阅读说明：本技术 基于相位和幅值衰减的多探头阵列检测供水管道漏损的方法及系统 (Method and system for detecting leakage of water supply pipeline by multi-probe array based on phase and amplitude attenuation ) 是由 崔昊 袁一星 张鹏 于 2021-09-13 设计创作，主要内容包括：基于相位和幅值衰减的多探头阵列检测供水管道漏损的方法及系统,它属于供水管道漏损检测领域。本发明解决了传统声学检漏方法在供水管网检漏方面操作条件受限、检测精度以及检测效率低下的问题。本发明在地面摆布可移动式的多探头声学传感器,并通过特定的传感器排列方式采集地面振动噪声信号作为检测管道是否产生泄漏的声信号,通过快速傅里叶变换进行信号处理,获得声信号的相位谱特征与幅值谱特征后,基于获得的特征构造特征向量,将构造的特征向量输入BP神经网络获得对待测管道的漏损检测结果。本发明可以应用于对供水管道进行漏损检测。(A method and a system for detecting water supply pipeline leakage based on a multi-probe array with phase and amplitude attenuation belong to the field of water supply pipeline leakage detection. The invention solves the problems of limited operation conditions, low detection precision and low detection efficiency of the traditional acoustic leak detection method in the aspect of leak detection of the water supply network. The method comprises the steps of arranging movable multi-probe acoustic sensors on the ground, collecting ground vibration noise signals in a specific sensor arrangement mode to serve as acoustic signals for detecting whether the pipeline leaks, conducting signal processing through fast Fourier transform to obtain phase spectrum characteristics and amplitude spectrum characteristics of the acoustic signals, constructing characteristic vectors based on the obtained characteristics, and inputting the constructed characteristic vectors into a BP neural network to obtain leakage detection results of the pipeline to be detected. The invention can be applied to the leakage detection of the water supply pipeline.)

1. The method for detecting the leakage of the water supply pipeline based on the multi-probe array with the attenuated phase and amplitude is characterized by comprising the following steps:

step one, collecting pipeline signals by adopting a three-probe sensor array formed by A, B, C three sensors;

the acquired pipeline signals comprise pipeline signals under the laboratory condition and pipeline signals under the actual condition, and the pipeline signals comprise leakage pipeline signals and non-leakage pipeline signals;

secondly, amplifying the acquired signals, and connecting the amplified signals to an upper computer;

thirdly, the upper computer sequentially carries out denoising and preprocessing on the signals to obtain preprocessed signals;

extracting phase characteristics and amplitude attenuation characteristics of the preprocessed signals, constructing a phase characteristic vector based on the extracted phase characteristics, and constructing an amplitude attenuation characteristic vector based on the extracted amplitude attenuation characteristics;

the specific process of extracting the phase characteristics of the preprocessed signals is as follows:

after signals acquired by A, B, C by the three sensors simultaneously under the same experimental condition are processed in the second step and the third step, Fourier transformation is respectively carried out on the preprocessed signals corresponding to the sensor A, the sensor B and the sensor C, so that a Fourier transformed signal corresponding to the sensor A, a Fourier transformed signal corresponding to the sensor B and a Fourier transformed signal corresponding to the sensor C are obtained;

respectively calculate the frequency f₁、f₂、f₃、f₄、f₅Phase difference (A-B) A Phase difference (B-C) value, wherein Phase difference (A-B) represents a Phase spectrum difference value between a Fourier transformed signal corresponding to sensor A and a Fourier transformed signal corresponding to sensor B, and Phase difference (B-C) represents a Phase spectrum difference value between a Fourier transformed signal corresponding to sensor B and a Fourier transformed signal corresponding to sensor C;

will frequency f₁The Phase difference (A-B)/Phase difference (B-C) value is marked as x 1;

will frequency f₂The Phase difference (A-B)/Phase difference (B-C) value is marked as x 2;

will frequency f₃The Phase difference (A-B)/Phase difference (B-C) value is marked as x 3;

will frequency f₄The Phase difference (A-B)/Phase difference (B-C) value is marked as x 4;

will frequency f₅The Phase difference (A-B)/Phase difference (B-C) value is marked as x 5;

constructing a phase eigenvector X under the current experimental conditions (X1, X2, X3, X4, X5);

for the collected signals under other experimental conditions and actual conditions, the phase characteristic extraction mode and the phase characteristic vector construction mode are the same;

assigning values to the pipeline signals acquired in the step one each time, taking the constructed phase characteristic vector and amplitude attenuation characteristic vector as the input of a BP neural network, taking the corresponding assignment result as the output of the BP neural network, and training the BP neural network;

and step six, collecting a pipeline signal to be detected by using a three-probe sensor array, extracting phase characteristics and amplitude attenuation characteristics of the pipeline signal to be detected, constructing a characteristic vector, inputting the characteristic vector of the pipeline signal to be detected into the trained BP neural network, and obtaining a leakage detection result of the pipeline to be detected.

2. The method for detecting water supply pipeline leakage based on phase and amplitude attenuation multi-probe array of claim 1, wherein said A, B, C three sensors are arranged in a right triangle.

3. The method for detecting water supply pipeline leakage based on phase and amplitude attenuation multi-probe array of claim 2 wherein the lengths of the strands of the right angle triangle are 0.7m, 0.39m and 0.8m, respectively.

4. The method for detecting water supply pipeline leakage based on the phase and amplitude attenuation multi-probe array as claimed in claim 1, wherein the signal acquisition process in the laboratory case is as follows:

welding a box body by adopting a steel plate, taking the box body as a soil carrier, enabling the pipeline to penetrate through the box body from a position H away from the bottom of the box body along the longitudinal direction, and circularly supplying water to the pipeline by adopting a water tank;

the three-probe sensor array is arranged right above the pipeline, the size and the orientation of a leakage opening of the pipeline and the burial depth of the pipeline in the box body are continuously adjusted, and signals are acquired by the three-probe sensor array during each adjustment.

5. The method for detecting water supply pipeline leakage based on a phase and amplitude attenuation multi-probe array of claim 1 wherein said preprocessing includes pre-emphasis, framing and windowing.

6. The method for detecting water supply pipeline leakage based on the multi-probe array with phase and amplitude attenuation as claimed in claim 1, wherein the specific process of extracting the amplitude attenuation characteristics of the preprocessed signals is as follows:

after signals acquired by A, B, C by the three sensors simultaneously under the same experimental condition are processed in the second step and the third step, respectively performing Fourier transform on the preprocessed signals corresponding to the sensor A, the sensor B and the sensor C to obtain a signal time frequency spectrum corresponding to the sensor A, a signal time frequency spectrum corresponding to the sensor B and a signal time frequency spectrum corresponding to the sensor C;

constructing a function y which is transfer function (A-B)/transfer function (B-C), wherein transfer function (A-B) represents the difference of the time spectrum amplitude of the sensor A and the sensor B, and transfer function (B-C) represents the difference of the time spectrum amplitude of the sensor B and the sensor C;

respectively calculate the frequency f₁、f₂、f₃、f₄、f₅The following y-values:

will frequency f₁The lower transfer function (A-B)/transfer function (B-C) value was designated as y 1;

will frequency f₂The lower transfer function (A-B)/transfer function (B-C) value was designated as y 2;

will frequency f₃The lower transfer function (A-B)/transfer function (B-C) value was designated as y 3;

will frequency f₄The lower transfer function (A-B)/transfer function (B-C) value was designated as y 4;

will frequency f₅The lower transfer function (A-B)/transfer function (B-C) value was designated as y 5;

constructing an amplitude attenuation characteristic vector Y under the current experimental condition (Y1, Y2, Y3, Y4 and Y5);

for signals under other acquired experimental conditions and under actual conditions, the amplitude attenuation feature extraction mode and the amplitude attenuation feature vector construction mode are the same.

7. The method for detecting water supply pipeline leakage based on the phase and amplitude attenuation multi-probe array of claim 1, wherein the pipeline signal collected each time in the first step is assigned a value of 0 or 1.

8. The method for detecting water supply pipeline leakage based on the multi-probe array with phase and amplitude attenuation of claim 1, wherein the number of input layer nodes of the BP neural network is 10, the number of hidden layer nodes is 12, and the number of output layer nodes is 1.

9. The method for detecting water supply pipeline leakage based on the phase and amplitude attenuation multi-probe array of claim 1, wherein the amplified signal is connected to an upper computer through a dynamic acquisition analyzer.

10. A system for detecting water supply pipeline leakage based on a phase and amplitude attenuation multi-probe array, wherein the system is adapted to perform the method for detecting water supply pipeline leakage based on a phase and amplitude attenuation multi-probe array as claimed in any one of claims 1 to 9.

Technical Field

The invention belongs to the field of water supply pipeline leakage detection, and particularly relates to a method and a system for detecting water supply pipeline leakage based on a multi-probe array with phase and amplitude attenuation.

Background

The urban water supply network is an important municipal field which is concerned with national living health level and national economy, but at present, partial domestic cities still have the problems that water supply facilities are old, pipe networks are seriously aged, the updating speed is slow, and even the pipe networks are buried and positioned to be lost, and leakage accidents occur occasionally, so that a large amount of treated clean water resources are wasted due to loss, and huge loss is caused to the national economy. At present, leakage detection methods adopted at home and abroad are various, wherein the acoustic leakage detection method is more commonly applied. Acoustic leak detection methods can be broadly classified into hardware-based leak detection methods and software-based leak detection methods. The method for detecting the leakage based on hardware mainly comprises an acoustic leakage detecting method based on a pipeline leakage acoustic signal at present and is a positioning method widely applied at present, adopted instruments mainly comprise a leakage listening rod and an electronic leakage listening instrument, the method is accumulated by the experience of leakage detection workers seriously, a qualified technician can be cultured in a long time, the method has high requirement on environmental noise and can work only when the person is quiet at night, and certain influence is caused on the hearing and the body of the workers for a long time. Therefore, it is necessary to develop a leakage detection method which is accurate in positioning and suitable for the situation of the Chinese pipe network, and further develop an instrument and equipment which are simple in operation.

A leak point positioning method based on software is a leak positioning and identification method through model and signal processing, such as a correlation analysis method, and the method is a method which is applied more generally in the world at present, a time delay estimation method is applied, two sensors are respectively arranged at two ends of a pipeline or at a valve and a fire hydrant and used for receiving vibration sound wave signals transmitted along the pipeline, and the position of a leak point is accurately calculated through the time difference of the sound wave signals received by the two sensors and various input parameters such as the length, the material and the pipe diameter of the pipeline. The method is simple to operate, accurate in positioning and widely applied to the leakage detection and positioning field all over the world, but for the complex pipe network condition of China and the sensor installation distance requirement of a correlation instrument, the position where two sensors can be installed is sometimes difficult to find, and the specific length of a pipeline between two point positions cannot be determined, if other fittings such as an elbow and a valve exist on a pipe section between the two sensors, the analysis result is also influenced, and the whole set of correlation analyzer is expensive, and for some counties and towns with low economic level in China, a water supply company cannot bear related cost, so the method is difficult to popularize in China.

In summary, a leakage positioning method based on a leakage acoustic signal has been widely applied, but in an actual leakage detection process, the method is influenced by pipeline conditions and environmental noise, detection efficiency, detection accuracy and operation conditions of a correlation analysis method and a ground leakage acoustic detection method are limited, and the ground leakage acoustic detection method is used as a method most widely applied in an actual leakage detection work, and it is urgently needed to realize intellectualization by combining a modern sensing technology and a signal processing means.

Disclosure of Invention

The invention aims to solve the problems of limited operating conditions, low detection precision and low detection efficiency of the traditional acoustic leak detection method in the aspect of leak detection of a water supply network, and provides a method and a system for detecting leakage of a water supply pipeline by using a multi-probe array based on phase and amplitude attenuation.

The technical scheme adopted by the invention for solving the technical problems is as follows:

a method for detecting water supply pipeline leakage based on a multi-probe array with phase and amplitude attenuation specifically comprises the following steps:

step one, collecting pipeline signals by adopting a three-probe sensor array formed by A, B, C three sensors;

the acquired pipeline signals comprise pipeline signals under the laboratory condition and pipeline signals under the actual condition, and the pipeline signals comprise leakage pipeline signals and non-leakage pipeline signals;

secondly, amplifying the acquired signals, and connecting the amplified signals to an upper computer;

thirdly, the upper computer sequentially carries out denoising and preprocessing on the signals to obtain preprocessed signals;

extracting phase characteristics and amplitude attenuation characteristics of the preprocessed signals, constructing a phase characteristic vector based on the extracted phase characteristics, and constructing an amplitude attenuation characteristic vector based on the extracted amplitude attenuation characteristics;

the specific process of extracting the phase characteristics of the preprocessed signals is as follows:

after signals acquired by A, B, C by the three sensors simultaneously under the same experimental condition are processed in the second step and the third step, Fourier transformation is respectively carried out on the preprocessed signals corresponding to the sensor A, the sensor B and the sensor C, so that a Fourier transformed signal corresponding to the sensor A, a Fourier transformed signal corresponding to the sensor B and a Fourier transformed signal corresponding to the sensor C are obtained;

respectively calculate the frequency f₁、f₂、f₃、f₄、f₅A Phase difference (A-B)/Phase difference (B-C) value, wherein Phase difference (A-B) represents a Phase spectrum difference value between a Fourier transformed signal corresponding to sensor A and a Fourier transformed signal corresponding to sensor B, and Phase difference (B-C) represents a Phase spectrum difference value between a Fourier transformed signal corresponding to sensor B and a Fourier transformed signal corresponding to sensor C;

will frequency f₁The Phase difference (A-B)/Phase difference (B-C) value is marked as x 1;

will frequency f₂The Phase difference (A-B)/Phase difference (B-C) value is marked as x 2;

will frequency f₃The Phase difference (A-B)/Phase difference (B-C) value is marked as x 3;

will frequency f₄The Phase difference (A-B)/Phase difference (B-C) value is marked as x 4;

will frequency f₅The Phase difference (A-B)/Phase difference (B-C) value is marked as x 5;

constructing a phase eigenvector X under the current experimental conditions (X1, X2, X3, X4, X5);

for the collected signals under other experimental conditions and actual conditions, the phase characteristic extraction mode and the phase characteristic vector construction mode are the same;

assigning values to the pipeline signals acquired in the step one each time, taking the constructed phase characteristic vector and amplitude attenuation characteristic vector as the input of a BP neural network, taking the corresponding assignment result as the output of the BP neural network, and training the BP neural network;

and step six, collecting a pipeline signal to be detected by using a three-probe sensor array, extracting phase characteristics and amplitude attenuation characteristics of the pipeline signal to be detected, constructing a characteristic vector, inputting the characteristic vector of the pipeline signal to be detected into the trained BP neural network, and obtaining a leakage detection result of the pipeline to be detected.

Further, the A, B, C three sensors are arranged in a right triangle.

Further, the lengths of the strands of the right triangle are 0.7m, 0.39m and 0.8m respectively.

Further, the acquisition process of the signals in the laboratory situation is as follows:

welding a box body by adopting a steel plate, taking the box body as a soil carrier, enabling the pipeline to penetrate through the box body from a position H away from the bottom of the box body along the longitudinal direction, and circularly supplying water to the pipeline by adopting a water tank;

the three-probe sensor array is arranged right above the pipeline, the size and the orientation of a leakage opening of the pipeline and the burial depth of the pipeline in the box body are continuously adjusted, and signals are acquired by the three-probe sensor array during each adjustment.

Further, the preprocessing includes pre-emphasis, framing, and windowing.

Further, the specific process of extracting the amplitude attenuation characteristic of the preprocessed signal is as follows:

after signals acquired by A, B, C by the three sensors simultaneously under the same experimental condition are processed in the second step and the third step, respectively performing Fourier transform on the preprocessed signals corresponding to the sensor A, the sensor B and the sensor C to obtain a signal time frequency spectrum corresponding to the sensor A, a signal time frequency spectrum corresponding to the sensor B and a signal time frequency spectrum corresponding to the sensor C;

constructing a function y which is transfer function (A-B)/transfer function (B-C), wherein transfer function (A-B) represents the difference of the time spectrum amplitude of the sensor A and the sensor B, and transfer function (B-C) represents the difference of the time spectrum amplitude of the sensor B and the sensor C;

respectively calculate the frequencyRate f₁、f₂、f₃、f₄、f₅The following y-values:

will frequency f₁The lower transfer function (A-B)/transfer function (B-C) value was designated as y 1;

will frequency f₂The lower transfer function (A-B)/transfer function (B-C) value was designated as y 2;

will frequency f₃The lower transfer function (A-B)/transfer function (B-C) value was designated as y 3;

will frequency f₄The lower transfer function (A-B)/transfer function (B-C) value was designated as y 4;

will frequency f₅The lower transfer function (A-B)/transfer function (B-C) value was designated as y 5;

constructing an amplitude attenuation characteristic vector Y under the current experimental condition (Y1, Y2, Y3, Y4 and Y5);

for signals under other acquired experimental conditions and under actual conditions, the amplitude attenuation feature extraction mode and the amplitude attenuation feature vector construction mode are the same.

Further, the pipeline signal acquired in each step in the first step is assigned with a value of 0 or 1.

Further, the number of nodes of an input layer of the BP neural network is 10, the number of nodes of a hidden layer is 12, and the number of nodes of an output layer is 1.

Further, the amplified signal is connected to an upper computer through a dynamic acquisition analyzer.

A system for detecting water supply pipeline leakage based on a phase and amplitude attenuated multi-probe array, said system being adapted to perform a method for detecting water supply pipeline leakage based on a phase and amplitude attenuated multi-probe array.

The invention has the beneficial effects that:

arranging a movable multi-probe acoustic sensor on the ground, collecting ground vibration noise signals in a specific sensor arrangement mode to serve as acoustic signals for detecting whether leakage occurs in a pipeline, performing signal processing through fast Fourier transform, and constructing a feature vector based on obtained features after phase spectrum features and amplitude spectrum features of the acoustic signals are obtained; and inputting the constructed characteristic vector into the BP neural network to obtain a leakage detection result of the pipeline to be detected by establishing the BP neural network. The method can be realized only by collecting the acoustic signals on the ground without being limited by operating conditions, and leakage detection can be more accurate while the detection efficiency is improved by comprehensively considering the phase spectrum characteristics and the amplitude spectrum characteristics.

Drawings

FIG. 1 is a flow chart of a method of detecting water supply pipeline leaks based on a phase and amplitude attenuated multi-probe array of the present invention.

Detailed Description

First embodiment this embodiment will be described with reference to fig. 1. The method for detecting leakage of a water supply pipeline based on the multi-probe array with attenuated phase and amplitude comprises the following steps:

step one, collecting pipeline signals by adopting a three-probe sensor array formed by A, B, C three sensors;

the acquired pipeline signals comprise pipeline signals under the laboratory condition and pipeline signals under the actual condition, and the pipeline signals comprise leakage pipeline signals and non-leakage pipeline signals;

secondly, amplifying the acquired signals, and connecting the amplified signals to an upper computer;

thirdly, the upper computer sequentially carries out denoising and preprocessing on the signals to obtain preprocessed signals;

extracting phase characteristics and amplitude attenuation characteristics of the preprocessed signals, constructing a phase characteristic vector based on the extracted phase characteristics, and constructing an amplitude attenuation characteristic vector based on the extracted amplitude attenuation characteristics;

the specific process of extracting the phase characteristics of the preprocessed signals is as follows:

after signals acquired by A, B, C at the same time by the three sensors under the same experimental condition (the same experimental condition refers to the condition that the sizes and the directions of the leakage openings are the same as the buried depth in the box body) are processed in the second step and the third step, the preprocessed signals corresponding to the sensor A, the sensor B and the sensor C are subjected to Fourier transform respectively to obtain a Fourier transformed signal corresponding to the sensor A, a Fourier transformed signal corresponding to the sensor B and a Fourier transformed signal corresponding to the sensor C;

respectively calculate the frequency f₁、f₂、f₃、f₄、f₅A Phase difference (A-B)/Phase difference (B-C) value, wherein Phase difference (A-B) represents a Phase spectrum difference value between a Fourier transformed signal corresponding to sensor A and a Fourier transformed signal corresponding to sensor B, and Phase difference (B-C) represents a Phase spectrum difference value between a Fourier transformed signal corresponding to sensor B and a Fourier transformed signal corresponding to sensor C; phase difference (A-B)/Phase difference (B-C) represents the ratio of Phase difference (A-B) to Phase difference (B-C);

will frequency f₁The Phase difference (A-B)/Phase difference (B-C) value is marked as x 1;

will frequency f₂The Phase difference (A-B)/Phase difference (B-C) value is marked as x 2;

will frequency f₃The Phase difference (A-B)/Phase difference (B-C) value is marked as x 3;

will frequency f₄The Phase difference (A-B)/Phase difference (B-C) value is marked as x 4;

will frequency f₅The Phase difference (A-B)/Phase difference (B-C) value is marked as x 5;

constructing a phase eigenvector X under the current experimental conditions (X1, X2, X3, X4, X5);

for the collected signals under other experimental conditions and actual conditions, the phase characteristic extraction mode and the phase characteristic vector construction mode are the same;

before the method of the invention starts, the numbers (A, B, C) of any three sensors can be given, and once the numbers are determined, the numbers are not changed in the whole treatment process;

since the fourier transformed (FFT) signal satisfies the following relationship:

(1) under the fixed frequency, the phase spectrum difference value between the Fourier transformed signal corresponding to the sensor A and the Fourier transformed signal corresponding to the sensor B is equal to the phase spectrum difference value between the Fourier transformed signal corresponding to the sensor B and the Fourier transformed signal corresponding to the sensor C;

(2) under different frequencies, the phase spectrum difference value between the Fourier transformed signal corresponding to the sensor A and the Fourier transformed signal corresponding to the sensor B linearly changes along with the distance between the three-probe sensor array and the position right above the leak point;

therefore, the phase characteristic extraction mode of the invention is designed;

assigning values to the pipeline signals acquired in the step one each time, taking the constructed phase characteristic vector and amplitude attenuation characteristic vector as the input of a BP neural network, taking the corresponding assignment result as the output of the BP neural network, and training the BP neural network;

the signals acquired each time have corresponding eigenvectors and assignments;

and step six, collecting a pipeline signal to be detected by using a three-probe sensor array, extracting phase characteristics and amplitude attenuation characteristics of the pipeline signal to be detected, constructing a characteristic vector, inputting the characteristic vector of the pipeline signal to be detected into the trained BP neural network, and obtaining a leakage detection result of the pipeline to be detected.

After the characteristic vector of the pipeline signal to be detected is input, the output value of the BP neural network is between 0 and 1, 0 represents that no leakage is determined, 1 represents that leakage is determined, and the numerical value between 0 and 1 represents the probability of leakage.

The propagation rule of the leakage signal on the ground meets the following conditions: displacement u-Ae^i(kx-wt)Here, k is an imaginary number, a real part re (k) of k is frequency/velocity w/v, and an imaginary part of k is an attenuation coefficient and is denoted as im (k). The phase is referred to as kx, and it can be seen that for t being the same, kx is a function of distance, and based on this principle, the phase and amplitude decay characteristics are taken as characteristics of the leakage detection.

The amplitude attenuation characteristic is greatly different under the conditions of pipeline leakage and no pipeline leakage, and can attenuate by 30dB within 2m from a leakage point, so that the amplitude attenuation characteristic is suitable for being used as a judgment basis of pipeline leakage, but the characteristic is easily interfered by noise. And the phase information characteristic is a characteristic which is more stable than the amplitude value in the presence of noise and is less affected by the noise. According to the invention, the amplitude attenuation characteristic and the phase information characteristic are selected as the consideration factors of the pipeline leakage detection, so that the precision of the leakage detection can be obviously improved, and the two characteristics are not compatible.

The second embodiment is as follows: the difference between this embodiment and the first embodiment is that the A, B, C three sensors are arranged in a right triangle.

Other steps and parameters are the same as those in the first embodiment.

The third concrete implementation mode: the difference between the present embodiment and the first or second embodiment is that the length of the strands of the right triangle is 0.7m, 0.39m and 0.8m, respectively.

This embodiment is through reasonable in design's colluding thigh length to guarantee the precision that detects.

Other steps and parameters are the same as those in the first or second embodiment.

The fourth concrete implementation mode: the difference between this embodiment and one of the first to third embodiments is that the signal acquisition process in the laboratory situation is as follows:

welding a box body by adopting a steel plate, taking the box body as a soil carrier, enabling the pipeline to penetrate through the box body from a position H away from the bottom of the box body along the longitudinal direction, and circularly supplying water to the pipeline by adopting a water tank;

the three-probe sensor array is arranged right above the pipeline, the size and the orientation of a leakage opening of the pipeline and the burial depth of the pipeline in the box body are continuously adjusted, and signals are acquired by the three-probe sensor array during each adjustment.

The water tank is adopted for circulating water supply, the vertical multi-stage pump supplies energy, the leaked water amount is supplemented by the water tank, the air pressure tank is adopted for supplying water, the water pump is firstly used for pressing water into the air pressure tank, and the water pressure is closed when reaching a set value. During the experiment, high-pressure water flow is pumped into the simulation pipeline through the air pressure tank, and the pressure in the pipeline is controlled to be constant through the pressure reducing valve and the pressure gauge. The pipeline is removable, can adjust leak opening size, orientation and position to acquire the signal under the various leakage conditions.

Other steps and parameters are the same as those in one of the first to third embodiments.

The fifth concrete implementation mode: this embodiment is different from one of the first to the fourth embodiments in that the preprocessing includes pre-emphasis, framing, and windowing.

Other steps and parameters are the same as in one of the first to fourth embodiments.

The sixth specific implementation mode: the difference between this embodiment and one of the first to fifth embodiments is that the specific process of extracting the amplitude attenuation characteristic of the preprocessed signal is as follows:

after signals acquired by A, B, C by the three sensors simultaneously under the same experimental condition are processed in the second step and the third step, respectively performing Fourier transform on the preprocessed signals corresponding to the sensor A, the sensor B and the sensor C to obtain a signal time frequency spectrum corresponding to the sensor A, a signal time frequency spectrum corresponding to the sensor B and a signal time frequency spectrum corresponding to the sensor C;

the two groups of attenuations are the same in principle, and due to the influence of body waves near leakage points, the closer to the leakage points, the larger the relative time frequency spectrum is;

constructing a function y which is transfer function (A-B)/transfer function (B-C), wherein transfer function (A-B) represents the difference of the time spectrum amplitude of the sensor A and the sensor B, and transfer function (B-C) represents the difference of the time spectrum amplitude of the sensor B and the sensor C;

respectively calculate the frequency f₁、f₂、f₃、f₄、f₅The following y-values:

will frequency f₁The lower transfer function (A-B)/transfer function (B-C) value was designated as y 1;

will frequency f₂The lower transfer function (A-B)/transfer function (B-C) value was designated as y 2;

will frequency f₃The lower transfer function (A-B)/transfer function (B-C) value was designated as y 3;

will frequency f₄The lower transfer function (A-B)/transfer function (B-C) value was designated as y 4;

frequency conversionRate f₅The lower transfer function (A-B)/transfer function (B-C) value was designated as y 5;

constructing an amplitude attenuation characteristic vector Y under the current experimental condition (Y1, Y2, Y3, Y4 and Y5);

for signals under other acquired experimental conditions and under actual conditions, the amplitude attenuation feature extraction mode and the amplitude attenuation feature vector construction mode are the same.

And obtaining the amplitude of the corresponding frequency from the frequency spectrum, constructing a transfer function, and obtaining a feature vector Y.

Other steps and parameters are the same as those in one of the first to fifth embodiments.

The seventh embodiment: the difference between this embodiment and one of the first to sixth embodiments is that the pipeline signal acquired in step one at each time is assigned with a value of 0 or 1.

The pipeline signal with a missing point is assigned a value of 1 and the ambient noise is assigned a value of 0.

Other steps and parameters are the same as those in one of the first to sixth embodiments.

The specific implementation mode is eight: the difference between this embodiment and one of the first to seventh embodiments is that the number of nodes of the input layer of the BP neural network is 10, the number of nodes of the hidden layer is 12, and the number of nodes of the output layer is 1.

Other steps and parameters are the same as those in one of the first to seventh embodiments.

The specific implementation method nine: the difference between this embodiment and the first to eighth embodiment is that the amplified signal is connected to an upper computer through a dynamic acquisition analyzer.

Other steps and parameters are the same as those in one to eight of the embodiments.

The detailed implementation mode is ten: the system for detecting water supply pipeline leakage based on the multi-probe array with attenuated phase and amplitude is used for executing the method for detecting water supply pipeline leakage based on the multi-probe array with attenuated phase and amplitude.

The above-described calculation examples of the present invention are merely to explain the calculation model and the calculation flow of the present invention in detail, and are not intended to limit the embodiments of the present invention. It will be apparent to those skilled in the art that other variations and modifications of the present invention can be made based on the above description, and it is not intended to be exhaustive or to limit the invention to the precise form disclosed, and all such modifications and variations are possible and contemplated as falling within the scope of the invention.