r is not to be assumed₃＞r₁The following analysis is above this assumption. It can be seen that for the first coil L1, the third coil L3 has

The distance between the first coil L1 and the third coil L3 is a fixed value.

And for two phases have

Lexi L₁And xi L₃All can be obtained by electronic measurement, and can be obtained from the above formula

I.e. the distance from the three-rod probe to the submarine optical cable. The distance of the probe from the submarine optical cable can be obtained by measuring the amplitudes of the induced electrical signals of the first coil L1 and the third coil L3 at the same time and then applying the formula.

If the first coil L1 is used, the third coil L3 is not aligned with the magnetic field but at an angle η, i.e. the axis does not pass through the undersea optical fiber cable. Then the first coil L1 and the third coil L3 induce electromotive force, and the cos η term is counted. Since they are parallel to each other but the axis does not pass through the undersea optical cable, their respective angles are different. As shown in fig. 3.

At this time, the magnitudes of the induced electromotive forces of the first coil L1 and the third coil L3 are set to

And satisfy r₃ cos′₃ir₁ cos′₁L is given by

Can be simplified to

Further reduction is carried out to obtain r1 as,

the distance sought cannot then be calculated using direct measurement methods, since for example η is included₁，η₃. Due to the formula in practical applications when the axis line passes through the undersea optical fiber cable. The numerical relationship between them will be found next. For convenience of comparison, assume here thatThen there is

As can be seen from the figure, η is satisfied₁≥η₃，cosη₃≥cosη₁The equal sign is true only when the axis line passes through the undersea optical cable. Then there is

Also known as the triangle geometric inequality, r₃≤r₁+ l, as applied to the above formula,

due to the fact thatCan be further simplified into

The above formula is at₁＝η₃When the number is equal, the equal sign is established.

It can be seen that the result of the calculation using the formula in the measurement is always greater than the actual r₁The value is large. It is then stated that when the three-rod probe is positioned around the sea floor cable above, the probe is rotated and the minimum distance obtained is the distance from the probe to the sea floor cable. By using the inequality, the method can be used for judging when the axis line of the probe passes through the submarine optical cable, and the calculated value is the actual value. Since the second coil L2 is not in the horizontal orientation at this time, it can be seen that L2 is temporarily not the basis for judgment.

From the above analysis, it can be seen that the distance between the probe and the undersea optical fiber cable can be calculated by applying the formula (#) on data processing and display. When the probe is right above the submarine optical cable, and the axial lead of the three-rod probe passes through the submarine optical cable. The distance obtained if the formula (#) is applied is exactly the distance of the actual probe from the submarine optical cable. When the probe is above the submarine cable, but the probe axis does not pass through the submarine cable, it is offset at an angle around it. Then the resulting distance is calculated using the formulaWill be greater than the actual value r₁. Then if go to probe r'₁To the minimum. The submarine optical cable is in the direction of the axial lead of the probe. And r 'at this time'₁＝r₁. This rule applies to three barsThe posture of the probe is accurately adjusted to be right above the submarine optical cable, and the axis of the probe penetrates through the submarine optical cable. Accurate depth data is obtained.

The transmitter converts the alternating current in the commercial power through AC → DC → AC, outputs the alternating current as pure low-frequency sine wave, and the output frequency and the voltage are adjustable in a certain range. It is different from frequency-changing speed-regulating controller for motor speed regulation and common AC voltage-stabilizing power supply.

The ideal transmitter is characterized by stable frequency, stable voltage, zero internal resistance and pure sine wave (no distortion) voltage waveform. The transmitter is very close to an ideal ac power source.

The transmitter takes a microprocessor as a core, is manufactured in an SPWM mode, is designed by an active element IGBT module, adopts the technologies of digital frequency division, D/A conversion, instantaneous value feedback, sine pulse width modulation and the like, ensures that the capacity of a single machine can reach 350VA, increases the stability of the whole machine by isolating the output of a transformer, has the characteristics of strong load adaptability, good output waveform quality, simple and convenient operation, small volume, light weight and the like, and has the protection functions of short circuit, overcurrent, overload, overheating and the like so as to ensure the reliable operation of a power supply. The transmitter principle is shown in figure 4.

Pwm (pulse Width modulation) is a pulse Width modulation technique, and the theoretical basis is an area equivalent principle, i.e. when narrow pulses with equal impulse but different shapes are applied to a link with inertia, the effect is basically the same. The desired waveform can be equivalently obtained by modulating the width of a series of pulses. The equivalent PWM waveform of a full cycle of a sine wave is shown in fig. 5. According to the area equivalence principle, the sine wave can also be equivalent to the PWM wave in fig. 1, and if the amplitude of the equivalent output sine wave is to be changed, the pulse widths are changed in the same proportion. At present, the inverter circuit with medium and low power almost adopts the PWM technology, the technology is divided into a bipolar modulation technology and a unipolar modulation technology, and the output waveform corresponds to that of fig. 5 and 6.

The output waveform of SPWM (unipolar modulation) is as shown in fig. 6, positive half cycle outputs positive square wave, negative half cycle outputs negative square wave, and it is corresponding to the switching tube that: one high frequency arm and one low frequency arm are a pair, and a complementary switch is arranged between the high frequency arms and a complementary switch is also arranged between the low frequency arms, and only one pair of high frequency switches is used. The output waveform of the bipolar modulation technology is shown in fig. 5, the waveform drives a pair of diagonal switching tubes to be synchronously switched, the upper and lower tubes of the bridge arm are complementary switches except dead time, and 4 power tubes of the bridge arm work at a higher frequency (carrier frequency), so that larger switching loss is generated, and the higher the switching frequency is, the larger the loss is. Unipolar modulation has the characteristic of low loss relative to bipolar modulation.

After optical coupling isolation is performed on the 4 paths of SPWM waves generated in the graph 7, two IR2110 waves are used for driving four single-tube MOSFETs, a double-loop control strategy that an outer loop is a voltage effective value loop and an inner loop is a voltage instantaneous value loop is adopted in the invention, and a unipolar SPWM control mode is realized, as shown in the graph 8, the output power is 500W. And the MOS tube is protected by using a pulse absorption circuit, as shown in figure 9.

The hardware design of the invention comprises the steps of extraction, amplification, filtering, sampling, Digital Signal Processing (DSP) and communication of the induced electromotive force signals. The software design comprises the design of DSP system software, the design of DSP system bootstrap startup and the like. In addition, when a multi-channel signal system is designed, a hardware part design and a software part design which comprise multi-channel data synchronization are also designed.

The overall design of the multi-channel signal system is shown in fig. 10 below. The multi-channel signal system is mainly applied to a three-rod measurement mode. The multi-path signals are extracted from the induction coil, are respectively processed through each path of data, and then are transmitted to a display terminal through RS485 uplink, so that submarine optical cable routing and submarine optical cable depth data are further determined.

The hardware design content of the invention mainly comprises an analog part, a digital logic part and DSP peripheral hardware design. The analog part comprises an induction resonant circuit, a low noise amplifier, a variable gain amplifier, a low pass filter, AD (analog to digital) conversion and a DC power supply circuit. The digital logic part comprises a relay control circuit and a little simple sequential circuit. The DSP peripheral hardware design comprises a DSP core and IO power supply circuit, a DSP and Flash interface circuit. The following will describe the overall structure thereof. As shown in fig. 11:

the signal processing process of the hardware system is clearly illustrated in the figure, and the induced electrical signal is generated by the electrode pair, and the signal is sent to a low noise amplifier (in the figure, chopping amplification and PGA (variable gain amplifier)) by matching the appropriate capacitance. The amplified electric signal is filtered by a switched capacitor low-pass filter to remove the middle-high frequency part, and the rest low-frequency signal is sent to high-precision AD sampling. Then, the AD quantizes the analog signal into a digital signal at a certain fixed sampling frequency, and sends the data to the DSP serial port at intervals.

And storing the AD sampling data by properly configuring a DSP serial port. And discrete Fourier transform is carried out on the data of a short time at intervals of a fixed time to obtain the amplitude and the phase of a certain frequency. And then the data goes through the SPI and UART conversion circuit and is sent to the display terminal through the RS485 communication interface.

The noise model shown in FIG. 12 is used to represent the amplifier; but may also be used to describe transistors, valves, integrated circuit amplifiers, etc.

The diagram includes Vs of the signal source and Rs of the signal source. The amplified noise is fully represented by a voltage source En of zero impedance In series with the input and an infinite impedance current source In parallel with the input and a related system C (not shown). The thermal noise of the signal source is represented by source Et.

To fully understand the effect of the noise sources on the output, all three will be represented by the equivalent input noise Eni. Let the noise-free voltage gain be Av. Then the total noise output is

Thus, it is possible to provide

Definition ofThe transfer function from the signal source to the output is called the system gain Kt. It is different from Av, it is related to the input impedance of the amplifier and the impedance of the amplifier, and varies with frequency. Is provided with

Get the expression of Kt as

The expression of equivalent input noise is obtained by the total output noise Eno and the system gain

Then the equivalent input noise is

This equation is important for the analysis of noise problems. The method is suitable for systems of any active devices. From the above equation, the equivalent input noise is the sum of the mean square values of the three noise sources. However, the noise voltage source and the noise current source may not be completely independent.

The noise model analysis of the amplifier is important for the device selection of the amplifier and the matching of the pre-stage network and the post-stage network of the amplifier. Noise caused by poor connection can be greatly improved. The signal-to-noise ratio of the system is improved.

A noise figure, also called noise factor, is defined. Refers to the ratio of the total effective output noise power to the output noise power generated by the source resistance thermal noise Et. Denoted as F.

In addition to the noise model of the amplifier, the receiver design also knows the contribution of noise in each stage to the noise of the whole system, or the influence degree of noise in the multi-stage system. Thus, a method of how to reduce the system noise in design can be envisioned.

A system block diagram is shown in fig. 13. This consists of an internal noise signal source and a two-stage cascade network. Setting the gain of the network 1 as G1 and the noise as N1; the gain of network 2 is G2 and the noise is N2. Let the equivalent source thermal noise power of the network 2 be G₂kT Δ f. Then the total output noise is

The noise figure of the two-stage cascade is

WhereinTo obtain

Similarly, if a three-stage cascade network is analyzed, there is a noise figure of

It can then be concluded that the noise figure of the cascaded network is mainly affected by the first stage noise when the first stage gain is high. If the network 1 is composed of passive circuit elements, such as a coupling network, with effective power gain less than 1, the noise of the whole system is severely affected by the noise contribution represented by F2. The processing of the noise figure for the first stage is particularly prominent for the noise figure of a multi-stage system. As can be seen from this. For passive filtering to be present in an electronic system, it must be present in a second stage or other system. Otherwise the noise of the system will be severely amplified.

In the amplifying circuit, the introduction of the feedback network can be larger than the widening of the frequency band of the system, and the frequency response of the system is improved. Meanwhile, the stability of the system is greatly improved. Non-linear devices can be made to operate in a linear fashion. However, there is no big improvement in noise for feedback, and a new noise source is added in the introduced feedback network.

A two-stage feedback amplifier is shown. Let A1 and A2 denote the gains of each stage, E1, E2, … and E4 are the noise of each link, and β is the feedback coefficient of the feedback network. Then there are

If for the open-loop amplifier, let A1, A' 2 represent the gain of each stage, the output voltage of the open loop has

V_o＝A₁A′₂(V_in+E₁+E₂)+A′₂E₃+E₄

However, in practical designs, the requirements of the magnification in the design are met regardless of whether feedback is used or not. I.e. the same input-output voltage for the closed-loop and open-loop systems described above. Then there is

A′₂＝A₂/(1+A₁A₂β)

The output voltage of the open loop system has

It can be seen that the feedback does not contribute to the cancellation of the input noise of the amplifier. The noise introduced at the output of the amplifier is attenuated to some extent. Without any improvement in the noise introduced in the amplifier and in the feedback network. In fact, the output noise of the system is increased due to the thermal noise of the feedback network resistor and other elements.

Because the frequency adopted by the invention is in a low frequency range, the low frequency 1/f noise of the electronic noise is particularly prominent. The invention adopts the chopper-stabilized amplifier, and is better for low-frequency signals. However, in the process of actually applying submarine cable routing and depth measurement, because the signal variation range of the induction coil is overlarge, the phenomenon of saturation is formed when the signal is large due to fixed amplification of the front stage. For example, when the induced electrical signal is as small as a few uV and as large as a few tens mV. The difference between the front and the back is about 1 ten thousand times. In view of this, a chopper amplifier is no longer used when the measurement range is large. The gain programmable amplifier is directly used to connect with the LC loop. And ensures peak-to-peak noise levels below 1 uV. This corresponds to a mean square noise of about 175nV/√ Hz.

The gain programmable amplifier (PGA) adopted by the invention is LTC6915, the device takes a switched capacitor as sampling input, middle and high frequency signals are cut off to the maximum extent, and meanwhile, as a feedback amplifying resistance network is integrated in a chip, other equipment can control the gain of the chip through SPI. As applied to the present invention, the PGA gain is controlled using the serial port of the DSP. At higher gain, i.e. smaller signal, the amplifier stage equivalent noise peak-to-peak value is controlled below 1 uV. In the worst case, namely, the signal is relatively large, and the equivalent peak-to-peak noise of the amplifier stage is about 20uV in the range of dozens of millivolts. It can be seen that, for the amplifier, when the amplifier operates at a high gain, the equivalent input noise is small, which is beneficial to extracting the effective signal.

The role of switched capacitor filtering here is to achieve anti-aliasing filtering between AD samples. So that the quantized digital signal can normally reflect the input signal. As to why the switched capacitor technique is selected here, it will be analyzed in the following.

According to the sampling theorem, the sampling signal can be restored without distortion when the sampling frequency band is limited to within half of the sampling frequency. The role of anti-aliasing filtering is to ensure that the frequency band satisfies the sampling theorem. The sampling theorem and the necessity of anti-aliasing filtering are briefly explained below using Matlab.

As shown in fig. 15, x (t) is 11 points at 1024hz sampling rate for a 100hz sine wave, and x2(t) is 11 points at 1024hz sampling rate for a 1124hz sine wave. It can be seen that for the frequency f2 ═ f1+ N × fs/2(fs ═ 1024hz), the spectrum is folded by fs/2 to the lower frequencies. The signal with frequency 1124 can be found in the above diagram to fold at a frequency of 100hz at fs-1024. This is an aliasing phenomenon of the frequency.

To prevent aliasing, an anti-aliasing low-pass filter must be added before the AD sampling. To ensure that the AD samples are sent to the next unit without distortion.

The filtering effect of the traditional RC source filter is controlled by the absolute values of the resistor R and the capacitor C. However, in terms of process, the existence of the absolute value tolerance of the components considerably affects the accuracy and performance of the filter. And for the RC active filter realized by using the integrated circuit, the error of realizing the resistance in the integrated circuit is large (20-30 percent), and the linearity is poor. So that the application of RC active filters is considerably limited. Some have to implement RC active filters in discrete component-plus-op-amp approaches. It can be seen that once the frequency is fixed, it is not easily adjustable. The switched capacitor technology realizes active filtering by using a method of simulating a resistor R by using an MOS switch and a capacitor, and the performance of the switched capacitor technology depends on the ratio of the capacitors and is irrelevant to the absolute value of the capacitors. The switched capacitor filter has high accuracy and the cut-off frequency can be controlled by an external clock.

The invention basically has the requirement on a low-pass filter, and is flat in a low-frequency band, so that a Butterworth fourth-order switched capacitor low-pass filter TLC14 is selected. The input is connected to the output of the programmable gain controller and the output is connected to the input of the AD.

After signal processing and amplification, quantization and further operation are required. The process of quantization is a sampling system in a digital system. The selection of the sampling frequency in a sampling system must satisfy the sampling theorem. Since the present invention requires that 25Hz signals be effectively resolved, the sampling rate must be more than 2 times the maximum. If the sampling rate is chosen to be low, the requirement on the anti-aliasing filter of the previous stage is high. While the accuracy of the AD sampling is not high. In order to meet the indexes of the task, the invention samples the AD with high precision of 24 bits to complete the sampling work.

In chip selection of 24-bit AD, the effective accuracy of many chips is limited by noise and sampling technology, and the accuracy can not reach higher bit number all the time. Wherein the accuracy of the AD data can be improved by using an oversampling technique. The oversampling means that a signal is sampled at a sampling frequency which is several times of the original requirement, so that the frequency band of the pre-filter can be greatly widened, and in addition, the AD chip adopting the oversampling technology averages a certain amount of data during output, thereby further improving the accuracy of the AD data.

In an AD chip that employs oversampling and can average over a certain data, the present invention selects 24-bit ADs 1210. Which is effective at 23 bits of accuracy at a 10hz data output rate (sampled at a fixed 300 khz). At 1khz, the effective precision can reach 20 bits. And it has its own digital filter, self-rectifying, self-zeroing, synchronization capability, and communicates with the DSP using SPI. The use of the measuring device can ensure accurate measurement without error in the mu V level.

The invention uses SPI of ADS1210 to communicate with DSP master slave mode. The ADS1210 is used as a master and sends data to the slave at a fixed 1024hz frequency. And the slave computer (DSP) interrupts the received data by the serial port.

The DSP selected by the invention does not have the function of the ROM or the Flash. That is, the developer cannot store the developed program thereon. The developed program can be stored on the storage chip such as Flash and the like only through the external connection of the DSP according to certain specifications. And in cooperation with the BootLoder function of the DSP, under the condition that the system is powered on, a user program stored in the external Flash is loaded into the RAM of the DSP through a BootLoader program on the DSP chip to operate.

The SST39VF020 is selected, has 256Kx8 storage space and is enough to store 128Kx16 DSP programs and data contents.

The general flow chart of the software design is shown in fig. 16, the quantized data is obtained from the AD, the data is stored in the data storage area through the AD data processing unit, when the FFT calculation is ready, the data in the data storage area is copied to the data calculation area, then a series of FFT operations are performed, the amplitude and phase data required by the subject are transferred to the result storage area, and are handed to the communication protocol part to interact with the upper computer, and the whole flow is finished. However, it also involves the control of the PGA, the control of the synchronization signals, and the preprocessing of the data in the data storage area before copying to the calculation area.

The subject matter has two alternative software framework modes. One is a foreground and background system. The application task is an infinite loop, and the loop calls a corresponding function to complete a corresponding operation, which can be regarded as a background system (background). The foreground is generally an interrupt service routine used to process some applications with high real-time requirements. The other is a multitasking real-time operating system. The multiple tasks are mutually switched and scheduled, and processor resources are fully utilized. In many cases, a task corresponds to a function, and the system kernel switches between running and suspending. The task level is modularized, and the structure is clearer.

Foreground and background system execution is shown in fig. 17. The background is always circularly executed, and the foreground executes the program for the interrupt of the hardware request. The flow chart of the foreground and background system of the DSP is shown in FIG. 18: the system software hierarchy diagram is shown in FIG. 19; the application part is the high level abstract flow seen in the software flow diagram. The hierarchical diagram is an operational interface provided by the DSP system from the hardware layer to the software application layer. For foreground and background systems, all interfaces provided at the system level require the developer to process the interfaces. While developers must also understand the behavior of the following hardware layers, and problems that may occur.

The real-time operating system is different from the foreground and background systems in that the application module can directly use the calling of the operating system and can mutually share the same processor in multitask. The process flow of the implementation using the real-time operating system is the same as that of the foreground and background, except that each task can use a device which is provided by the operating system and is irrelevant to the platform, such as a MailBox mechanism and the like. A system interface hierarchy is shown in fig. 20.

Obviously, compared with a foreground system and a background system, the real-time operating system can freely add the number of tasks, can automatically schedule the tasks among the tasks, shares processor resources, and has simple interfaces and easy operation.

The realization of two software frameworks is compared by combining the design. For example, the RTOS system adopts uCOS-II. Because the time for FFT operation in the task requirement is relatively high, the time interval for two requests of data is within 20 ms. The general outline of the 1024-point 32-bit complex FFT is 1-2 ms. It is also shown that more than 2ms must be allocated to the FFT task to compute within 20 ms. This is poorly controlled for RTOS operating systems. Typically the time slice for operating the system is 10ms, which is reasonable for the subject. If the time slice size is modified, for example, to 100 us. The time for task scheduling of the operating system will continue to increase. In the practical application process, the efficiency and pipeline structure of the DSP are greatly weakened.

Based on the consideration of finishing the FFT operation of the DSP in real time and stably, the invention selects a foreground and background system as a software framework.

The AD data sampling part uses the serial port interruption of the DSP as an entrance of a program, and when data is in the serial port, an interruption is directly generated to the DSP. And pointing to the execution of the serial port interrupt subprogram through the interrupt vector table. As shown in fig. 21. When the 24-bit data of the AD is fetched by the DSP, it is stored in a data storage unit (buffer), and then data preprocessing is performed. For example, whether the data in this period is too large or not is checked. To determine whether the amplified signal is full. Whether the amplified signal falls within a suitable voltage range. If the amplified signal is too large, the data is sent to the PGA, and the amplification factor is reduced. If the amplified signal is too small, the PGA is controlled to increase the amplification factor. Through data comparison in the link, the automatic gain control of the DSP signals is realized.

The automatic gain control is realized by controlling a programmable gain controller of a preceding stage. The analog signal is operated in the most suitable state. The implementation principle is similar to that of a Schmitt trigger, and when the amplitude of a signal is larger than a set maximum value, the amplification factor is reduced and the signal is controlled to be in a proper position. When it is low to the set minimum, it is increased by the magnification. So that it works in a better position. The method can effectively prevent the signal from jumping for many times when the pure threshold system swings around a set value. The number of times of changing the voltage amplification is reduced.

Then, in the submarine cable dead point (point where the signal is zero) method of the present problem, the dead point may become very small in a short time and may become very large. It is desirable that it not jump during the process. Then, the corresponding external environment is solved in some time, and a proper upper threshold and a proper lower threshold are set, and a set value to be jumped when meeting the upper threshold and the lower threshold. The adaptive capacity of automatic gain control is improved.

In addition, when the programmable gain controller switches the gain, the dc output of the programmable gain controller will be different under different gains and will not be proportional to the gain. This is equivalent to applying a step signal to the signal during the FFT computation. This is superimposed on the frequency bins in the frequency domain. Has an impact on the accuracy of the FFT calculation.

FFT processing is a fast algorithmic implementation of the discrete fourier transform. The discrete fourier series is an effective approximation of a continuous fourier series. It is a digital implementation of fourier analysis. Within a certain range of engineering indexes, the method can well reflect the frequency spectrum of the signal. Because the orthogonal basis of the discrete Fourier series is limited, the corresponding frequency spectrum is also in a certain range, and therefore, the phenomenon of frequency spectrum aliasing exists in the actual process. This is also why anti-aliasing filtering is applied in front of the data samples.

The invention adopts 32-bit complex Fourier transform, and the output rate of AD data is 1024 hz. I.e. the corresponding data sampling rate is 1024hz, the spectral resolution can be controlled to 1hz for a 1024-point FFT. And expands the data that is AD-fed into 32 bits. And then carrying out 1024-point complex Fourier transform. A spectrally corresponding 25hz complex number (comprising magnitude and phase data) is stored.

The frequency domain expression of the rectangular window is

As shown in fig. 22. The fourier transform of the signal is equal to the convolution of the original infinite time length signal spectrum with the rectangular window spectrum, which shows that energy leakage occurs. The width of the main lobe is 4 pi/N, and at the 1024Hz sampling frequency of the design, 1024 points of FFT are carried out, and 4 pi/N is 2 Hz. However, due to the influence of the barrier effect, the rectangular window function is 0 at the frequency point of the discrete fourier, and is not shown at this time. Therefore, the FFT spectrum analysis for all integer multiples of the frequency signal would ideally be very accurate. However, if the frequency of the source signal does not fall exactly on the frequency bin of the FFT, the signal frequency, e.g., 50hz, is shifted, e.g., 50.5 hz. This is often the case when a power cable is present near the return cable being tested. As shown in fig. 23. It can be seen that since 50.5hz of signal energy leaks onto 25hz by approximately 0.0009, which is quite close to its original signal magnitude of 0.001. I.e. to illustrate that energy leakage can completely drown out the useful signal. In other words, if there is a 100mV 50.5hz signal from the outside world during the electronic measurement, then it is found that there is always a nearly 0.9mV signal added to the 25hz marine cable measurement. This is not known when the electrical signal is sensed below 1 mV. In fact, it is common to measure signals with a submarine cable induced electrical signal amplitude below 1 mV. As can be seen, the leakage at 1hz for the rectangular window is predominantly near 1/4. About 1/100 at 50hz to 25 hz. It can be seen that the decay rate of the curve is slow and the leakage is severe. Furthermore, if the external interference frequency is around 30hz (with an amplitude of 0.1V), the leakage signal to 25hz is about 1/20, which is 0.0045, and completely overlaps the source signal. These problems are disadvantageous in actual measurement.

In addition, if the sampling frequency Fs is not well 1024hz due to hardware implementation, the frequency points on the time frequency spectrum are no longer integer multiples of hz, and leakage occurs on the external 50hz spectrum at this time. The window function is improved so that it can be attenuated quickly in the spectrum without causing significant leakage in other frequency points.

Other methods of windowing may be used instead in this respect, as shown in fig. 24. It can be seen that for the hanning window function and the spectrum, it is added. The 100mV 50.5hz signal has only 0.425uV leakage at 25hz (note: the hanning window function will attenuate the main resolution by 0.5, so the source signal spectrum here is only 0.5 mV). Overall, this leakage voltage is very small for the source signal. Clearly, the attenuation of the hanning window leakage is fast, while its main resolution is wider.

In the implementation process of the invention, the hanning window is found to improve the leakage noise of the frequency spectrum and improve the accuracy of the frequency spectrum. The performance is very good.

The panning window function employed in the present invention is

Although many window functions may be used to improve these problems, each is suitable for the environment in which it is used. The selected window function varies from environment to environment. All as required by the design. Since the detection frequency used in this design is at 25 hz. The external interference which usually affects the electronic system at most 50 hz. 50 and 25hz have a certain frequency distance. Therefore, the requirement for main discrimination needs to be relaxed, but the attenuation must be larger for a certain bandwidth, so as to effectively prevent the leakage of 50hz to other two frequency points.

The task of the invention is to accurately measure the electric signal of microvoltage level, and the signal is required to be conditioned according to the characteristics of the original signal so that the original signal can be properly and accurately collected and processed, thereby causing noise among electronic elements and causing random noise of a semiconductor. The noise level of the signal loop has to be analyzed to further suppress or eliminate the unwanted noise and to achieve good results with proper device selection and circuit design. According to the invention, weak induction signals from the sensor are subjected to LC resonance and low-noise chopping amplification, and the amplification factor is subjected to program control according to the characteristic of large dynamic range of the detected signals.

In the laboratory, the filtering characteristics of 4Hz, 25Hz and 133Hz are carefully analyzed and the direct current output of PWM is simply tested, and the detection signal processing system simultaneously samples four channels at the moment. The sampling channel tested was optional, channel 1 was used in this experiment, and the rest were shorted (if not stated, its channel results showed less than 10 uV).

When the frequency of the sampled signal is 4Hz, according to the filtering algorithm in the third chapter, when the single-channel sampling rate is 32Hz, an amplitude-frequency response simulation curve of the signal after the signal is subjected to a correlation algorithm (sampling processing time is 1s, and mapping resolution is 0.01Hz) and a Blackman-Harris window interpolation algorithm can be obtained, as shown in fig. 25.

As can be seen from fig. 25, after the 4Hz detection signal is filtered, the peak value at the frequency doubling position does not decay, and the frequency doubling point f is 32n ± 4 (Hz). This is because the sampling rate of the system is 32Hz, and the bandwidth of the spectrum to be examined tends to be infinite, which results in the generation of spectrum mirror effect, and usually requires the addition of low-pass filtering suppression.

In a laboratory, the detection frequency of a submarine optical cable buried depth detection signal processing system is set to be 4Hz, the amplitude value of an output signal of a stable signal generator is set to be 500mV (namely 500000uV), the stable signal generator is connected into the system, then the frequency of the output signal is slowly changed, and the experimental result displayed by a liquid crystal of the system is recorded in an appendix. The data in the appendix are calculated by Matlab, and the actual amplitude-frequency response curve of the system for the 4Hz detection signal can be obtained, as shown in fig. 26.

According to an experimental curve, when the frequency is 4Hz, the maximum value of 507540uV can be obtained, the peak value oscillation at the frequency doubling position is attenuated due to the fact that FIR low-pass filtering is added into the system, and the frequency doubling peak value interference is almost completely attenuated to 0 after 1 KHz. Meanwhile, a Blackman-Harris window interpolation algorithm is added, and the sidelobe attenuation at the frequency doubling position is obviously increased, but the main lobe is widened. In the experiment, the difference between the maximum peak value and the input value is the problem of the correction coefficient of the sampling channel, because the signal generators used in the debugging and the final experiment are different, and the error between the signal generators causes the correction coefficient of each sampling channel to be different. Because of the problems of time and equipment, the calibration is not carried out again in the experiment, but the measurement analysis of the system is not influenced, and the same reason is not repeated. And when the stable signal frequency is 4Hz, the input amplitude of the signal generator is 500mV, and due to the temperature drift effect of the signal generator, the jitter display error of the system is about dozens of uV at the moment, but when the input is changed into 1uV, the jitter display error is 1uV, the precision reaches about 20 bits, and the design requirement is met.

When the frequency point of the sampled signal is 25Hz, according to the filtering algorithm, when the single-channel sampling rate is 200Hz, we can obtain an amplitude-frequency response simulation curve of the signal after the signal passes through a correlation algorithm (sampling processing time is 1s, and mapping resolution is 0.01Hz) and a Blackman-Harris window interpolation algorithm, as shown in fig. 27, the curve also has a spectrum mirror effect, and the frequency multiplication point f is 200n ± 25 (Hz). In a laboratory, the detection frequency of a submarine optical cable buried depth detection signal processing system is set to be 25Hz, the amplitude value of an output signal of a stable signal generator is set to be 500mV (namely 500000uV), the stable signal generator is connected into the system, then the frequency of the output signal is slowly changed, and the experimental result displayed by a liquid crystal of the system is recorded in an appendix. By Matlab calculation, the actual amplitude-frequency response curve of the system to the 25Hz detection signal can be obtained, as shown in fig. 28.

According to an experimental curve, the frequency doubling points of a 25Hz detection signal are far apart due to the fact that the sampling frequency is high, the low-pass filtering effect is good, and the attenuation of frequency doubling peak interference after 1KHz can be ignored. Also when measuring at a fixed frequency, the jitter error of the first sampling channel is found to be several tens of uV due to the temperature drift effect of the signal generator. And the minimum resolution voltage is 1uV, the system precision reaches 20 bits, and the design requirement is met. But channel two always had a signal perturbation of 30uV present, which was not present in the experimental analysis of the 4Hz detected signal. Probably because the sampling rate of 25Hz switching four channels is as high as 800Hz, and the sampling rate of 4Hz switching four channels is 128Hz, the fast switching rate of the former may cause the ADS1256 to carry a small amount of charges when switching the channel switch, which may generate common mode voltage interference.

When the frequency point of the sampled signal is 133Hz, according to the filtering algorithm, when the single-channel sampling rate is 532Hz, the amplitude-frequency response simulation curve of the signal after the signal passes through the correlation algorithm (sampling processing time is 1s, and mapping resolution is 0.01Hz) and the Blackman-Harris window interpolation algorithm can be obtained, as shown in fig. 29,

the 133Hz detection signal also has a spectrum mirror effect, the frequency multiplication part is not attenuated, and the frequency multiplication point f is 532n +/-133 (Hz). In a laboratory, setting the detection frequency point of a submarine optical cable buried depth detection signal processing system to be 133Hz, and the amplitude value of an output signal of a stable signal generator to be 500mV (namely 500000uV), accessing the stable signal generator into the system, then slowly changing the frequency of the output signal, and recording the experimental result displayed by a liquid crystal of the system. By Matlab calculation, we can obtain the actual amplitude-frequency response curve of the system to the 25Hz detection signal, as shown in FIG. 30, the 133Hz detection signal has larger sampling frequency, so the frequency doubling points are far apart, the low-pass filtering effect is also better, and the attenuation of the frequency doubling peak interference after 3KHz can be ignored. When the frequency is measured at a fixed frequency, the jitter error of the first sampling channel is found to be dozens of uV due to the temperature drift effect of the signal generator, and the system precision can reach 20 bits, so that the design requirement is met. But channel two has signal interference of 30uV because of common mode voltage interference.

In the present system, the PWM can be programmed to convert a single channel signal amplitude processing value or a function of multiple channel amplitude processing values to a dc output. The PWM end is connected with a mechanical meter head at the same time, and the direct current output size is represented by the rotation of the pointer, so that visual display is provided for reflecting the strength of the detection signal. In the system, a detection frequency point is set to be 25Hz, and the PWM direct current output represents the input voltage of the sampling channel 1. In a laboratory, a signal generator is connected into a system, a PWM is connected with a 2K omega resistor load in a connecting mode, a voltmeter is connected in parallel, the output frequency of a signal source is 25Hz, the effect of PWM direct current output is observed by changing the amplitude of the signal, and the experimental result is shown in table 1.

Table 1 PWM output test table 1

The output amplitude of the stable signal generator is 500mv, the frequency is changed, and the measured experimental results are shown in table 2.

Table 2 PWM output test table 2

The experimental result shows that when the frequency of the input signal is stabilized at 25Hz, the direct current output of the PWM is approximately in direct proportion to the amplitude value of the input signal, the intensity change of the input signal can be reflected, and the design requirement is met; when the amplitude value of the input signal is stable, the direct current output of the PWM can be maximum at 25Hz, then the direct current output is attenuated along with the diffusion of the frequency to the left side and the right side of 25Hz, the right side cut-off bandwidth is 29Hz, the left side cut-off bandwidth is 21Hz, the PWM load output of the side lobe is obviously smaller than the peak amplitude, the theoretical analysis of the algorithm is met, and meanwhile the filter characteristic of the algorithm is also met with the design requirement. When the system is applied to submarine optical cable burial depth detection, the burial depth value can be set to be PWM output, and therefore the change trend of the burial depth can be reflected visually.

The outdoor performance of the invention is further analyzed below and the correctness of the biorthogonal coil model is verified. And (3) only setting the detection frequency point of the system to be 25Hz, selecting a test channel to be a sampling channel 1, carrying out simulation buried depth detection and model analysis, and not carrying out analysis on the detection frequency points of the other two systems. The experimental design was as follows:

(1) a long wire is used for passing 25Hz alternating current to simulate the submarine optical cable.

(2) The search coil was modeled as an iron core coil of 10000 turns. The coil is mainly used for measuring induction voltage at two ends of the coil by considering an available submarine optical cable burial depth detection signal processing system, so that the burial depth is calculated.

The basic experimental schematic diagram is shown in fig. 31:

in fig. 31, the transmitter generates an ac signal at 25Hz that is injected into the conductor. We place the probe on a fixed height platform and measure the induced voltage across the probe in different directions relative to the cable. The results are shown in tables 3, 4 and 5.

TABLE 3 induced Voltage values in Probe at different angles

(Probe height 0.460m, current in wire 60mA)

TABLE 4 induced Voltage values in Probe at different angles

(Probe height 0.895m, current in wire 60mA)

TABLE 5 induced Voltage values in Probe at different angles

(Probe height 1.359m, current in wire 60mA)

From the results in the above table, it can be seen that the rms values of the coils that intersect at 90 ° are all approximately equal, and when η is 0 °, the coils induce the maximum value, which is consistent with our biorthogonal coil model analysis. We will next further analyze the sounding performance of the system.

The detection coil position during the experiment is shown in fig. 32

In actual detection, only two coils can be ensured to be perpendicular to each other, but the angles of the two coils relative to the submarine optical cable cannot be ensured, and when a biorthogonal coil model is applied, only a fixed value of the distance between two groups of coils can be ensured. Therefore, in order to research the actual depth measurement performance of the system, the data is processed according to the following method:

(1) positioning the cable buried depth by using the root mean square value of the voltage obtained by different heights of the coil which forms the same angle with the cable;

(2) all the measured burial depth values of the same height are averaged, and the relative error between the buried depth values and the real burial depth is calculated.

Let us assume that the cable buried depth is h1, and first, the sounding depth of three different angles is obtained according to the root mean square value and the height value of the voltage of the double coils on the horizontal planes a and B:

when the double coils and the cable form 0 degree and 90 degree, the sounding depth is 0.421 m;

when the double coils and the cable form an angle of 30 degrees and an angle of 120 degrees, the sounding depth is 0.407 m;

when the double coils and the cable form 60 degrees and 150 degrees, the sounding depth is 0.426 m;

the average depth of measurement is calculated to be 0.418m, and the relative error with the actual buried depth h1 is-9.13 percent, which is within the engineering allowable range.

In a similar way, the sounding of three different angles is obtained according to the root mean square value and the height value of the voltage of the double coils on the horizontal planes A and C:

when the double coils and the cable form an angle of 0 degree and an angle of 90 degrees, the depth of measurement is 0.415 m;

when the double coils and the cable form an angle of 30 degrees and an angle of 120 degrees, the sounding depth is 0.407 m;

when the double coils and the cable form 60 degrees and 150 degrees, the sounding depth is 0.415 m;

the average depth of sounding was found to be 0.412m, and the relative error with the actual depth of burial h1 was found to be-10.43%, within the engineering allowable range.

Finally, assuming that the cable buried depth is h2, the sounding of three different angles is obtained according to the root mean square value and the height value of the voltage of the double coils on the horizontal planes B and C:

when the double coils and the cable form 0 degree and 90 degree, the depth measurement is 0.833 m;

when the double coils and the cable form an angle of 30 degrees and an angle of 120 degrees, the sounding depth is 0.840 m;

when the double coils and the cable form 60 degrees and 150 degrees, the sounding depth is 0.819 m;

the average depth of sounding was found to be 0.831m, and the relative error with the actual depth of burial h2 was found to be-7.15%, within the engineering allowable range.

The relative errors of the sounding are negative errors, so that obvious system errors exist in the experiment, and the output of signal voltage is unstable, which causes large errors. Meanwhile, whether the coils are perpendicular to each other or not, whether the coil centers are right above the cable or not, whether the measurement of the coil heights is accurate or not and the like are not well mastered in an experiment, and certain errors can be caused. Meanwhile, the experimental result can prove the correctness of the biorthogonal coil sounding method and the reliability of the designed buried depth detection signal processing system on engineering.

Details not described in this specification are within the skill of the art that are well known to those skilled in the art.

38页详细技术资料下载

上一篇：一种医用注射器针头装配设备

下一篇：车辆轮对踏面擦伤及不圆度并联接触式检测系统

Submarine optical cable burial depth detection equipment

相关技术

网友询问留言