Method for automatically detecting wall thickness protection layer of electric pole

Abstract

The invention discloses a method for automatically detecting a wall thickness protection layer of an electric pole, and the method comprises the steps: adjusting the ultrasonic sound velocity and detection parameters through combining environment data, and dynamically adjusting the ultrasonic frequency and duty ratio according to the material of the electric pole; the multi-channel ultrasonic probe works synchronously to acquire the wall thickness and surface defect data of the electric pole in real time; analog signals are converted into TTL digital signals through the FPGA, and signal abnormity is detected in real time through a dynamic threshold value method; intelligently evaluating the wall thickness data by using a deep learning model, and identifying defect types according to ultrasonic echo features; an electric pole wall thickness and defect risk assessment result is displayed in real time through a display screen, and an alarm signal is triggered when abnormity is found; recording a task state, and generating a wall thickness change trend chart based on historical data; the historical trend of the change of the wall thickness of the electric pole is generated by using detection data, and potential risk points in the future are evaluated through a prediction algorithm. The method has the characteristics of high precision, high efficiency and intellectualization, and can be widely applied to the field of electric power facility detection.

Description

Technical Field

[0001] The invention relates to the technical field of pole detection, and in particular to a method for automatically detecting a protective layer of a pole wall. Background Art

[0002] Electric poles are an important part of power facilities. The integrity of their protective layer and wall thickness is directly related to the safety and stability of power transmission. However, due to long-term exposure to complex environments, electric poles may experience wall thinning or protective layer shedding due to corrosion, wear or mechanical damage, which poses a potential threat to the structural strength and service life of the electric poles. Therefore, regular inspection of the wall thickness and protective layer status of electric poles is an important measure to ensure the safe operation of the power system.

[0003] Traditional pole inspection methods mostly rely on manual inspection or single-point inspection equipment, which is not only inefficient and limited in coverage, but also prone to insufficient detection accuracy and consistency due to human factors. At the same time, the existing inspection equipment is insufficient in signal processing, data transmission and defect identification, resulting in weak intelligent analysis and real-time monitoring capabilities for the state of the pole protective layer, which cannot meet the needs of modern power systems for high-precision, full coverage and intelligent inspection.

[0004] In response to the above problems, the present invention provides a method for automatically detecting the thickness of the protective layer of the pole wall. The automated detection system combined with a multi-channel ultrasonic probe, an FPGA signal processing unit and a deep learning algorithm can realize all-round, efficient and accurate detection of poles in complex environments, significantly improving the efficiency and intelligence level of detection. Summary of the invention

[0005] In view of the above problems, the present invention provides a method for automatically detecting the thickness of the protective layer of the pole wall, so as to solve the problem that the prior art has weak intelligent analysis and real-time monitoring capabilities for the status of the pole protective layer, and cannot meet the modern power system's requirements for high-precision, full coverage and intelligent detection.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] The present invention provides a method for automatically detecting the thickness of a protective layer of a pole wall, comprising the following steps:

[0008] Step S1, installing the detection device and verifying its operating status, adjusting the ultrasonic sound velocity and detection parameters in combination with environmental data, and dynamically adjusting the ultrasonic frequency and duty cycle according to the pole material;

[0009] Wherein step S1 also includes the following sub-steps:

[0010] S1-1, install the detection device on the pole, make the sensor module fully contact with the pole surface, start the equipment self-check function, and verify the operating status of the sensor module, drive module, and data processing unit;

[0011] S1-2, dynamically correct the ultrasonic sound velocity according to the temperature and humidity parameters of the detection environment, set the number of channels, detection accuracy and alarm threshold through the main control device, and adjust the ultrasonic frequency and duty cycle according to the pole material, as shown in formula (1)-formula (2):

[0012] v c =v0(1+α·ΔT) Formula (1)

[0013]

[0014] Among them, v c is the ultrasonic sound velocity, v0 is the reference sound velocity, α is the temperature coefficient, ΔT is the temperature difference, f0 is the ultrasonic frequency, t w is the excitation pulse width.

[0015] Step S2, the sensor module moves along the axial direction of the pole under the control of the driving module and performs a 360° surround scan, while the multi-channel ultrasonic probe works synchronously to collect the pole wall thickness and surface defect data in real time;

[0016] Wherein step S2 also includes the following sub-steps:

[0017] S2-1, after the driving module is started, the sensor module is controlled to move smoothly along the axial direction of the pole through the preset path and motion parameters, and a 360° surround scan is performed at the same time. During the scanning process, the motion trajectory of the sensor module maintains a constant distance from the pole surface to ensure that the sensor can collect complete pole surface data;

[0018] S2-2, multi-channel ultrasonic probes work synchronously, using a distributed layout to collect the wall thickness and surface defect data of the pole in real time. The probes emit ultrasonic signals at a high frequency. When the signals encounter different interfaces of the poles, echoes are generated. The ultrasonic propagation time is calculated based on the collected echo signals, as shown in formula (3):

[0019]

[0020] Where d is the thickness of the pole, t is the propagation time, v c is the corrected sound velocity;

[0021] S2-3, all signals collected by the sensor module are transmitted to the data processing unit through a high-speed data interface. During the data transmission process, the sound speed parameters are dynamically corrected according to the environmental data to reduce the interference of the environment on the detection results. The original data is optimized through signal enhancement and denoising algorithms to eliminate possible noise and interference signals.

[0022] Step S3, converting the analog signal into a TTL digital signal through FPGA, calculating the wall thickness and defect depth using the time interval, and detecting signal anomalies in real time using the dynamic threshold method;

[0023] Wherein step S3 also includes the following sub-steps:

[0024] S3-1, in the data processing unit, the collected ultrasonic signal is filtered, amplified and feature extracted, a low-pass filter is used to remove high-frequency noise, and the amplitude of the effective signal is enhanced through the signal amplification circuit. The key feature points are extracted in the signal time domain, including the surface wave arrival time and the primary bottom wave arrival time, and the time difference between the two is calculated, as shown in formula (4):

[0025] Δt base =t b1 -t s Formula (4)

[0026] Among them, t s is the surface wave arrival time, t b1 is the arrival time of a bottom wave;

[0027] S3-2, using the FPGA processing unit to digitally convert the ultrasonic echo signal, converting the analog signal into a TTL digital signal. The specific conversion process is as follows:

[0028]

[0029] Among them, V is the analog signal voltage value collected by the sensor, V threshold is the voltage threshold for signal conversion, V TTL It is the converted digital signal, 1 represents high level and 0 represents low level;

[0030] S3-3, using FPGA to synchronously process the digital signal, calculate the time interval between the surface wave, the bottom wave and the defect wave, and calculate the wall thickness and defect depth according to the ultrasonic propagation time, as shown in formula (6) to formula (9):

[0031] Δt base =t b1 -t s Formula (6)

[0032] Δt defect =t defect-t s Formula (7)

[0033]

[0034] Among them, Δt base is the time interval between the surface wave and the bottom wave, Δt defect is the ultrasonic round trip time, t defect is the defect echo arrival time, Δt base is the bottom wave arrival time difference, H is the wall thickness, v c is the corrected sound velocity, A defect is the defect depth;

[0035] S3-4, combining the data of multi-channel signals, integrating the detection results of each channel through the decision-level fusion algorithm, and using the dynamic threshold method to perform real-time detection of signal anomalies, as shown in formula (10):

[0036] Δt threshold =mean(Δt)+k·std(Δt) Formula (10)

[0037] Where, mean(Δt) is the mean of the time interval, std(Δt) is the standard deviation of the time interval, and k is the adjustment parameter.

[0038] Step S4, using a deep learning model to intelligently evaluate the wall thickness data, identifying the defect type based on the ultrasonic echo characteristics, and evaluating the risk level based on the defect depth and classification results;

[0039] Wherein step S4 also includes the following sub-steps:

[0040] S4-1, using the deep learning algorithm model to take the wall thickness data as input, determine whether the wall thickness is within the safe range and evaluate the wall thickness distribution, as shown in formula (11):

[0041] P(y|X)=σ(W·X+b) Formula (11)

[0042] Among them, X is the input data, W and b are model parameters, σ is the activation function, and P(y|X) is the defect probability;

[0043] S4-2, based on the waveform characteristics of the ultrasonic echo signal, identify the defect type and evaluate the risk level, and use support vector machine (SVM) to classify the defect type, as shown in formula (12)-formula (13):

[0044]

[0045] y=argmax(softmax(Z)) Formula (13)

[0046] Among them, Adefect is the actual depth of the defect, Δt defect is the defect echo time interval, Z is the classification network output, argmax is used to determine the most likely defect type y, and y is the classification result;

[0047] S4-3, based on the data collected from multiple channels, the three-dimensional wall thickness and defect distribution model of the pole is constructed using spatial coordinates, as shown in equations (14) to (15):

[0048]

[0049] Among them, x, y are the two-dimensional position coordinates, H(x, y) is the wall thickness of the corresponding point, and Δt(x, y) is the time interval between the corresponding positions.

[0050] Step S5, presenting the pole wall thickness and defect risk assessment results in real time on a display screen, and triggering an alarm signal when an abnormality is found;

[0051] Wherein, in step S5, the following sub-steps are also included:

[0052] S5-1, after the inspection is completed, the wall thickness inspection results and defect risk assessment report of the pole are displayed in real time on the LCD screen, and the wall thickness value and defect size are calculated, as shown in formula (16):

[0053] V display (t)=H(t)+A defect (t) Formula (16)

[0054] Among them, V display (t) is the displayed value, H(t) is the real-time wall thickness value, A defect (t) is the defect size;

[0055] S5-2, when abnormal wall thickness or high-risk defects are detected, the buzzer and light alarm are automatically triggered. The specific alarm conditions are as follows:

[0056]

[0057] Among them, Alarm is the set minimum wall thickness threshold, A max is the maximum allowable defect depth;

[0058] S5-3, the detection data is transmitted to the remote monitoring center in real time through the wireless communication module for maintenance personnel to analyze and archive. The transmitted data includes: wall thickness value, defect depth and time information.

[0059] Step S6, after the detection task is completed, the task status is recorded and the data is stored in the local and cloud databases, and a wall thickness change trend chart is generated based on the historical data;

[0060] Wherein, in step S6, the following sub-steps are also included:

[0061] S6-1, after the detection task is completed, the main control system stops the movement of the sensor module and the operation of the drive module through the termination instruction to ensure that the equipment is in a safe shutdown state;

[0062] The stop operation includes: recording the final detection status parameters; stopping all data collection and processing work, and disconnecting the communication connection with the remote monitoring center; restoring the equipment to the initial standby position to prepare for the next task;

[0063] S6-2, the detection data is stored in the local storage unit of the device, and the data is uploaded to the cloud database through the wireless communication module, and the pole wall thickness change trend is generated based on the stored data, as shown in formula (18):

[0064]

[0065] Among them, H trend (t) is the wall thickness variation trend, H baseline is the base wall thickness, ΔH i is the thickness change for each test;

[0066] S6-3, after the test is completed, check the surface cleanliness of the sensor module and remove any dust and foreign matter that may be attached; lubricate and perform performance tests on the slide rails and motors of the drive module; and determine whether key components need to be replaced or repaired based on the equipment's operating time and maintenance records.

[0067] Step S7, using the detection data to generate the historical trend of pole wall thickness changes, evaluating potential risk points in the future through a prediction algorithm, and proposing targeted maintenance suggestions.

[0068] Wherein, in step S7, the following sub-steps are also included:

[0069] S7-1, based on the multiple wall thickness data stored in the detection task, generates a historical change trend chart of the pole wall thickness. The historical trend chart intuitively displays the change of the pole wall thickness in long-term operation. The trend calculation is specifically shown in formula (19):

[0070]

[0071] Among them, H baseline is the initial wall thickness at the beginning of the test, ΔH i is the wall thickness change of the i-th detection, and n is the number of detections;

[0072] S7-2, combined with trend analysis and current detection data, uses prediction algorithms to evaluate potential risk points of poles. The prediction model is based on a time series algorithm to calculate future wall thickness changes, as shown in formula (20):

[0073] H future (t+Δt)=H(t)-R decay ·Δt Formula (20)

[0074] Among them, H future (t+Δt) is the expected wall thickness at a future time point, R decay is the wall thickness reduction rate, Δt is the predicted time interval, when the predicted wall thickness is lower than the set safety threshold H threshold Maintenance recommendations will be generated, including specific plans for strengthening or replacing poles, and marking pole areas for high priority treatment;

[0075] S7-3, conducts in-depth mining of all historical test data, analyzes the wall thickness variation patterns under different materials, pole types and environmental conditions, and generates material performance optimization suggestions or test method improvement plans through multi-dimensional data association, as follows:

[0076] Material optimization: According to the corrosion rate in different environments, it is recommended to use a more durable protective layer material; the detection method is improved, adding a filtering algorithm in a noisy environment and increasing the detection frequency in areas with severe wear.

[0077] Compared with the prior art, the present invention has the following beneficial effects:

[0078] The present invention introduces a deep learning algorithm to perform intelligent analysis on the detection data, which can determine in real time whether the wall thickness of the pole is within a safe range, and accurately identify the defect type through a classification algorithm, effectively improving the accuracy and reliability of defect identification.

[0079] The present invention reduces the influence of environmental factors on the detection result by dynamically adjusting the ultrasonic sound velocity and signal processing parameters, thereby ensuring the accuracy of detection in a complex environment.

[0080] The present invention generates a three-dimensional wall thickness and defect distribution model, which helps users quickly understand the health status of the pole through intuitive graphical display. At the same time, the detection data is transmitted to the cloud storage in real time, realizing remote monitoring and historical data management, and providing a scientific basis for subsequent maintenance.

[0081] The present invention combines historical data and trend analysis, uses a time series prediction algorithm to evaluate potential risk points of poles, generates maintenance plans, and proposes suggestions for material optimization and detection method improvement, which significantly improves the scientific nature and foresight of pole maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It is understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0083] Figure 1 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0084] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 making creative work belong to the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but is only for the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0085] Please refer to Figure 1 The present invention provides a method flow chart of automatically detecting the thickness of the protective layer of the pole wall, which includes the following steps:

[0086] Step S1, installing the detection device and verifying its operating status, adjusting the ultrasonic sound velocity and detection parameters in combination with environmental data, and dynamically adjusting the ultrasonic frequency and duty cycle according to the pole material;

[0087] Wherein step S1 also includes the following sub-steps:

[0088] S1-1, install the detection device on the pole, make the sensor module fully contact with the pole surface, start the equipment self-check function, and verify the operating status of the sensor module, drive module, and data processing unit;

[0089] S1-2, dynamically correct the ultrasonic sound velocity according to the temperature and humidity parameters of the detection environment, set the number of channels, detection accuracy and alarm threshold through the main control device, and adjust the ultrasonic frequency and duty cycle according to the pole material, as shown in formula (1)-formula (2):

[0090] v c =v0(1+α·ΔT) Formula (1)

[0091]

[0092] Among them, v c is the ultrasonic sound velocity, v0 is the reference sound velocity, α is the temperature coefficient, ΔT is the temperature difference, f0 is the ultrasonic frequency, t w is the excitation pulse width.

[0093] It should be noted that the operating status includes: the sensor module confirms whether its signal acquisition capability and contact status are normal, the drive module verifies its ability to move along the axial direction of the pole and rotate 360°, and the data processing unit checks whether its signal processing, storage and transmission functions are normal.

[0094] The number of channels determines the number of ultrasonic probes involved in the detection at the same time, which directly affects the detection coverage and efficiency. According to the diameter and length of the pole, an appropriate number of probes are selected. The main control device provides options, allowing users to switch between single-channel mode and multi-channel mode as needed.

[0095] The setting of detection accuracy determines the signal sampling rate and the resolution of data processing, which affects the detail and reliability of the detection results. It is adjusted according to the pole material and the detection target: for thicker protective layers, choose lower accuracy to speed up the detection; for thin layers or vulnerable areas, increase the sampling rate to improve accuracy.

[0096] The alarm threshold is used to determine whether the wall thickness or defect depth has reached a dangerous level and trigger the corresponding alarm mechanism. The setting basis is: the wall thickness threshold is set according to the pole design specifications, and the minimum safe wall thickness value is set; the defect depth threshold is set according to the pole bearing capacity, and the maximum allowable defect depth is set.

[0097] Step S2, the sensor module moves along the axial direction of the pole under the control of the driving module and performs a 360° surround scan, while the multi-channel ultrasonic probe works synchronously to collect the pole wall thickness and surface defect data in real time;

[0098] Wherein step S2 also includes the following sub-steps:

[0099] S2-1, after the driving module is started, the sensor module is controlled to move smoothly along the axial direction of the pole through the preset path and motion parameters, and a 360° surround scan is performed at the same time. During the scanning process, the motion trajectory of the sensor module maintains a constant distance from the pole surface to ensure that the sensor can collect complete pole surface data;

[0100] S2-2, multi-channel ultrasonic probes work synchronously, using a distributed layout to collect the wall thickness and surface defect data of the pole in real time. The probes emit ultrasonic signals at a high frequency. When the signals encounter different interfaces of the poles, echoes are generated. The ultrasonic propagation time is calculated based on the collected echo signals, as shown in formula (3):

[0101]

[0102] Where d is the thickness of the pole, t is the propagation time, v c is the corrected sound velocity;

[0103] S2-3, all signals collected by the sensor module are transmitted to the data processing unit through a high-speed data interface. During the data transmission process, the sound speed parameters are dynamically corrected according to the environmental data to reduce the interference of the environment on the detection results. The original data is optimized through signal enhancement and denoising algorithms to eliminate possible noise and interference signals.

[0104] It should be noted that a multi-channel probe refers to multiple ultrasonic probes arranged on the equipment. Each probe is responsible for a different detection area. The probes surround the pole in an equidistant manner, covering the circumferential and axial areas of the pole. The position and spacing of the probes are dynamically adjusted according to the diameter and material properties of the pole.

[0105] The probes of each channel are triggered periodically and synchronously through the control system to emit ultrasonic signals. The excitation signals for multiple probes are realized through the PWM signal generator to avoid interference between the signals of each channel. The echo signals received by each probe are collected independently, filtered and feature extracted by the data processing unit, and multi-channel data are simultaneously transmitted to the FPGA for real-time processing to ensure that the time difference error between the probes is minimized.

[0106] The drive module housing is made of IP68 grade protection and is equipped with a position sensor to monitor its position on the pole in real time to ensure the accuracy of the detection path. It is equipped with high-precision guide rails and ball sliders to ensure the stability of the drive module when moving on the pole surface. It can automatically adjust the movement speed to improve efficiency according to the complexity of the detection area.

[0107] Step S3, converting the analog signal into a TTL digital signal through FPGA, calculating the wall thickness and defect depth using the time interval, and detecting signal anomalies in real time using the dynamic threshold method;

[0108] Wherein step S3 also includes the following sub-steps:

[0109] S3-1, in the data processing unit, the collected ultrasonic signal is filtered, amplified and feature extracted, a low-pass filter is used to remove high-frequency noise, and the amplitude of the effective signal is enhanced through the signal amplification circuit. The key feature points are extracted in the signal time domain, including the surface wave arrival time and the primary bottom wave arrival time, and the time difference between the two is calculated, as shown in formula (4):

[0110] Δt base =t b1 -t s Formula (4)

[0111] Among them, t s is the surface wave arrival time, t b1 is the arrival time of a bottom wave;

[0112] S3-2, using the FPGA processing unit to digitally convert the ultrasonic echo signal, converting the analog signal into a TTL digital signal. The specific conversion process is as follows:

[0113]

[0114] Among them, V is the analog signal voltage value collected by the sensor, V threshold is the voltage threshold for signal conversion, V TTL It is the converted digital signal, 1 represents high level and 0 represents low level;

[0115] S3-3, using FPGA to synchronously process the digital signal, calculate the time interval between the surface wave, the bottom wave and the defect wave, and calculate the wall thickness and defect depth according to the ultrasonic propagation time, as shown in formula (6) to formula (9):

[0116] Δt base =t b1 -t s Formula (6)

[0117] Δt defect =t defect -t s Formula (7)

[0118]

[0119] Among them, Δt base is the time interval between the surface wave and the bottom wave, Δt defect is the ultrasonic round trip time, t defect is the defect echo arrival time, Δt base is the bottom wave arrival time difference, H is the wall thickness, v c is the corrected sound velocity, A defect is the defect depth;

[0120] S3-4, combining the data of multi-channel signals, integrating the detection results of each channel through the decision-level fusion algorithm, and using the dynamic threshold method to perform real-time detection of signal anomalies, as shown in formula (10):

[0121] Δt threshold =mean(Δt)+k·std(Δt) Formula (10)

[0122] Where, mean(Δt) is the mean of the time interval, std(Δt) is the standard deviation of the time interval, and k is the adjustment parameter.

[0123] It should be noted that FPGA is a data processing unit, which is mainly used to receive and process echo signals transmitted by ultrasonic probes to achieve high-speed, high-precision real-time signal processing. Its functions include: converting analog signals into digital signals, filtering, feature extraction and time difference calculation, multi-channel signal synchronization processing and fusion, and logical control of data storage and transmission.

[0124] FPGA collects and processes signals from multiple channels simultaneously, ensuring that the time synchronization error between channels is less than 1 microsecond. It uses data fusion algorithms to integrate multi-channel results to improve detection accuracy, and stores the processed data directly in the device's cache, which is then transmitted to the master device or the cloud via a Gigabit Ethernet interface or wireless module.

[0125] The data processing unit is usually integrated in the main control device. As the core part of the entire detection system, it is responsible for receiving signals from the sensor module and multi-channel probe, performing preliminary data processing, connecting the high-speed data interface and the sensor module, maintaining communication with the drive module, and coordinating the sensor's detection tasks and position status.

[0126] The data processing unit receives the original analog signal transmitted from the sensor module, performs preliminary calibration on the signal according to environmental parameters, removes high-frequency noise through a low-pass filter, and enhances the signal amplitude using an amplifier circuit to ensure that key signal points are highlighted.

[0127] The preprocessed signal is transmitted to the FPGA processing unit through a high-speed data interface for further processing. The FPGA processing unit is usually integrated on the hardware motherboard, close to the data processing unit, to ensure efficient data transmission and real-time processing capabilities. It is connected to the data processing unit through a high-speed interface and directly interacts with the display module, alarm system, and storage module.

[0128] Step S4, using a deep learning model to intelligently evaluate the wall thickness data, identifying the defect type based on the ultrasonic echo characteristics, and evaluating the risk level based on the defect depth and classification results;

[0129] Wherein step S4 also includes the following sub-steps:

[0130] S4-1, using the deep learning algorithm model to take the wall thickness data as input, determine whether the wall thickness is within the safe range and evaluate the wall thickness distribution, as shown in formula (11):

[0131] P(y|X)=σ(W·X+b) Formula (11)

[0132] Among them, X is the input data, W and b are model parameters, σ is the activation function, and P(y|X) is the defect probability;

[0133] S4-2, based on the waveform characteristics of the ultrasonic echo signal, identify the defect type and evaluate the risk level, and use support vector machine (SVM) to classify the defect type, as shown in formula (12)-formula (13):

[0134]

[0135] y=argmax(softmax(Z)) Formula (13)

[0136] Among them, A defect is the actual depth of the defect, Δt defect is the defect echo time interval, Z is the classification network output, argmax is used to determine the most likely defect type y, and y is the classification result;

[0137] S4-3, based on the data collected from multiple channels, the three-dimensional wall thickness and defect distribution model of the pole is constructed using spatial coordinates, as shown in equations (14) to (15):

[0138]

[0139] Among them, x, y are the two-dimensional position coordinates, H(x, y) is the wall thickness of the corresponding point, and Δt(x, y) is the time interval between the corresponding positions.

[0140] It should be noted that defect type identification extracts feature values ​​for defect classification by analyzing the waveform characteristics of ultrasonic echoes, including amplitude, frequency, duration, and energy. The support vector machine model classifies defects based on high-dimensional feature space. The specific defect types are as follows:

[0141] Surface defects usually appear as high-frequency short-time signals, such as cracks and spalling; internal defects appear as low-frequency long-time signals, such as bubbles and voids; boundary anomalies appear as sudden changes in signal amplitude, such as delamination and edge damage.

[0142] Using 3D graphics processing tools, the data is converted into an interactive distribution model, and the defect distribution is superimposed on the 3D wall thickness model with color coding:

[0143] Normal areas are green, moderately defective areas are yellow, and severely defective areas are red.

[0144] Step S5, presenting the pole wall thickness and defect risk assessment results in real time on a display screen, and triggering an alarm signal when an abnormality is found;

[0145] Wherein, in step S5, the following sub-steps are also included:

[0146] S5-1, after the inspection is completed, the wall thickness inspection results and defect risk assessment report of the pole are displayed in real time on the LCD screen, and the wall thickness value and defect size are calculated, as shown in formula (16):

[0147] V display (t)=H(t)+A defect (t) Formula (16)

[0148] Among them, V display (t) is the displayed value, H(t) is the real-time wall thickness value, A defect (t) is the defect size;

[0149] S5-2, when abnormal wall thickness or high-risk defects are detected, the buzzer and light alarm are automatically triggered. The specific alarm conditions are as follows:

[0150]

[0151] Among them, Alarm is the set minimum wall thickness threshold, A max is the maximum allowable defect depth;

[0152] S5-3, the detection data is transmitted to the remote monitoring center in real time through the wireless communication module for maintenance personnel to analyze and archive. The transmitted data includes: wall thickness value, defect depth and time information.

[0153] It should be noted that the buzzer alarm emits a high-frequency sound to remind the operator to pay attention to the test results immediately. The alarm modes include: continuous sound indicates the existence of multiple high-risk areas, intermittent sound indicates a single high-risk area or medium risk, and the volume adjustment can adjust the alarm volume according to the ambient noise of the test site.

[0154] The light alarm provides intuitive alarm information through color changes and flashing modes. The color codes include: green for normal status without abnormalities; yellow for medium risk requiring attention; red for high risk requiring immediate attention; the flashing mode means that the flashing frequency increases with the risk level, with red light having the highest frequency.

[0155] The display screen is linked. When the alarm is triggered, the display screen automatically locates the abnormal point, highlights the specific value of the wall thickness or defect depth, and generates a real-time alarm log.

[0156] Grading alarm, providing graded alarm according to risk level:

[0157] Level 1 alarm (low risk): only the display screen is marked, no sound or light alarm;

[0158] Level 2 alarm (medium risk): the light flashes yellow and the display screen is highlighted;

[0159] Level 3 alarm (high risk): flashing red light + buzzer alarm.

[0160] Step S6, after the detection task is completed, the task status is recorded and the data is stored in the local and cloud databases, and a wall thickness change trend chart is generated based on the historical data;

[0161] Wherein, in step S6, the following sub-steps are also included:

[0162] S6-1, after the detection task is completed, the main control system stops the movement of the sensor module and the operation of the drive module through the termination instruction to ensure that the equipment is in a safe shutdown state;

[0163] The stop operation includes: recording the final detection status parameters; stopping all data collection and processing work, and disconnecting the communication connection with the remote monitoring center; restoring the equipment to the initial standby position to prepare for the next task;

[0164] S6-2, the detection data is stored in the local storage unit of the device, and the data is uploaded to the cloud database through the wireless communication module, and the pole wall thickness change trend is generated based on the stored data, as shown in formula (18):

[0165]

[0166] Among them, H trend (t) is the wall thickness variation trend, H baseline is the base wall thickness, ΔH i is the thickness change for each test;

[0167] S6-3, after the test is completed, check the surface cleanliness of the sensor module and remove any dust and foreign matter that may be attached; lubricate and perform performance tests on the slide rails and motors of the drive module; and determine whether key components need to be replaced or repaired based on the equipment's operating time and maintenance records.

[0168] It should be noted that the trend analysis of pole wall thickness changes reveals the change pattern of wall thickness over time by comparing multiple test data. It is used for: monitoring of corrosion or wear: identifying the long-term degradation of pole protective layers and materials; risk warning: predicting possible future problems of insufficient wall thickness or defect expansion; maintenance optimization: providing a basis for formulating scientific maintenance plans and material improvements.

[0169] Data visualization includes trend chart generation, heat map and three-dimensional model. Trend chart generation uses line chart or bar chart to intuitively display the change of wall thickness. The x-axis represents time (detection cycle), and the y-axis represents wall thickness or change rate. The key point annotation is set as the time node when the threshold is first reached.

[0170] The thermal map and three-dimensional model use the thermal map combined with trend analysis to intuitively display the area and severity of wall thickness changes on the pole surface, add the time dimension to the three-dimensional wall thickness model, and dynamically display the wall thickness degradation.

[0171] The local storage unit is integrated in the main control device and is located inside the hardware platform of the main control device. It is directly connected to the data processing unit and the FPGA processing unit, and is connected to the data processing unit and the FPGA unit through a high-speed bus SPI interface to ensure fast data writing and reading. It is equipped with a non-volatile memory (such as NAND flash memory or EEPROM) to ensure that data will not be lost in the event of a power outage. Key data is backed up through a RAID configuration or a dual partition structure to further improve storage security. It supports regular cleaning of expired data to release storage space, or allows manual export or deletion of historical data through the user interface.

[0172] Step S7, using the detection data to generate the historical trend of pole wall thickness changes, evaluating potential risk points in the future through a prediction algorithm, and proposing targeted maintenance suggestions.

[0173] Wherein, in step S7, the following sub-steps are also included:

[0174] S7-1, based on the multiple wall thickness data stored in the detection task, generates a historical change trend chart of the pole wall thickness. The historical trend chart intuitively displays the change of the pole wall thickness in long-term operation. The trend calculation is specifically shown in formula (19):

[0175]

[0176] Among them, H baseline is the initial wall thickness at the beginning of the test, ΔH i is the wall thickness change of the i-th detection, and n is the number of detections;

[0177] S7-2, combined with trend analysis and current detection data, uses prediction algorithms to evaluate potential risk points of poles. The prediction model is based on a time series algorithm to calculate future wall thickness changes, as shown in formula (20):

[0178] H future (t+Δt)=H(t)-R decay ·Δt Formula (20)

[0179] Among them, H future (t+Δt) is the expected wall thickness at a future time point, R decay is the wall thickness reduction rate, Δt is the predicted time interval, when the predicted wall thickness is lower than the set safety threshold H threshold Maintenance recommendations will be generated, including specific plans for strengthening or replacing poles, and marking pole areas for high priority treatment;

[0180] S7-3, conducts in-depth mining of all historical test data, analyzes the wall thickness variation patterns under different materials, pole types and environmental conditions, and generates material performance optimization suggestions or test method improvement plans through multi-dimensional data association, as follows:

[0181] Material optimization: According to the corrosion rate in different environments, it is recommended to use a more durable protective layer material; the detection method is improved, adding a filtering algorithm in a noisy environment and increasing the detection frequency in areas with severe wear.

[0182] It should be noted that the sources of multidimensional data include:

[0183] Wall thickness change data, wall thickness trend and change rate in different time periods;

[0184] Defect characteristic data, defect type, depth, distribution and expansion law;

[0185] Environmental condition data, including temperature, humidity, corrosive gas concentration and other environmental factors;

[0186] Material property data, composition, density and durability characteristics of pole protective layer materials.

[0187] Suggestions for optimizing material properties: The optimization direction is to improve corrosion resistance. For highly corrosive environments, it is recommended to use protective layer materials with higher corrosion resistance, such as acid- and alkali-resistant composite materials; to increase mechanical strength, for crack-prone areas, it is recommended to use high-strength fiber-reinforced materials to improve crack resistance; to reduce the effects of thermal expansion and contraction, in areas with large temperature differences, it is recommended to use materials with higher thermal stability.

[0188] It should be noted that the sensor module is fixed on the surface of the pole through a mounting device. For poles of different diameters, a retractable fixing bracket is used and equipped with an elastic clamp to ensure that the sensor module is in close contact with the pole surface. A universal adjustment base is added at the connection between the bracket and the sensor module to allow the sensor to fine-tune the angle after installation. To avoid signal interference or data distortion caused by improper installation, ensure that the contact surface of the sensor module is flat during installation, and use an adjustable clamp or soft gasket to adapt to the curvature of the pole surface to prevent loose installation.

[0189] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention has various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for automatically detecting the thickness of a protective layer of a pole wall, characterized in that: The following steps are involved: Step S1, installing the detection device and verifying its operating status, adjusting the ultrasonic sound velocity and detection parameters in combination with environmental data, and dynamically adjusting the ultrasonic frequency and duty cycle according to the pole material; Step S2, the sensor module moves along the axial direction of the pole under the control of the driving module and performs a 360° surround scan, while the multi-channel ultrasonic probe works synchronously to collect the pole wall thickness and surface defect data in real time; Step S3, converting the analog signal into a TTL digital signal through FPGA, calculating the wall thickness and defect depth using the time interval, and detecting signal anomalies in real time using the dynamic threshold method; Step S4, using a deep learning model to intelligently evaluate the wall thickness data, identifying the defect type based on the ultrasonic echo characteristics, and evaluating the risk level based on the defect depth and classification results; Step S5, presenting the pole wall thickness and defect risk assessment results in real time on a display screen, and triggering an alarm signal when an abnormality is found; Step S6, after the detection task is completed, the task status is recorded and the data is stored in the local and cloud databases, and a wall thickness change trend chart is generated based on the historical data; Step S7, using the detection data to generate the historical trend of pole wall thickness changes, evaluating potential risk points in the future through a prediction algorithm, and proposing targeted maintenance suggestions.

2. The method for automatically detecting the thickness of the protective layer of the pole wall according to claim 1 is characterized in that: Wherein step S1 also includes the following sub-steps: S1-1, install the detection device on the pole, make the sensor module fully contact with the pole surface, start the equipment self-check function, and verify the operating status of the sensor module, drive module, and data processing unit; S1-2, dynamically correct the ultrasonic sound velocity according to the temperature and humidity parameters of the detection environment, set the number of channels, detection accuracy and alarm threshold through the main control device, and adjust the ultrasonic frequency and duty cycle according to the pole material, as shown in formula (1)-formula (2): v c =v0(1+α·ΔT) Equation (1) Among them, v c is the ultrasonic sound velocity, v0 is the reference sound velocity, α is the temperature coefficient, ΔT is the temperature difference, f0 is the ultrasonic frequency, t w is the excitation pulse width.

3. The method for automatically detecting the thickness of the protective layer of the pole wall according to claim 1 is characterized in that: Wherein step S2 also includes the following sub-steps: S2-1, after the driving module is started, the sensor module is controlled to move smoothly along the axial direction of the pole through the preset path and motion parameters, and a 360° surround scan is performed at the same time. During the scanning process, the motion trajectory of the sensor module maintains a constant distance from the pole surface to ensure that the sensor can collect complete pole surface data; S2-2, multi-channel ultrasonic probes work synchronously, using a distributed layout to collect the wall thickness and surface defect data of the pole in real time. The probes emit ultrasonic signals at a high frequency. When the signals encounter different interfaces of the poles, echoes are generated. The ultrasonic propagation time is calculated based on the collected echo signals, as shown in formula (3): Where d is the thickness of the pole, t is the propagation time, v c is the corrected sound velocity; S2-3, all signals collected by the sensor module are transmitted to the data processing unit through a high-speed data interface. During the data transmission process, the sound speed parameters are dynamically corrected according to the environmental data to reduce the interference of the environment on the detection results. The original data is optimized through signal enhancement and denoising algorithms to eliminate possible noise and interference signals.

4. The method for automatically detecting the thickness of the protective layer of the pole wall according to claim 1 is characterized in that: Wherein step S3 also includes the following sub-steps: S3-1, in the data processing unit, the collected ultrasonic signal is filtered, amplified and feature extracted, a low-pass filter is used to remove high-frequency noise, and the amplitude of the effective signal is enhanced through the signal amplification circuit. The key feature points are extracted in the signal time domain, including the surface wave arrival time and the primary bottom wave arrival time, and the time difference between the two is calculated, as shown in formula (4): Δt base =t b1 -t s Formula (4) Among them, t s is the surface wave arrival time, t b1 is the arrival time of a bottom wave; S3-2, using the FPGA processing unit to digitally convert the ultrasonic echo signal, converting the analog signal into a TTL digital signal. The specific conversion process is as follows: Among them, V is the analog signal voltage value collected by the sensor, V threshold is the voltage threshold for signal conversion, V TTL It is the converted digital signal, 1 represents high level and 0 represents low level; S3-3, using FPGA to synchronously process the digital signal, calculate the time interval between the surface wave and the bottom wave and the defect wave, and calculate the wall thickness and defect depth according to the ultrasonic propagation time, as shown in formula (6) to formula (9): Δt base =t b1 -t s Formula (6) Δt defect =t defect -t s Formula (7) Among them, Δt base is the time interval between the surface wave and the bottom wave, Δt defect is the ultrasonic round trip time, t defect is the defect echo arrival time, Δt base is the bottom wave arrival time difference, H is the wall thickness, v c is the corrected sound velocity, A defect is the defect depth; S3-4, combining the data of multi-channel signals, integrating the detection results of each channel through the decision-level fusion algorithm, and using the dynamic threshold method to perform real-time detection of signal anomalies, as shown in formula (10): Δt threshold = mean(Δt) + k·std(Δt), Equation (10) Where, mean(Δt) is the mean of the time interval, std(Δt) is the standard deviation of the time interval, and k is the adjustment parameter.

5. The method for automatically detecting the thickness of the protective layer of the pole wall according to claim 1 is characterized in that: Wherein, in step S4, the following sub-steps are also included: S4-1, using the deep learning algorithm model to take the wall thickness data as input, determine whether the wall thickness is within the safe range and evaluate the wall thickness distribution, as shown in formula (11): P(y|X)=σ(W·X+b) Formula (11) Where X is the input data, W and b are model parameters, σ is the activation function, and P(y|X) is the defect probability; S4-2, based on the waveform characteristics of the ultrasonic echo signal, identify the defect type and evaluate the risk level, and use support vector machine (SVM) to classify the defect type, as shown in formula (12)-formula (13): y=argmax(softmax(Z)) Formula (13) Among them, A defect is the actual depth of the defect, Δt defect is the defect echo time interval, Z is the classification network output, argmax is used to determine the most likely defect type y, and y is the classification result; S4-3, based on the data collected from multiple channels, the three-dimensional wall thickness and defect distribution model of the pole is constructed using spatial coordinates, as shown in equations (14) to (15): Among them, x, y are the two-dimensional position coordinates, H(x, y) is the wall thickness of the corresponding point, and Δt(x, y) is the time interval between the corresponding positions.

6. The method for automatically detecting the thickness of the protective layer of the pole wall according to claim 1 is characterized by: Wherein, in step S5, the following sub-steps are also included: S5-1, after the inspection is completed, the wall thickness inspection results and defect risk assessment report of the pole are displayed in real time on the LCD screen, and the wall thickness value and defect size are calculated, as shown in formula (16): V display V(t) = H(t) + A defect V(t) Equation (16) Among them, V display (t) is the displayed value, H(t) is the real-time wall thickness value, A defect (t) is the defect size; S5-2, when abnormal wall thickness or high-risk defects are detected, the buzzer and light alarm are automatically triggered. The specific alarm conditions are as shown in formula (17): Among them, Alarm is the set minimum wall thickness threshold, A max is the maximum allowable defect depth; S5-3, the detection data is transmitted to the remote monitoring center in real time through the wireless communication module for maintenance personnel to analyze and archive. The transmitted data includes: wall thickness value, defect depth and time information.

7. The method for automatically detecting the thickness of the protective layer of the pole wall according to claim 1 is characterized by: Wherein, in step S6, the following sub-steps are also included: S6-1, after the detection task is completed, the main control system stops the movement of the sensor module and the operation of the drive module through the termination instruction to ensure that the equipment is in a safe shutdown state; The stop operation includes: recording the final detection status parameters; stopping all data collection and processing work, and disconnecting the communication connection with the remote monitoring center; restoring the equipment to the initial standby position to prepare for the next task; S6-2, the detection data is stored in the local storage unit of the device, and the data is uploaded to the cloud database through the wireless communication module, and the pole wall thickness change trend is generated based on the stored data, as shown in formula (18): Among them, H trend (t) is the wall thickness variation trend, H baseline is the base wall thickness, ΔH i is the thickness change for each test; S6-3, after the test, check the surface cleanliness of the sensor module and remove any dust and foreign matter that may be attached; lubricate and perform performance tests on the slide rails and motors of the drive module; and determine whether key components need to be replaced or repaired based on the equipment's operating time and maintenance records.

8. The method for automatically detecting the thickness of the protective layer of the pole wall according to claim 1 is characterized by: Wherein, in step S7, the following sub-steps are also included: S7-1, based on the multiple wall thickness data stored in the detection task, generates a historical change trend chart of the pole wall thickness. The historical trend chart intuitively displays the change of the pole wall thickness in long-term operation. The trend calculation is specifically shown in formula (19): Among them, H baseline is the initial wall thickness at the beginning of the test, ΔH i is the wall thickness change of the i-th detection, and n is the number of detections; S7-2, combined with trend analysis and current detection data, uses prediction algorithms to evaluate potential risk points of poles. The prediction model is based on a time series algorithm to calculate future wall thickness changes, as shown in formula (20): H future (t + Δt) = H(t) - R decay ·Δt Equation (20) Among them, H future (t+Δt) is the expected wall thickness at a future time point, R decay is the wall thickness reduction rate, Δt is the predicted time interval, when the predicted wall thickness is lower than the set safety threshold H threshold Maintenance recommendations will be generated, including specific plans for strengthening or replacing poles, and marking pole areas for high priority treatment; S7-3, conducts in-depth mining of all historical test data, analyzes the wall thickness variation patterns under different materials, pole types and environmental conditions, and generates material performance optimization suggestions or test method improvement plans through multi-dimensional data association, as follows: Material optimization: According to the corrosion rate in different environments, it is recommended to use a more durable protective layer material; the detection method is improved, adding a filtering algorithm in a noisy environment and increasing the detection frequency in areas with severe wear.

Images