阅读说明：本技术 基于小波分析的电荷感应风速检测方法 (Charge induction wind speed detection method based on wavelet analysis ) 是由 刘丹丹 汤春瑞 于 2018-08-01 设计创作，主要内容包括：基于小波分析的电荷感应风速检测方法,解决可直接准确测量风速、无需标定,而且具有永不堵塞、免维护的特征。能够适应恶劣环境、大管道高流速风速、风量的在线检测。交流耦合静电感应检测获得的电信号反应粉尘粒子的运动状态及带电属性,在检测探头设计固定的情况下,也会出现较复杂的波动及不明显的周期性特点,反应为正弦信号的多次谐波的叠加(也包含有空间电磁干扰)。在粉尘浓度相对稳定及流场相对均匀的情况下,信号会表现出较好的周期性。而信号的波动正好反映了粉尘浓度大小,因此从算法规划方向上,通过小波分析或短时傅立叶变化后,主要提取信号的偏差(反映浓度的变化程度),通过一定时间的累积,反映该段时间段内粉尘的平均浓度,并实时跟踪累积的变化。(The charge induction wind speed detection method based on wavelet analysis solves the problems that the wind speed can be directly and accurately measured, calibration is not needed, and the charge induction wind speed detection method has the characteristics of never blocking and maintenance. The device can be suitable for online detection of severe environment, high flow speed and high air volume of a large pipeline. The electric signal obtained by the ac coupling electrostatic induction detection reflects the motion state and the charged property of the dust particles, and under the condition that the design of the detection probe is fixed, the characteristics of complex fluctuation and unobvious periodicity also appear, and the signal is reflected as the superposition of multiple harmonics of a sinusoidal signal (including space electromagnetic interference). Under the conditions of relatively stable dust concentration and relatively uniform flow field, the signal shows better periodicity. And the fluctuation of the signal just reflects the dust concentration, so that in the algorithm planning direction, after wavelet analysis or short-time Fourier change, the deviation of the signal (reflecting the change degree of the concentration) is mainly extracted, the average concentration of the dust in a period of time is reflected through accumulation in a certain time, and the accumulated change is tracked in real time.)

1. The charge induction wind speed detection method based on wavelet analysis adopts the same detection method at a fixed distance to obtain an induction signal of dust passing through two detection points. For the same dust cloud, signals obtained at two points theoretically have good correlation, and the passing time of dust is obtained through a wavelet transform disturbance quantity correlation threshold detection algorithm to calculate the wind speed and the wind quantity.

For the one-dimensional signal C (x), assume { C_i(x) Is the scalar product of the signal c (x) and a scaling function ψ (x), which is in effect a low pass filter. The signal C (x) is filtered for the first time to obtain C₁(x)，ω₁(x)＝C₀(x)-C₁(x) Containing information between these two scales, ω₁(x) Referred to as the first wavelet surface, is also the result of the wavelet transform corresponding to the scale function. And the wavelet function phi (x) has the following relationship with the scale function phi (x):

φ[x/2]/2＝ψ(x)-ψ[x/2]/2 (1)

the difference between adjacent scales is two times, and C is obtained after i times of filtering_i(x) Comprises the following steps:

ω_i(x)＝C_i-1(x)-C_i(x) (3)

in the formula, ω_i(x) Is the wavelet coefficient at the scale i; c_i(x) Is an approximate signal at the scale i; h is a low-pass filter which satisfies the following equation with the scale function ψ (x):

atrous discrete wavelet decomposes the signal to generate a set of adjacent wavelet coefficients { omega } with different resolution_iAnd the approximation signal.

Technical Field

The invention relates to a wind speed detection method, in particular to a method for measuring wind speed under the conditions of an underground coal mine ultrasonic wave detection principle and high environmental requirements of a mechanical velocimeter wind speed sensor.

Background

At present, an air velocity sensor applied to a coal mine mainly adopts an ultrasonic detection principle, and a vane type is also applied to a portable velocimeter. However, the ultrasonic wind speed sensor probe has a large volume, and has large interference on a pipeline flow field, and the test range is less than 30 m/s. The detection of the pipeline wind speed adopts differential pressure type and thermal type wind speed sensors, but has higher requirements on the testing environment, is easy to be polluted underground, and the thermal type sensors have slower reaction and can generate temperature drift after long-term use. Wind speed and flue gas flow rate measurements are important parameters in industrial automation processes. Wind speed measuring devices are important equipment for industrial automation process control.

At present, widely used in industry wind speed and wind volume measuring devices, such as pitot tube, power bar, back flute tube, venturi, multi-throat stiffness, wing device, etc., all adopt the P measuring method, i.e. full pressure measuring tube and static pressure measuring tube, one end of a differential pressure transmitter is connected with a full pressure sensing tube, the other end is connected with a static pressure sensing tube, and the flow speed is measured by measuring the differential pressure between full pressure and static pressure.

The tail end of the pressure sensing pipe is required to be connected with a differential pressure transmitter for pressure sensing, so that the pressure sensing pipe is easy to block, particularly in measurement of dusty airflow, the blockage is easy, the maintenance amount is large, and the measurement is difficult.

Disclosure of Invention

In order to overcome the defects, the invention provides a charge induction wind speed detection method based on wavelet analysis by utilizing a charge induction technology and a related principle, which not only can directly and accurately measure the wind speed without calibration, but also has the characteristics of no blockage and no maintenance. The device can be suitable for online detection of severe environment, high flow speed and high air volume of a large pipeline.

And obtaining the induction signals of dust passing through two detection points by adopting the same wavelet disturbance amount detection method at a fixed distance, and obtaining the dust passing time to calculate the wind speed and the wind volume.

When the pulverized coal medium flow in the pipeline passes through the charge induction sensor, the charge signal received by the sensor comprises impact charge, friction charge and induction charge of the medium to the probe. But since the surface area of the sensor mounted on the pipe is very small relative to the area of the pipe, most of the received charge signal comes from induced charge. The strength of the induced charge signal is related to the coal fines concentration and flow rate. As shown in the following formula, the charge signal intensity is a binary function of the coal concentration and the wind flow rate, and is influenced by the size of the coal concentration and the change of the wind flow rate.

E＝f(N,V)

Wherein:

e-charge induced signal strength

Concentration of N-coal powder

V-velocity of wind powder

The alternating current charge induction type coal dust concentration measuring device adopts an alternating current charge coupling technology, measures the alternating current disturbance quantity of the induced charge around the charge average value when primary air dust flows through a probe, and uses the alternating current disturbance quantity as a basic signal for measuring the coal dust concentration and the flow velocity. In the ac coupling technique, the positive and negative averages of the induced charge are filtered out, and the system then detects the electric field, peak value, root mean square value, and other various mixed variations of the residual perturbation signal. Among the above values, the rms value can accurately reflect the standard deviation of the signal. The higher the coal dust concentration and the higher the flow velocity, the larger the standard deviation of the signal, so the AC coupling technology determines the disturbance quantity of the AC signal by monitoring the standard deviation of the charge signal and determines the instantaneous total quantity of the coal dust by the magnitude of the instantaneous disturbance quantity. The standard deviation of the charge signal has an accurate linear relation with the instantaneous total amount of the pulverized coal, and the stable pulverized coal concentration is obtained by dividing the instantaneous primary air flow speed measured at the same time and the same measuring point by the standard deviation of the charge signal.

As shown in the following formula: coal powder concentration is coal powder total amount/(flow rate pipe cross section area)

The relative concentration value of the pulverized coal in the primary air pipe can be accurately measured, and the absolute value of the concentration of the pulverized coal is obtained after the calibration is carried out on site by a total calibration method. All concentration measurement devices need to be calibrated on site after installation.

The invention achieves the technical scheme that an electric signal obtained by AC coupling electrostatic induction detection reflects the motion state and the charged property of dust particles, under the condition that a detection probe is designed and fixed, more complex fluctuation and unobvious periodic characteristics can also appear, the signal is reflected as superposition of multiple harmonics of sinusoidal signals (also including space electromagnetic interference), under the condition that the dust concentration is relatively stable and a flow field is relatively uniform, the signal can show better periodicity, and the fluctuation of the signal just reflects the dust concentration, therefore, in the direction of algorithm planning, after wavelet analysis or short-time Fourier change, the deviation of the signal (reflecting the change degree of the concentration) is mainly extracted, the average concentration of the dust in the period is reflected through accumulation of a certain time, and the accumulated change is tracked in real time, the wind speed detection adopts a two-point detection and calculation related algorithm, as shown in figure 1, the same wavelet disturbance quantity detection method is adopted in a fixed distance, the dust passes through two induction signals of two beams, the same induction signal cloud of the same beam, theoretically, the signal obtained through the same wavelet disturbance quantity detection method, the same wavelet detection method is adopted, the two sets of the air speed detection and the two sets of the induction signal processing unit, namely, the two sets of the two induction sensors are designed and the two sets of the induction signal processing units, the two sets of the dust particles are processed according to obtain the high-time induction signal processing algorithm, namely, the principle of the dust particle induction signal processing unit, the dust particle processing unit, the dust processing unit is the dust processing unit, the dust processing;

the installation distance between the upstream and downstream sensors is constant L, and the absolute flow rate of airflow can be accurately calculated by using the formula V (m/s) ═ L (m)/△ t(s). The volume flow can be measured after the absolute flow rate is input into the cross section area of the air duct, and the signal processing unit can be converted into an analog signal representing the flow rate by 4-20mA through D/A conversion or can be output to other systems through a field bus.

The invention has the beneficial effects that: the wind speed is detected by adopting an induction type principle, the test range is wide, and the theory can reach 60 m/s; the probe is a metal rod, and the interference of the large pipeline to a flow field is extremely small; the wind speed continuous online detection device is particularly suitable for the environment with dust, can not cause dust deposition after long-term operation, is maintenance-free, has strong adaptability, and can be widely applied to the continuous online detection of the wind speed of each point in the underground coal mine. And the pipeline wind speed sensor based on the electrostatic induction is developed by adopting the same detection method and processing through a related algorithm, the measurement range is 2-40 m/s, the absolute error of the measured value is less than or equal to +/-0.5 m/s, and the rate of differentiation is 0.1 m/s.

The present invention will be described in detail with reference to the accompanying drawings.

Drawings

FIG. 1 is a schematic diagram of a charge induction wind speed detection device system based on wavelet analysis.

FIG. 2 is a schematic diagram of a test precision verification system of the wind speed detector.

The specific implementation mode is as follows:

because the dust particles in the dust-containing airflow are charged to generate induced charges on the probe, the dust concentration can be detected by measuring the change of the disturbance quantity of a charge signal around the charge average value. Different from the direct-current coupling static induction technology, the alternating-current coupling static induction technology adopts the disturbance quantity (fluctuation signal) of the detected dust charge instead of the total static induction quantity (the sum of friction static and induction static, and the total charge quantity average value after filtering out the alternating-current signal), so the method is not influenced by the accumulation of the dust of the probe, has high signal intensity, can adopt more algorithms to process and identify the disturbance quantity, and easily annihilate the direct-current coupling direct-current signal in the disturbance signal of the environment, and has lower identification precision.

And obtaining the induction signals of the dust passing through the two detection points by adopting the same detection method at a fixed distance. For the same dust cloud, signals obtained at two points theoretically have good correlation, and the passing time of dust is obtained through a wavelet transform disturbance quantity correlation threshold detection algorithm to calculate the wind speed and the wind quantity. The charge induction wind speed detection device based on wavelet analysis is shown in fig. 1:

1) the calculation space and time requirements are reasonable, and programming is easy to realize; 2) the method has two-dimensional isotropy, and the transformation process can be realized through filtering; 3) sampling and interpolation are not needed in the calculation, and the detailed characteristics of the signals are favorably acquired;

for the one-dimensional signal C (x), assume { C_i(x) Is the scalar product of the signal c (x) and a scaling function ψ (x), which is in effect a low pass filter. The signal C (x) is filtered for the first time to obtain C₁(x)，ω₁(x)＝C₀(x)-C₁(x) Containing information between these two scales, ω₁(x) Referred to as the first wavelet surface, is also the result of the wavelet transform corresponding to the scale function. And the wavelet function phi (x) has the following relationship with the scale function phi (x):

φ[x/2]/2＝ψ(x)-ψ[x/2]/2 (1)

the difference between adjacent scales is two times, and C is obtained after i times of filtering_i(x) Comprises the following steps:

ω_i(x)＝C_i-1(x)-C_i(x) (3)

in the formula, ω_i(x) Is the wavelet coefficient at the scale i; c_i(x) Is an approximate signal at the scale i; h is a low-pass filter which satisfies the following equation with the scale function ψ (x):

atrous discrete wavelet decomposes the signal to generate a set of adjacent wavelet coefficients { omega } with different resolution_iAnd the approximation signal.