FPGA-based active and passive cooperative ultrasonic vehicle-mounted pipeline inspection monitoring method and system

Abstract

The application discloses a kind of based on FPGA's active and passive collaborative ultrasonic vehicle-mounted pipeline inspection monitoring method and system, first combination of each hardware module of flaw detection inspection trolley, so that the trolley can walk, send excitation signal, and build signal receiving array on pipeline wall;Then passive monitoring pipeline state using signal receiving array, when receiving defect signal, feedback defect position to trolley, or use FPGA to control trolley movement, with pipeline wall coupling with excitation sensor at a certain angle, actively detect pipeline state;After obtaining defect position, trolley fixed-point motion gathers defect depth information, forms data set and carries out wavelet transform, extracts time-frequency domain feature, establishes neural network model and carries out training and prediction, and outputs final result.The ultrasonic vehicle-mounted pipeline inspection monitoring system of the application uses FPGA to realize the collaborative control of structural health monitoring and nondestructive testing, improves the intelligent level, saves economic cost, and lays a good foundation for further processing of water delivery pipeline ultrasonic damage detection.

Description

Technical Field

[0001] This invention relates to the field of pipeline structure damage monitoring and non-destructive testing, and in particular to an FPGA-based active-passive collaborative ultrasonic vehicle-mounted pipeline inspection and monitoring method and system. Background Technology

[0002] Pipelines, as a vital transportation system, have made outstanding contributions to my country's economic development. However, with increasing service life, internal damage can occur in pipelines, posing a threat to public safety. Therefore, pipeline inspection has become a key engineering focus. Ultrasonic non-destructive testing (NDT) technology, due to its advantages of strong penetration and high sensitivity, is widely used in damage detection for underground pipelines. Underground pipeline damage detection based on ultrasonic NDT and structural health monitoring technologies can conveniently reflect the pipeline's condition, providing decision support for subsequent pipeline maintenance and repair.

[0003] Due to limitations in non-destructive testing (NDT) and health monitoring technologies, as well as the complexity of ultrasonic wave propagation within pipelines, ultrasonic testing and real-time monitoring of underground pipelines suffer from low efficiency, high cost, and low levels of intelligence, hindering their effective use. In recent years, the increasing demand for pipeline damage detection across various sectors has led to the emergence of several ultrasonic damage detection technologies. Some have improved detection accuracy by actively detecting pipeline damage using phased array technology, but these still require point-by-point inspection, resulting in low efficiency and an inability to inspect long-distance or inaccessible pipeline systems, thus failing to achieve continuous monitoring. The key to solving the problem of applying ultrasonic NDT and health monitoring in real-time scenarios lies in how to improve detection efficiency and intelligence while ensuring accuracy and saving economic costs. Summary of the Invention

[0004] The problem to be solved by this invention is to provide an active-passive collaborative ultrasonic vehicle-mounted pipeline inspection and monitoring method and system based on FPGA, which can improve the accuracy, efficiency and intelligence level of pipeline damage detection, save economic costs, and perform pipeline damage detection better and faster.

[0005] This invention adopts the following technical solution: a method for active-passive coordinated ultrasonic vehicle-mounted pipeline inspection and monitoring based on FPGA, comprising the following steps:

[0006] S1. Set up the inspection system: Combine the hardware modules of the flaw detection and inspection vehicle and build a signal receiving array on the inner wall of the pipeline to be inspected.

[0007] S2. Pipeline Defect Signal Monitoring: The pipeline status is passively monitored using a signal receiving array. When a defect signal is received, the signal receiving array feeds back the defect location to the flaw detection inspection trolley. The flaw detection inspection trolley is controlled by an FPGA to move and the excitation sensor is coupled to the pipeline wall at a certain angle to actively detect the pipeline status. After obtaining the defect location, the flaw detection inspection trolley moves to a fixed point to collect defect depth information and obtain dataset X1.

[0008] S3. Data Transformation and Denoising: Take dataset X1 as input, perform wavelet transform to obtain wavelet coefficients, perform soft and hard thresholding denoising, and use the Marat algorithm to perform wavelet reconstruction to obtain the denoised dataset X2.

[0009] S4. Defect Depth Prediction: Using dataset X2 as input, extract its frequency domain and time domain features, establish a neural network model for training and prediction, obtain the defect depth, and output the final result.

[0010] Furthermore, in step S1, the inspection system is established, including the following sub-steps:

[0011] S1.1 Combine the various hardware modules of the flaw detection inspection trolley to enable the flaw detection inspection trolley to move inside the pipeline under the control of FPGA and send excitation signals through the excitation sensors on the trolley.

[0012] S1.2. Using a coupling agent, piezoelectric wafers are spaced apart on the inner wall of one end of the pipeline to be inspected to form a signal receiving array.

[0013] Furthermore, in step S2, pipeline defect signal monitoring includes the following sub-steps:

[0014] S2.1 In passive monitoring mode, the pipeline status is passively detected by the signal receiving array. When a defect signal is received, the signal receiving array feeds back the defect location to the FPGA. The FPGA controls the movement of the flaw detection inspection trolley. After reaching the defect location, it rotates to excite the sensor angle, couples with the pipeline wall, measures the defect depth, and sends it to the signal receiving array.

[0015] S2.2 In active detection mode, the FPGA controls the movement of the flaw detection inspection trolley through the drive module. It stops at the same distance, adjusts the angle of the excitation sensor by rotation, and couples it with the pipe wall. After completion, it issues an excitation command, as follows;

[0016] S2.2.1 During the inspection process, when the motor needs to be driven, the FPGA control module of the flaw detection inspection trolley outputs a confirmation signal, which is sent to the drive module via the USB serial port.

[0017] S2.2.2 According to actual needs, the FPGA control module determines the PWM wave period and triggering conditions, which are used as input parameters and stored in the adder counter and digital comparator respectively.

[0018] S2.2.3 After the confirmation signal passes through the digital comparator of the FPGA control module and outputs a logic trigger signal, it passes through the adder counter to start the timer interrupt process and outputs a periodic signal.

[0019] S2.2.3 When the periodic signal passes through the microstepping counter of the FPGA control module, it outputs a PWM wave with a defined pulse width and duty cycle. The PWM wave passes through the drive circuit and outputs corresponding signal pulses, which in turn control the motor to operate, so that the flaw detection inspection trolley moves to the designated position and stops.

[0020] S2.2.4 The swing fine-tuning structure of the drive module lifts the excitation sensor with a certain angle and couples it with the pipe wall. After completion, it issues an excitation command.

[0021] S2.3 After receiving the excitation command, the excitation module of the flaw detection inspection trolley sends a sound beam to the pipeline. When a defect is detected, the defect location causes the ultrasonic wave to be reflected, scattered, and attenuated. Utilizing its attenuation characteristics, the signal receiving array receives the waveforms around the excitation point and analyzes them to obtain the location of the pipeline defect, as detailed below:

[0022] S2.3.1 The FPGA control module of the flaw detection inspection trolley imports the quantized data of the obtained waveform into the waveform memory according to the waveform function expression, and saves the specified signal frequency in the frequency synthesizer.

[0023] S2.3.2 During the inspection process, when an excitation is required, the address counter in the FPGA control module of the flaw detection inspection trolley sequentially reads the data in the memory, and outputs the desired waveform data through the frequency synthesizer.

[0024] S2.3.3 The output digital waveform data is sent to the digital-to-analog converter to be converted into an analog waveform signal. The analog waveform signal passes through the precision load resistor R to convert the output current signal into a voltage signal, and then outputs the AC signal in a differential output manner.

[0025] S2.3.3 The AC signal passes through the intensity algorithm control unit to obtain a gain control digital signal, which is then converted into an analog voltage signal by a digital-to-analog converter.

[0026] S2.3.4. The analog voltage signal is passed through a variable gain amplifier to obtain an analog waveform signal. The high-speed operational amplifier outputs an analog waveform signal with a fixed gain value, which is then sent to a push-pull power amplifier to output an analog signal waveform with a fixed bandwidth to the array element.

[0027] S2.3.5. The array element emits a sound beam into the pipe according to the emission focusing law. Since the excitation sensor has a certain tilt angle, the emitted sound beam is decomposed into ultrasonic waves in the horizontal direction and ultrasonic waves in the vertical direction.

[0028] S2.3.6 When encountering defects or damage in the material, the defect location will cause ultrasonic waves to be reflected, scattered, and attenuated. That is, the ultrasonic waves will propagate in different directions in the material around the damage and will be weakened. Using the signal receiving array around the excitation point, the attenuation degree of ultrasonic waves at different locations is measured and compared, the scattering mode and intensity are analyzed, the location of the pipe defect is obtained, and the result is fed back to the signal receiving array.

[0029] S2.4 After the signal receiving array receives the location of the defect, repeat step S2.1 to measure the depth of the pipe defect;

[0030] S2.5 Combine the defect depth and defect location to form dataset X1, which is used for subsequent signal processing and model training.

[0031] Furthermore, in step S3, the data transformation and denoising process includes the following sub-steps:

[0032] S3.1. Using dataset X1 as input, analyze the signal through discrete wavelet transform. Expand dataset X1 under the wavelet basis to obtain wavelet coefficients W(a,b), where a is the scaling parameter and b is the translation factor, used to control the scale and position of the wavelet function.

[0033] S3.2. Using W(a,b) as input, soft threshold denoising and hard threshold denoising are used to remove signal interference to obtain the denoised wavelet coefficients W(a,b)'. Specifically: when the signal contains a lot of low-frequency noise, hard threshold denoising is used, and when it contains a lot of high-frequency noise, soft threshold denoising is used.

[0034] S3.3. Using W(a,b)' as input, perform wavelet reconstruction using the Marat algorithm to obtain the denoised dataset X2.

[0035] The technical solution of the present invention also includes: an FPGA-based active-passive cooperative ultrasonic vehicle-mounted pipeline inspection and monitoring system, used to implement the active-passive cooperative ultrasonic vehicle-mounted pipeline inspection and monitoring method described in any of the above claims, including: the pipeline to be inspected, a flaw detection and inspection vehicle, a signal receiving array, and a computer processor;

[0036] The signal receiving array is built on the inner wall of one end of the pipeline to be inspected. It is used to receive the defect signals of the pipeline and feed back the defect location to the flaw detection and inspection trolley.

[0037] The flaw detection inspection trolley includes an FPGA control module, a drive module, an excitation module, and a power supply module. The excitation module also includes a rotatable excitation sensor. Under the control of the FPGA control module, the flaw detection inspection trolley moves inside the pipeline. By rotating, it adjusts the angle of the excitation sensor, couples with the pipeline wall, and sends out excitation commands.

[0038] The computer processor performs wavelet transform and denoising on the data, extracts its frequency and time domain features, establishes a neural network model for training and prediction, obtains the defect depth, and outputs the final result.

[0039] Furthermore, inside the flaw detection and inspection vehicle, the FPGA control module is connected to the microcontroller via a USB interface, exchanges information via USB serial communication, and then controls the motor drive via the drive module. When the motor needs to be driven, the FPGA control module outputs an acknowledgment signal, which reaches the microcontroller's drive module via the USB serial port, driving the flaw detection and inspection vehicle to move to the designated position inside the pipeline.

[0040] Furthermore, inside the flaw detection and inspection vehicle, the on-chip structure of the FPGA control module includes: a frequency synthesizer, an address counter, a waveform memory, and an intensity algorithm control unit. The intensity algorithm control unit is connected to the programmable amplifier on the excitation module and is used to control the movement of the vehicle and to generate excitation.

[0041] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects:

[0042] 1. This invention proposes an active-passive collaborative ultrasonic vehicle-mounted pipeline inspection and monitoring method and system based on FPGA. It applies ultrasonic waves to a flaw detection and inspection vehicle, reducing the high signal-to-noise ratio caused by ultrasonic guided wave dispersion. At the same time, it solves the problems of difficulty in ultrasonic guided wave propagation in actual pipelines and the inability of personnel to enter the pipeline. The system is convenient, easy to operate, low in cost, and has a wide range of applications.

[0043] 2. The active-passive coordinated ultrasonic vehicle-mounted pipeline inspection and monitoring method of this invention adopts an active-passive coordinated approach, combining structural health monitoring with non-destructive testing. It can operate in both passive and active modes. In passive mode, the trolley is controlled to move at a fixed point to measure the defect depth upon detection. In active mode, the trolley is controlled to position excitation sensors at a specific angle to measure the defect location. This achieves real-time monitoring of the pipeline while accurately measuring defect depth, improving detection efficiency and reducing errors. It solves the problems of passive measurement alone being unable to achieve all-time monitoring and active measurement alone generating numerous data points that are difficult to pinpoint accurately. It provides richer structural damage information, laying a solid foundation for subsequent processing.

[0044] 3. The active-passive coordinated ultrasonic vehicle-mounted pipeline inspection and monitoring system of this invention adopts a digital waveform transmission method based on FPGA, which can select excitation signals of arbitrary intensity and waveform, laying the foundation for intensity focusing or deflection of ultrasonic phased array at defects, improving detection resolution and signal-to-noise ratio. At the same time, the excitation module and drive module of the vehicle are integrated together and controlled by FPGA, which can better achieve active-passive coordinated operation, improve the system intelligence level, prevent signal crosstalk, reduce errors, and reduce costs. Attached Figure Description

[0045] Figure 1 This is a flowchart of the FPGA-based active-passive collaborative ultrasonic vehicle-mounted pipeline inspection and monitoring method of the present invention;

[0046] Figure 2 This is a diagram showing the composition of the inspection trolley in the FPGA-based active-passive collaborative ultrasonic vehicle-mounted pipeline inspection and monitoring system of the present invention.

[0047] Figure 3 This is a schematic diagram of the signal excitation of the FPGA-based active-passive cooperative ultrasonic vehicle-mounted pipeline inspection and monitoring system of the present invention.

[0048] Figure 4 This is a neural network model diagram of the FPGA-based active-passive collaborative ultrasonic vehicle-mounted pipeline inspection and monitoring system of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the application will be further described in detail below with reference to the accompanying drawings. The described embodiments are only a part of the embodiments involved in this invention. All non-innovative embodiments based on these embodiments by other researchers in the art are within the protection scope of this invention. Furthermore, the step numbers in the embodiments of this invention are only set for ease of explanation and do not limit the order of the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0050] like Figure 1 As shown, this invention provides an FPGA-based active-passive cooperative ultrasonic vehicle-mounted pipeline inspection and monitoring method, comprising the following steps:

[0051] S1. Assemble the various hardware modules of the combined flaw detection and inspection trolley to enable the trolley to move and emit excitation signals. At the same time, set up a signal receiving array at the pipe opening to complete the overall system construction, as detailed below:

[0052] S1.1 Combine the various hardware modules of the flaw detection inspection trolley to enable the flaw detection inspection trolley to move inside the pipeline under the control of FPGA and send excitation signals through the excitation sensors on the trolley.

[0053] The flaw detection and inspection trolley runs along the pipe wall; the trolley's hardware system is as follows: Figure 2 As shown, it mainly consists of a control module, an excitation module, a drive module, and a power supply module.

[0054] The function of the excitation module is to modulate the analog waveform signal emitted by the control module into an analog waveform signal with a certain strength and power. It consists of a digital-to-analog converter, a programmable amplifier, a power amplifier, and an excitation sensor.

[0055] The drive module has two functions. First, it controls the movement of the trolley, consisting of a microcontroller and a motor drive voltage regulator module. Second, it is responsible for driving the excitation sensor to move left and right and up and down to complete specific tasks, which is called the swing fine-tuning structure.

[0056] The control module is the central module of the system. It adopts a programmable logic device (FPGA) and its main functions are to generate digital excitation signals and construct intensity compensation algorithms, output specific analog waveform signals and compensation intensity parameters; generate motion confirmation signals; control the swing fine-tuning structure of the drive module to drive the excitation module to emit ultrasonic signals; and control the drive module to realize the functions of the car's forward, backward, left and right turns, and stop, that is, to realize task allocation and control of the entire drive process.

[0057] S1.2. Using a coupling agent, piezoelectric wafers are spaced apart on the inner wall of one end of the pipeline to be inspected to form a signal receiving array.

[0058] S2. Pipeline Defect Signal Monitoring: The pipeline status is passively monitored using a signal receiving array. When a defect signal is received, the array feeds back the defect location to the flaw detection inspection trolley. The FPGA controls the movement of the flaw detection inspection trolley, coupling the excitation sensor to the pipeline wall at a certain angle to actively detect the pipeline status. After obtaining the defect location, the flaw detection inspection trolley moves to a fixed point to collect defect depth information, obtaining dataset X1. This includes the following sub-steps:

[0059] S2.1 In passive monitoring mode, the pipeline status is passively detected by the signal receiving array. When a defect signal is received, the signal receiving array feeds back the defect location to the FPGA. The FPGA controls the movement of the flaw detection inspection trolley. After reaching the defect location, it rotates to excite the sensor angle, couples with the pipeline wall, measures the defect depth, and sends it to the signal receiving array.

[0060] S2.2 In active detection mode, the FPGA controls the movement of the flaw detection inspection trolley through the drive module. It stops at the same distance every time, adjusts the angle of the excitation sensor by rotation, and couples with the pipe wall. After completion, it sends an excitation command.

[0061] The FPGA connects to the microcontroller via a USB interface and uses serial communication for information exchange between the host computer and the slave computer. Then, the motor is driven and controlled by the drive unit.

[0062] During the inspection process, when the motor needs to be driven, the FPGA control unit outputs a confirmation signal, which is transmitted to the microcontroller's drive module via the USB serial port.

[0063] Based on actual needs, the PWM wave period and triggering conditions are determined and stored as input parameters in the adder counter and digital comparator, respectively.

[0064] The confirmation signal passes through a digital comparator, outputs a logic trigger signal, and then passes through an adder counter to start a timer interrupt process, outputting a periodic signal.

[0065] When the periodic signal passes through the microstepping counter, it outputs a PWM wave with a defined pulse width and duty cycle, such as... Figure 3 As shown, the PWM wave passes through the drive circuit and outputs corresponding signal pulses, which in turn control the motor to move the car to the designated position and then stop.

[0066] The swing fine-tuning structure of the drive module lifts the excitation sensor at a certain angle, couples it with the pipe wall, and then issues an excitation command.

[0067] S2.3 After receiving the excitation command, the excitation module of the flaw detection inspection trolley sends a sound beam to the pipeline. When a defect is found, the defect location causes the ultrasonic wave to be reflected, scattered, and attenuated. Utilizing its attenuation characteristics, the signal receiving array receives the waveforms around the excitation point and analyzes them to obtain the location of the pipeline defect.

[0068] The on-chip structure of the FPGA includes: a frequency synthesizer, an address counter, a waveform memory, and an intensity algorithm control unit, wherein the intensity algorithm control unit is connected to the programmable amplifier on the excitation module;

[0069] First, based on the waveform function expression, the quantized data of the obtained waveform is imported into the waveform memory in advance, and the specified signal frequency is stored in the frequency synthesizer.

[0070] During the inspection process, when an excitation is required, the address counter sequentially reads the data from the memory, and after passing through the frequency synthesizer, outputs the desired waveform data.

[0071] The output digitized waveform data is sent to a digital-to-analog converter (DAC) to be converted into an analog waveform signal. The analog waveform signal passes through a precision load resistor R, converting the output current signal into a voltage signal, and then outputting an AC signal using a differential output method.

[0072] The AC signal passes through the intensity algorithm control unit to obtain a gain control digital signal, which is then converted into an analog voltage signal by a digital-to-analog converter (DAC).

[0073] The analog voltage signal is passed through a variable gain amplifier to obtain an analog waveform signal with a certain intensity. This signal is then passed through a high-speed operational amplifier, where it is increased to a fixed gain value and fed into a push-pull power amplifier. Finally, an analog signal waveform with increased bandwidth is output to the array elements.

[0074] The array element emits a sound beam into the pipe according to the emission focusing law. Since the excitation sensor has a certain tilt angle, the emitted sound beam can be roughly decomposed into waves in the horizontal direction and waves in the vertical direction.

[0075] When encountering defects or damage in a material, the defect location causes ultrasonic waves to be reflected, scattered, and attenuated. This means the ultrasonic waves propagate in different directions within the material surrounding the damage and weaken. By using a signal receiving array around the excitation point, the attenuation level of the ultrasonic waves at different locations is measured and compared. The scattering patterns and intensities are analyzed to determine the defect location, which is then fed back to the signal receiving array.

[0076] S2.4 After the signal receiving array receives the location of the defect, repeat step S2.1 to measure the depth of the pipe defect;

[0077] S2.5 Combine the defect depth and defect location to form dataset X1, which is used for subsequent signal processing and model training.

[0078] S3. Data Transformation and Denoising: Using dataset X1 as input, wavelet transform is performed to obtain wavelet coefficients. Soft and hard thresholding denoising is then applied, and the Marat algorithm is used for wavelet reconstruction to obtain the denoised dataset X2, as detailed below:

[0079] S3.1. Using dataset X1 as input, perform wavelet transform, expand under the wavelet basis to obtain wavelet coefficients W(a,b), and analyze the signal using discrete wavelet transform (DWT). Expand dataset X1 under the wavelet basis, the expression is:

[0080]

[0081] In the formula, X1(n) is the input dataset, N is the dataset length, W(a,b) are the wavelet coefficients of the signal, a is the scale parameter, and b is the translation factor, which is used to control the scale and position of the wavelet function.

[0082] S3.2. Using W(a,b) as input, soft threshold denoising and hard threshold denoising are used to remove signal interference and obtain the denoised wavelet coefficients W(a,b)'.

[0083] Hard thresholding is used for denoising when the signal contains a lot of low-frequency noise, and soft thresholding is used when it contains a lot of high-frequency noise. The noise is subjected to wavelet transform. Even at the wavelet threshold, it still exhibits strong randomness and is generally considered Gaussian white noise. Compared to the coefficients corresponding to the effective signal, the coefficients corresponding to the noise are very small. Therefore, the soft and hard thresholding methods are selected to process the coefficients of W(a,b), outputting the denoised coefficients W(a,b)', as follows:

[0084] Hard thresholding denoising method: When the absolute value of the wavelet coefficient is less than a given threshold λ, set it to zero; when it is greater than the threshold λ, leave it unchanged. The formula is as follows:

[0085]

[0086] Soft thresholding denoising method: When the absolute value of the wavelet coefficients is less than a given threshold λ, set it to zero; when it is greater than the threshold λ, subtract the threshold from each coefficient. The formula is as follows:

[0087]

[0088] Wherein, sgn(W(a,b)) represents the sign function of W(a,b), that is, it is 1 when W(a,b) is greater than 0, 0 when it is equal to 0, and -1 when it is less than 0.

[0089] S3.3. Using W(a,b)' as input, perform wavelet reconstruction using the Marat algorithm to obtain the denoised dataset X2.

[0090] S4. Defect Depth Prediction:

[0091] like Figure 4 As shown, the dataset X2 is used as input to extract its frequency domain and time domain features. The dataset is divided into a test set, a training set, and a validation set. A neural network model is built for training and prediction. After evaluation, accuracy and recall are calculated to judge the model performance, and the model is optimized. Finally, the depth of pipeline defects is obtained, the final result is output, and pipeline defects are predicted and applied.

[0092] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for active-passive coordinated ultrasonic vehicle-mounted pipeline inspection and monitoring based on FPGA, characterized in that, The steps include the following: S1. Set up the inspection system: Combine the hardware modules of the flaw detection and inspection vehicle and build a signal receiving array on the inner wall of the pipeline to be inspected. S2. Pipeline defect signal monitoring and detection: The pipeline status is passively monitored using a signal receiving array. When a defect signal is received, the signal receiving array feeds back the defect location to the flaw detection inspection vehicle. Using FPGA to control the movement of the flaw detection inspection trolley, the excitation sensor is coupled to the pipe wall at a certain angle to actively detect the pipe status and obtain the defect location. After obtaining the defect location, the flaw detection inspection trolley moves to a fixed point to collect defect depth information and obtain dataset X1. S3. Data Transformation and Denoising: Take dataset X1 as input, perform wavelet transform to obtain wavelet coefficients, perform soft and hard thresholding denoising, and use the Marat algorithm to perform wavelet reconstruction to obtain the denoised dataset X2. S4. Defect Depth Prediction: Using dataset X2 as input, extract its frequency domain and time domain features, build a neural network model for training and prediction, obtain the defect depth, and output the final result. Step S2, pipeline defect signal monitoring, includes the following sub-steps: S2.1 In passive monitoring mode, the pipeline status is passively detected by the signal receiving array. When a defect signal is received, the signal receiving array feeds back the defect location to the FPGA. The FPGA controls the movement of the flaw detection inspection trolley. After reaching the defect location, it rotates to excite the sensor angle, couples with the pipeline wall, measures the defect depth, and sends it to the signal receiving array. S2.2 In active detection mode, the FPGA controls the movement of the flaw detection inspection trolley through the drive module. It stops at the same distance every time, adjusts the angle of the excitation sensor by rotation, and couples with the pipe wall. After completion, it sends an excitation command. S2.3 After receiving the excitation command, the excitation module of the flaw detection inspection trolley sends a sound beam to the pipeline. When a defect is found, the defect location causes the ultrasonic wave to be reflected, scattered, and attenuated. Utilizing its attenuation characteristics, the signal receiving array receives the waveforms around the excitation point and analyzes them to obtain the location of the pipeline defect. S2.4 After the signal receiving array receives the location of the defect, repeat step S2.1 to measure the depth of the pipe defect; S2.5 Combine the defect depth and defect location to form dataset X1, which is used for subsequent signal processing and model training.

2. The FPGA-based active-passive collaborative ultrasonic vehicle-mounted pipeline inspection and monitoring method according to claim 1, characterized in that, Step S1 involves setting up an inspection system, which includes the following sub-steps: S1.1 Combine the various hardware modules of the flaw detection inspection trolley to enable the flaw detection inspection trolley to move in the pipeline under FPGA control and issue excitation commands through the excitation sensors on the trolley. S1.

2. Using a coupling agent, piezoelectric wafers are spaced apart on the inner wall of one end of the pipeline to be inspected to form a signal receiving array.

3. The FPGA-based active-passive collaborative ultrasonic vehicle-mounted pipeline inspection and monitoring method according to claim 1, characterized in that, In step S2.2, the active detection mode is as follows: S2.2.1 During the inspection process, when the motor needs to be driven, the FPGA control module of the flaw detection inspection trolley outputs a confirmation signal, which is sent to the drive module via the USB serial port. S2.2.2 According to actual needs, the FPGA control module determines the PWM wave period and triggering conditions, which are used as input parameters and stored in the adder counter and digital comparator respectively. S2.2.3 After the confirmation signal passes through the digital comparator of the FPGA control module and outputs a logic trigger signal, it passes through the adder counter to start the timer interrupt process and outputs a periodic signal. S2.2.4 When the periodic signal passes through the microstepping counter of the FPGA control module, it outputs a PWM wave with a defined pulse width and duty cycle. The PWM wave passes through the drive circuit and outputs corresponding signal pulses, which in turn control the motor to operate, so that the flaw detection inspection trolley moves to the designated position and stops. S2.2.5 The swing fine-tuning structure of the drive module lifts the excitation sensor with a certain angle, couples it with the pipe wall, and then issues an excitation command.

4. The FPGA-based active-passive coordinated ultrasonic vehicle-mounted pipeline inspection and monitoring method according to claim 3, characterized in that, Step S2.3 includes the following steps: S2.3.1 The FPGA control module of the flaw detection inspection trolley imports the quantized data of the obtained waveform into the waveform memory according to the waveform function expression, and saves the specified signal frequency in the frequency synthesizer. S2.3.2 During the inspection process, when an excitation command needs to be issued, the address counter in the FPGA control module of the flaw detection inspection trolley sequentially reads the data in the memory, and outputs the desired waveform data through the frequency synthesizer. S2.3.3 The output digital waveform data is sent to the digital-to-analog converter to be converted into an analog waveform signal. The analog waveform signal passes through the precision load resistor R to convert the output current signal into a voltage signal, and then outputs the AC signal in a differential output manner. S2.3.4 The AC signal passes through the intensity algorithm control unit to obtain a gain control digital signal, which is then converted into an analog voltage signal by a digital-to-analog converter. S2.3.

5. The analog voltage signal is passed through a variable gain amplifier to obtain an analog waveform signal. The high-speed operational amplifier outputs an analog waveform signal with a fixed gain value, which is then sent to a push-pull power amplifier to output an analog signal waveform with a fixed bandwidth to the array element. S2.3.

6. The array element emits a sound beam into the pipe according to the emission focusing law. Since the excitation sensor has a certain tilt angle, the emitted sound beam is decomposed into ultrasonic waves in the horizontal direction and ultrasonic waves in the vertical direction. S2.3.7 When encountering defects or damage in the material, the defect location will cause the ultrasonic waves to be reflected, scattered, and attenuated. That is, the ultrasonic waves will propagate in different directions in the material around the damage and will be weakened. Using the signal receiving array around the excitation point, the attenuation degree of the ultrasonic waves at different locations is measured and compared, the scattering mode and intensity are analyzed, the location of the pipe defect is obtained, and the data is fed back to the signal receiving array.

5. The FPGA-based active-passive coordinated ultrasonic vehicle-mounted pipeline inspection and monitoring method according to claim 1, characterized in that, Step S3, data transformation and denoising processing, includes the following sub-steps: S3.

1. Using dataset X1 as input, analyze the signal through discrete wavelet transform. Expand dataset X1 under the wavelet basis to obtain wavelet coefficients. , where a is the scale parameter and b is the translation factor, used to control the scale and position of the wavelet function; S3.2, will As input, soft-threshold denoising and hard-threshold denoising methods are used to remove signal interference, resulting in denoised wavelet coefficients. ; S3.3, will As input, wavelet reconstruction is performed using the Marat algorithm to obtain the denoised dataset X2.

6. The FPGA-based active-passive coordinated ultrasonic vehicle-mounted pipeline inspection and monitoring method according to claim 5, characterized in that, In step S3.2, hard thresholding is used when the proportion of low-frequency noise in the signal is greater than that of high-frequency noise, and soft thresholding is used when the proportion of high-frequency noise is greater than that of low-frequency noise, as detailed below: The hard threshold denoising method: when the absolute value of the wavelet coefficient is less than a given threshold λ, it is set to zero; when it is greater than the threshold λ, it remains unchanged. The formula is: ； The soft thresholding denoising method works as follows: when the absolute value of the wavelet coefficients is less than a given threshold λ, set it to zero; when it is greater than the threshold λ, subtract the threshold from each coefficient. The formula is as follows: ； in, express The sign function, that is, when A value greater than 0 is 1, a value equal to 0 is 0, and a value less than 0 is -1.

7. An FPGA-based active-passive cooperative ultrasonic vehicle-mounted pipeline inspection and monitoring system, used to implement the active-passive cooperative ultrasonic vehicle-mounted pipeline inspection and monitoring method according to any one of claims 1-6, characterized in that, include: The pipeline to be inspected, the flaw detection and inspection vehicle, the signal receiving array, and the computer processor; The signal receiving array is built on the inner wall of one end of the pipeline to be inspected, and is used to receive the defect signal of the pipeline and feed back the defect location to the flaw detection and inspection trolley. The flaw detection inspection trolley includes an FPGA control module, a drive module, an excitation module, and a power supply module. The excitation module also includes a rotatable excitation sensor. Under the control of the FPGA control module, the flaw detection inspection trolley moves inside the pipeline. By rotating, it adjusts the angle of the excitation sensor, couples with the pipeline wall, and issues excitation commands. The computer processor performs wavelet transform and denoising on the data, extracts its frequency and time domain features, establishes a neural network model for training and prediction, obtains the defect depth, and outputs the final result.

8. The FPGA-based active-passive cooperative ultrasonic vehicle-mounted pipeline inspection and monitoring system according to claim 7, characterized in that, Inside the flaw detection and inspection vehicle, the FPGA control module is connected to the microcontroller via a USB interface, exchanges information via USB serial communication, and then controls the motor drive via the driver module. When the motor needs to be driven, the FPGA control module outputs an acknowledgment signal, which is sent to the microcontroller's driver module via the USB serial port. Drive the flaw detection and inspection trolley to move to the designated position inside the pipeline.

9. The FPGA-based active-passive collaborative ultrasonic vehicle-mounted pipeline inspection and monitoring system according to claim 8, characterized in that, Inside the flaw detection and inspection vehicle, the on-chip structure of the FPGA control module includes: a frequency synthesizer, an address counter, a waveform memory, and an intensity algorithm control unit. The intensity algorithm control unit is connected to the programmable amplifier on the excitation module and is used to control the movement of the vehicle and issue excitation commands.

Images