An Ultrasonic Detection Method for Gas Leaks at Ultra-Long Distance

Abstract

The present invention discloses an ultra-long-distance gas leakage ultrasonic detection method, including a sound focusing structure, data pre-processing, background noise processing, detection signal processing, and leakage signal judgment processing. In the present invention, by continuously collecting leakage ultrasonic signals transmitted through the air, a time-frequency analysis of the signal data is proposed, and the peak values ​​of the characteristic frequency band are accumulated and averaged in the time domain as the normal limit value of the background noise. The detection signal is then analyzed in the frequency domain, and the cumulative peak values ​​of the characteristic values ​​in the time domain are judged, and the positioning is performed through inspection. At the same time, an interface with a mixed transition of a parabola and a hyperboloid is used to form an outward-expanding speaker sound focusing tube, and an aluminum-plated film is attached to its surface to reflect and focus the ultrasonic signals transmitted over a long distance to the focal plane of the curved surface, thereby realizing ultra-long-distance detection of weak signals. At the same time, in order to improve the sensitivity of signal detection, a sensor with multiple probes connected in series is set at the focus.

Description

Technical Field

[0001] The invention relates to the technical field of sealing performance testing, and in particular to an ultra-long-distance gas leakage ultrasonic detection method. Background Art

[0002] At present, the performance of non-contact portable ultrasonic leak detectors is mainly detection sensitivity, and its long-distance detection is widely used in sealing performance detection of power plants, airtight structures, etc. It mainly directly reflects the value at a specific frequency by collecting the data received by the ultrasonic sensor in real time, and judges the occurrence of leakage by the change of the value. The detection method is fixed and cannot change with the change of distance and detection size, which is easy to affect the leakage detection sensitivity of the instrument. Ultra-long-distance detection is directly related to the sensitivity of the sensor, the signal-to-noise ratio of the preamplifier circuit, the circuit signal amplification performance, and the detection and recognition algorithm.

[0003] The existing ultra-long-distance ultrasonic detection method for gas leaks still has many shortcomings. First, the received signal is directly amplified and the sound pressure value or sound signal intensity at a specific frequency is obtained after frequency analysis. Under normal circumstances, for small leaks, the detectable distance is very limited, only a few meters. Secondly, to achieve ultra-long-distance ultrasonic leakage detection, it is necessary to use air-coupled sensors with very high sensitivity, increase the number of sensors, increase the receiving surface to 500mm in diameter, and even change the preamplifier circuit. Not only does it put forward high requirements on circuit performance and sensor performance, but the detection results are often not ideal.

[0004] Therefore, it is necessary to seek a simple and direct algorithm and method that can effectively reduce the requirements on the performance of circuits, sensors, etc. Under the performance of conventional ultrasonic leak detectors, ultra-long-distance ultrasonic leak detection can be achieved by changing the internal algorithm. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the ultra-long-distance gas leakage ultrasonic detection method in the past, which is difficult to implement, has high hardware requirements, and is not accurate, so as to propose an ultra-long-distance gas leakage ultrasonic detection method based on time domain accumulation. This measurement method can be applied to ultra-long-distance detection developed by various domestic ultrasonic leak detectors. During the detection process, there is no need to replace the detection device, and the test performance index of ultra-long-distance leak detection sensitivity can be directly achieved through inspection, which greatly improves the detection sensitivity, reduces the hardware cost, saves a lot of test time and funds, and provides an outward expansion sound-gathering structure based on a mixed transition of a parabola and a hyperboloid, which can improve the sound-gathering effect and positioning accuracy in space.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] An ultra-long-distance gas leakage ultrasonic detection method includes a sound focusing structure, data pre-processing, background noise processing, detection signal processing and leakage signal judgment processing; the data pre-processing is to pre-amplify and filter the signal detected by the air-coupled ultrasonic sensor, pass through an AD converter, sample at a sampling rate of 3M / s, send the sampled digital signal to FPGA for processing, perform software filtering pre-processing on the processed signal, and perform FFT transformation; the background noise processing is to continuously detect the signal for 1 minute in a leakage-free signal or detection background acoustic environment as the leakage-free background noise, and the specific processing method is to use FFT transformation to accumulate the peak value of the characteristic value of 37kHz-43kHz in the frequency domain for 1 minute, and then average it as the background noise effective value; the detection signal processing refers to accumulating the peak signal of the characteristic value of 100 cycles of the detection signal as the effective value of the detection signal during detection, and the leakage signal judgment processing is to subtract the effective value of the detection signal from 100 times the effective value of the background noise to obtain a relative value, and when the relative value is positive and greater than 1.5 times the effective value of the background noise, the presence of leakage is effectively detected.

[0008] Preferably, the pre-amplification and filtering processing of the data pre-processing is implemented by circuit hardware, with a two-stage amplification circuit, the first stage is fixed at 100 times, and the second stage is variable at 100 times, achieving a 10,000-fold amplification factor, and filtering at 20kHz to 100kHz is performed through a bandpass filter.

[0009] Preferably, the background noise processing is to automatically and continuously monitor the data for 1 minute after starting up, perform time-frequency analysis FFT, accumulate and average the data, and take the average as the effective value of the background noise without leakage.

[0010] Preferably, the detection signal processing refers to automatically accumulating the peak signals of the characteristic values ​​of 100 cycles of the detection signal after the background noise is self-determined 1 minute after power-on and the sensitivity detection mode is executed. The accumulated value specifically refers to the peak voltage value of the characteristic value between 37kHz-43kHz in the frequency domain after FFT of each detection cycle signal. 100 characteristic value data points are accumulated in the time domain as the effective value of the detection signal.

[0011] Preferably, the detection signal is collected continuously in real time, using a push-pop mode to incorporate the newly collected data into a 100-accumulation point data matrix, with the earliest collected data exiting the matrix, thereby ensuring the real-time nature of signal detection and effectively locating the location of gas leakage.

[0012] Preferably, the leakage signal judgment processing is to subtract the effective value of the detection signal from 100 times the effective value of the background noise to obtain a relative value. If the relative value is a negative number, it is automatically judged and displayed as a random number of 0 to 1. If the relative value is a positive number, the actual value is displayed. When the actual value is greater than 1.5 times the average value of the background noise, an alarm is triggered.

[0013] Preferably, the numerical value obtained by the leakage signal judgment processing is a relative value rather than an actual signal value; it is used to judge the occurrence and location of leakage, and its size is proportional to the size of the leakage hole.

[0014] Preferably, the sound-gathering structure is a 3D printed one-piece structure, and the sound-gathering structure includes an end mounting end, a front detection port, a parabolic curve surface, a hyperbolic curve surface, a transition surface and an aluminum-plated film. The sound-gathering structure is calculated by the method of total reflection of sound waves to obtain a focus, and the focus is the receiving position of the sensor. When the sound-gathering structure is put on the sensor probe, the sound waves can be effectively gathered; the sound-gathering structure is an outward-expanding trumpet mouth, and its inner surface is a smooth transition from a parabola to a hyperbolic surface; the material is nylon 1010, coated with black paint on the outside, and an aluminum-plated film is attached to the inner surface. The focus of the sound-gathering structure is the calculated design focus. When the sleeve is put on the sensor probe, the sensor receiving surface is at the focus of the sound-gathering structure.

[0015] Preferably, the terminal mounting end of the sound-focusing structure is connected to the sensor probe, and the inner diameter of the terminal end is the outer diameter of the sensor probe.

[0016] An ultra-long-distance gas leakage ultrasonic detection method comprises the following steps:

[0017] S1: An outward-expanding sound-gathering structure based on a mixed transition of a parabola and a hyperboloid is placed on the front end of the sensor probe of the ultrasonic leak detector, so that the sensor receiving surface is at the focus of the sound-gathering structure;

[0018] S2: Turn on the leak detector, and the received signal is pre-amplified and then AD converted and transmitted to the FPGA to complete the data pre-processing of signal amplification and filtering;

[0019] S3: Real-time continuous monitoring of the signal for 1 minute, after FFT processing by FPGA, the data is transmitted to the ARM core for cumulative averaging of the peak values ​​of the eigenvalues ​​of 37kHz-43kHz in the frequency domain to complete the background noise processing;

[0020] S4: Adjust to the sensitivity test mode, perform data acquisition and processing for 100 cycles continuously, and accumulate the peak signal of the characteristic value as the effective detection signal;

[0021] S5: Leakage signal judgment processing: Compare the effective detection signal data matrix with the background noise data matrix; subtract the effective value of the detection signal from 100 times the effective value of the background noise; process the obtained relative value, set negative numbers to random numbers between 0 and 1, and positive numbers to be actually greater than the background noise; realize real-time collection and processing through the first-in-first-out principle to ensure that the detection data during inspection can effectively reflect the value higher than the background noise at the center of the leakage.

[0022] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0023] 1. In this application, the weak leakage ultrasonic signal is used to utilize the cumulative effect of the time domain. Through high-speed acquisition and time domain accumulation, the effective signal that is difficult to distinguish under the background noise is accumulated and revealed, thereby achieving effective resolution of the tiny signal. Moreover, its algorithm is relatively simple, occupies little hardware resources, and has low hardware requirements, which overcomes the problem of high hardware requirements of the past direct detection method and solves the problem of ultra-long-distance detection of tiny leaks. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of an ultra-long-distance gas leakage ultrasonic detection method provided in accordance with an embodiment of the present invention is shown;

[0025] Figure 2 A schematic diagram of an outward-expanding sound focusing structure based on a hybrid transition of a parabola and a hyperbola provided according to an embodiment of the present invention is shown.

[0026] Legend:

[0027] 1. Terminal installation end; 2. Front detection port; 3. Parabolic curve surface; 4. Hyperbolic curve surface; 5. Transition surface; 6. Aluminum-plated film; 7. Sound-gathering structure; 8. Data pre-processing; 9. Background noise processing; 10. Detection signal processing; 11. Leakage signal judgment processing. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] See also Figure 1-2 , the present invention provides a technical solution:

[0030] An ultra-long-distance gas leakage ultrasonic detection method includes a sound focusing structure 7, data pre-processing 8, background noise processing 9, detection signal processing 10 and leakage signal judgment processing 11; the data pre-processing 8 is to pre-amplify and filter the signal detected by the air-coupled ultrasonic sensor, pass through an AD converter, sample at a sampling rate of 3M / s, send the sampled digital signal to FPGA for processing, perform software filtering pre-processing on the processed signal, and perform FFT transformation; the background noise processing 9 is to continuously detect for 1 minute in the absence of leakage signal or detection background acoustic environment. The signal is used as the background noise without leakage. The specific processing method is to use FFT transformation to accumulate the peak values ​​of the characteristic values ​​of 37kHz-43kHz in the frequency domain for 1 minute, and then average them as the effective value of the background noise; the detection signal processing 10 refers to accumulating the peak signals of the characteristic values ​​of 100 cycles of the detection signal during detection as the effective value of the detection signal, and the leakage signal judgment processing 11 is to subtract the effective value of the detection signal from 100 times the effective value of the background noise. When the relative value is positive and greater than 1.5 times the background, the existence of leakage is effectively detected;

[0031] The sound-gathering structure 7 is installed on the probe of the ultrasonic leak detector. The ultrasonic signal propagated in the space collected by the sound-gathering structure 7 is detected by the sensor and then subjected to data pre-processing 8. The received ultrasonic signal is amplified, filtered and AD-converted and transmitted to the FPGA. The FPGA then performs background noise processing 9. The cumulative average of the peak values ​​of the frequency domain characteristic values ​​(37kHz-43kHz) of the FFT of the signal 1 minute after startup is taken as the effective value of the background noise, and this noise level is taken as an indication of no leakage. Then the FPGA performs detection signal processing 10, enters the detection mode 1 minute after the power is turned on, and when ultra-long-distance detection is required, it switches to the sensitivity detection algorithm mode, performs data acquisition processing for 100 consecutive cycles, accumulates the peak signal of the characteristic value as an effective detection signal, and the signal uses a first-in-first-out stack method to update the data matrix in real time; then the leakage signal judgment processing 11 is performed, the leakage signal judgment processing 11 compares the effective detection signal data matrix with the background noise data matrix, and subtracts the effective value of the detection signal from 100 times the effective value of the background noise; the relative value obtained is processed, and negative numbers are set to random numbers of 0-1, and positive numbers are actually greater than the background noise. When the relative value is positive and greater than 1.5 times the background, the presence of leakage is effectively detected.

[0032] Specifically, Figure 1 and Figure 2As shown, the pre-amplification and filtering processing of the data pre-processing 8 is realized by circuit hardware, and has a two-stage amplification circuit. The first stage is fixed at 100 times, and the second stage is variable at 100 times, realizing a 10,000-fold amplification factor. It is filtered at 20kHz to 100kHz through a band-pass filter. The background noise processing 9 is to automatically continuously monitor the data for 1 minute after power-on, perform time-frequency analysis FFT, and then accumulate and average it as the effective value of the background noise without leakage. The detection signal processing 10 refers to the automatic accumulation of the peak signal of the characteristic value of 100 cycles of the detection signal after the background noise is set 1 minute after power-on, and the operation executes the sensitivity detection mode. The accumulated value specifically refers to the peak voltage value of the characteristic value between 37kHz and 43kHz in the frequency domain after the FFT of each detection cycle signal, which is performed within the scope of the time domain. The accumulation of 100 characteristic value data points is used as the effective value of the detection signal. The detection signal is collected in real time and continuously. The push-pop mode is adopted to incorporate the newly collected data into the 100 accumulation point data matrix. The earliest collected data is exited to ensure the real-time nature of signal detection and effectively locate the position of gas leakage. The leakage signal judgment processing 11 is to subtract the effective value of the detection signal from 100 times the effective value of the background noise to obtain a relative value. If the relative value is a negative number, it is automatically judged to be displayed as a random number of 0 to 1. If the relative value is a positive number, the actual value is displayed. When the actual value is greater than 1.5 times the effective value of the background noise, an alarm is triggered. The value obtained by the leakage signal judgment processing 11 is a relative value, not an actual signal value; it is used to judge the occurrence and location of leakage, and its size is proportional to the size of the leak.

[0033] Specifically, Figure 1 and Figure 2 As shown, the sound-gathering structure 7 is a 3D printed integrated structure, and the sound-gathering structure 7 includes an end mounting end 1, a front detection port 2, a parabolic curve surface 3, a hyperbolic curve surface 4, a transition surface 5 and an aluminum-plated film 6. The sound-gathering structure 7 is calculated by the method of total reflection of sound waves to obtain a focus, and the focus is the receiving position of the sensor. When the structure is put on the probe, the sound waves can be effectively gathered;

[0034] The sound-focusing structure 7 is an outward-expanding trumpet-shaped mouth, and its inner surface is first a parabola and then a hyperbolic surface with a smooth transition; the material of the sound-focusing structure 7 is nylon 1010, and the outer surface is sprayed with black paint. The outer surface is a conical trumpet-shaped mouth, and the inner surface is attached with an aluminum-plated film 6. The inner surface is first a parabolic curve surface 3, which smoothly transitions to a hyperbolic surface 4 and is then connected to a straight tube section at the end. The straight tube section marks the focal position and is directly mounted on the probe so that the sensor receiving surface coincides with the focal position. The focus of the sound-focusing structure 7 is the calculated design focus. When it is mounted on the sensor head, the sensor receiving surface is at the focus of the structure. The terminal mounting end 1 of the sound-focusing structure 7 is connected to the sensor probe, and the inner diameter of its end is the outer diameter of the sensor probe.

[0035] An ultra-long-distance gas leakage ultrasonic detection method, characterized in that it comprises the following steps:

[0036] S1: an outward-expanding sound-gathering structure 7 based on a hybrid transition of a parabola and a hyperboloid is placed on the front end of the ultrasonic leak detector probe, so that the sensor receiving surface is at the focus of the sound-gathering structure 7;

[0037] S2: Turn on the leak detector, and the received signal is pre-amplified and then AD converted and transmitted to the FPGA, completing the data pre-processing of signal amplification and filtering 8;

[0038] S3: Real-time continuous monitoring of the signal for 1 minute, after FFT processing through FPGA, the data is transmitted to the ARM core for cumulative averaging of characteristic values ​​37kHz-43kHz to complete the background noise processing 9;

[0039] S4: Adjust to the sensitivity test mode, perform data acquisition and processing for 100 cycles continuously, and accumulate the peak signal of the characteristic value as the effective detection signal;

[0040] S5: Leakage signal judgment processing 11 compares the effective detection signal data matrix with the background noise data matrix; subtracts the effective value of the detection signal from 100 times the effective value of the background noise; processes the obtained relative value, sets negative numbers to random numbers between 0 and 1, and positive numbers are actually greater than the background noise; through the principle of first in-first out, real-time collection and processing are achieved to ensure that the detection data during inspection can effectively reflect the value higher than the background noise at the center of the leakage.

[0041] Through inspection, the center of the leakage is determined based on the size of the relative comparison value. This method effectively collects, accumulates, compares and judges weak signals at ultra-long distances, overcomes the problems of weak signals being difficult to identify and high circuit performance requirements, and solves the problem of locating leakage points at ultra-long distances.

[0042] The above description of the embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An ultra-long distance gas leakage ultrasonic detection method, characterized in that: The invention comprises a sound gathering structure (7), data pre-processing (8), background noise processing (9), detection signal processing (10) and leakage signal judgment processing (11); the data pre-processing (8) is to pre-amplify and filter the signal detected by the air-coupled ultrasonic sensor, sample it at a sampling rate of 3M / s through an AD converter, send the sampled digital signal to FPGA for processing, perform software filtering pre-processing on the processed signal, and perform FFT transformation; the background noise processing (9) is to detect the signal continuously for 1 minute in the absence of leakage signal or detection background acoustic environment as the background noise of no leakage. The specific processing method is to use FFT transformation to accumulate the peak values ​​of the characteristic values ​​of 37kHz-43kHz in the frequency domain for 1 minute, and then average them as the effective value of the background noise; the detection signal processing (10) refers to accumulating the peak signals of the characteristic values ​​of 100 cycles of the detection signal during detection as the effective value of the detection signal, and the leakage signal judgment processing (11) is to subtract the effective value of the detection signal from 100 times the effective value of the background noise to obtain a relative value. When the relative value is positive and greater than 1.5 times the effective value of the background noise, the presence of leakage is effectively detected.

2. The method for ultra-long-distance gas leakage ultrasonic detection according to claim 1, characterized in that: The pre-amplification and filtering processing of the data pre-processing (8) is implemented by circuit hardware, and has a two-stage amplification circuit, the first stage is fixed at 100 times, and the second stage is variable at 100 times, achieving a 10,000-fold amplification factor, and filtering at 20kHz~100kHz is performed through a bandpass filter.

3. The method for ultra-long-distance gas leakage ultrasonic detection according to claim 1, characterized in that: The background noise processing (9) is to automatically and continuously monitor the data for 1 minute after starting up, perform time-frequency analysis FFT, accumulate and average the data, and use it as the background noise without leakage.

4. The method for ultra-long-distance gas leakage ultrasonic detection according to claim 1, characterized in that: The detection signal processing (10) refers to automatically accumulating the peak signals of the characteristic values ​​of 100 cycles of the detection signal after the background noise is self-determined 1 minute after the power is turned on, and after the sensitivity detection mode is executed. The accumulated value specifically refers to the peak voltage value of the characteristic value between 37kHz and 43kHz in the frequency domain after FFT of each detection cycle signal. 100 characteristic value data points are accumulated in the time domain as the effective value of the detection signal.

5. The method for ultra-long-distance gas leakage ultrasonic detection according to claim 1, characterized in that: The detection signal is collected continuously in real time, using a push-pop mode to incorporate the newly collected data into a 100-accumulation point data matrix, with the earliest collected data exiting the matrix, ensuring the real-time nature of signal detection and effectively locating the location of gas leaks.

6. The method for ultra-long-distance gas leakage ultrasonic detection according to claim 1, characterized in that: The leakage signal judgment process (11) is to subtract the effective value of the detection signal from 100 times the effective value of the background noise to obtain a relative value. If the relative value is a negative number, it is automatically judged to be displayed as a random number between 0 and 1. If the relative value is a positive number, the actual value is displayed. When the actual value is greater than 1.5 times the effective value of the background noise, an alarm is triggered.

7. The method for ultra-long-distance gas leakage ultrasonic detection according to claim 1, characterized in that: The numerical value obtained by the leakage signal judgment processing (11) is a relative value, not an actual signal value; it is used to judge the occurrence and location of leakage, and its size is proportional to the size of the leakage hole.

8. The method for ultra-long-distance gas leakage ultrasonic detection according to claim 1, characterized in that: The sound-gathering structure (7) is a 3D-printed one-piece structure, comprising a terminal mounting end (1), a front detection port (2), a parabolic curve surface (3), a hyperbolic curve surface (4), a transition surface (5) and an aluminum-plated film (6). The sound-gathering structure (7) is calculated by a method of total reflection of sound waves to obtain a focus, and the focus is the receiving position of the sensor. When the sound-gathering structure is put on the sensor probe, the sound waves can be effectively gathered; The sound-gathering structure (7) is an outward-expanding bell mouth, the inner surface of which is a smooth transition from a parabola to a hyperbolic surface; the material is nylon 1010, the outer surface is painted black, and the inner surface is attached with an aluminum-plated film (6); the focus of the sound-gathering structure (7) is a calculated design focus, and when the outer sleeve is mounted on the sensor probe, the sensor receiving surface is at the focus of the sound-gathering structure.

9. The method for ultra-long-distance gas leakage ultrasonic detection according to claim 8, characterized in that: The terminal mounting end (1) of the sound focusing structure (7) is connected to the sensor probe, and the inner diameter of the terminal end is the outer diameter of the sensor probe.

10. The method for ultra-long-distance gas leakage ultrasonic detection according to claim 1, characterized in that: The following steps are involved: S1: an outward-expanding sound-gathering structure (7) based on a hybrid transition of a parabola and a hyperbola is placed on the front end of a sensor probe of an ultrasonic leak detector, so that the receiving surface of the sensor is at the focus of the sound-gathering structure (7); S2: Turn on the leak detector, receive the signal, and transmit it to the FPGA after pre-amplification and AD conversion, completing the data pre-processing of signal amplification and filtering (8); S3: Real-time continuous monitoring of the signal for 1 minute, after FFT processing by FPGA, the data is transmitted to the ARM core for cumulative averaging of the peak values ​​of the eigenvalues ​​of 37kHz-43kHz in the frequency domain to complete the background noise processing (9); S4: Adjust to the sensitivity test mode, perform data acquisition and processing for 100 cycles continuously, and accumulate the peak signal of the characteristic value as the effective value of the detection signal; S5: Leakage signal judgment processing (11) Compare the effective detection signal data matrix with the background noise data matrix; subtract the effective value of the detection signal from 100 times the effective value of the background noise; process the obtained relative value, set negative numbers to random numbers between 0 and 1, and positive numbers to the actual value greater than the background noise; realize real-time collection and processing through the principle of first-in-first-out, and ensure that the detection data during inspection can effectively reflect the value higher than the background noise at the center of the leakage.

Images