Acoustic imaging positioning method for intelligent monitoring of electric power engineering regional faults

Abstract

The invention discloses an acoustic imaging positioning method for intelligent monitoring of electric power engineering regional faults, which relates to the technical field of fault monitoring, and comprises the following steps: arranging a microphone array to collect sound wave signals, and monitoring and recording sound data in real time; determining the spatial position of a sound source by using an ultrasonic sound source positioning technology; sound source distribution generated by the equipment fault is displayed; displaying the relationship between the three-dimensional position of the equipment fault point and the surrounding environment; fault sound sources are classified, and types and reasons are determined; a diagnosis result is fed back in real time, a fault position is displayed, and inspection and repair suggestions are provided; and detecting the health condition of the equipment in combination with a machine learning model, predicting potential faults and giving out early warning. According to the invention, a high-precision microphone array and an ultrasonic sound source positioning technology are utilized, the distribution state of a sound source in a space is displayed in real time through an imaging mode of superposing an acoustic image and a visible light image, a partial discharge point, a gas leakage point and an equipment abnormal sound site are rapidly determined, and the inspection efficiency and the maintenance speed are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power fault detection, and in particular to an acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering. Background Art

[0002] In modern power engineering, the stability and safety of equipment are of vital importance, and the timely detection and location of equipment failures are the key to ensuring the reliable operation of the power system.

[0003] Traditional methods of fault detection of power equipment mostly rely on manual inspection and traditional detection equipment. Manual inspection regularly checks the operating status of power equipment and relies on the experience and judgment of staff to identify signs of equipment failure. However, manual inspection has obvious limitations.

[0004] First, manual inspection relies on the experience of inspectors and their keen perception of equipment status, and is easily affected by subjective factors, resulting in missed detection and false detection. Second, manual inspection is usually carried out after a device failure occurs, and real-time and continuous fault monitoring cannot be achieved. Therefore, the discovery of faults is often delayed, resulting in serious equipment failures, and even shutdown and equipment damage, increasing maintenance costs and downtime. To this end, we propose an acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering projects. Summary of the invention

[0005] The purpose of the present invention is to provide an acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering projects to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: an acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering, comprising the following steps: Step 1: deploy a microphone array to collect sound wave signals in the power equipment area, and monitor and record sound data in real time; Step 2: Using ultrasonic sound source localization technology, the sound wave signal is subjected to time delay estimation and multi-channel signal processing to determine the spatial position of the sound source; Step 3: Construct an acoustic image according to the spatial position of the sound source to display the distribution of the sound source caused by the equipment failure; Step 4: Superimpose the acoustic image with the visible light image of the equipment's surrounding environment to obtain a comprehensive imaging map, accurately showing the relationship between the three-dimensional position of the equipment's fault point and the surrounding environment; Step 5: Classify different types of fault sound sources, and further determine the type and cause of the fault based on the frequency and intensity characteristics of the sound source; Step 6: Real-time feedback of fault diagnosis results, displaying the specific location of the fault point, and providing targeted inspection and repair suggestions; Step 7: Analyze sound source data with machine learning models to automatically detect equipment health, predict potential failures, and issue warnings.

[0007] Preferably, the specific steps of step 1 are: Step 1.1, deploy multiple microphone arrays in the equipment area to collect sound wave signals in the power equipment; Step 1.2: Perform preliminary filtering on the collected sound wave signals to remove background noise and environmental interference.

[0008] Preferably, the specific steps of step 2 are: Step 2.1, according to the delay difference of the multi-channel microphone array, the spatial position of the sound source is calculated by ultrasonic sound source localization technology, and the accurate sound source position is obtained by using the delay estimation algorithm; Step 2.2, using the sound wave propagation model, optimize the signal delay, and then use the least squares method to calculate the delay; Step 2.3: Estimate the sound source position according to the time delay difference calculation formula. The specific formula is: ; Among them, x s is the estimated sound source position, d i is the distance between the ith microphone and the sound source, d i (m) is the distance calculated according to the propagation model, and N is the number of microphones.

[0009] Preferably, the specific steps of step 3 are: Step 3.1, according to the spatial position of the sound source, generate an acoustic image to show the distribution of the sound source in the power equipment area, and use a three-dimensional image processing algorithm to draw a three-dimensional distribution map of the sound source; Step 3.2: Use an interpolation algorithm to smooth the sound source intensity to obtain a uniformly distributed sound source intensity map, and determine the specific position and intensity of the sound source in space.

[0010] Preferably, the specific steps of step 4 include: Step 4.1, obtain image data of the device area in real time through a visible light camera, and record the environmental image around the device in real time; Step 4.2: Process the visible light image to align it with the sound source position in the sonogram, and superimpose the two through image fusion technology to generate a comprehensive imaging image.

[0011] Preferably, the specific steps of step 5 include: Step 5.1, extract the frequency characteristics of the sound source through spectrum analysis, classify it according to the intensity change of the fault sound source, and determine the fault type; Step 5.2: Input the frequency and intensity data into the classification model to predict the fault type through the machine learning algorithm. The fault type includes but is not limited to partial discharge, gas leakage and abnormal noise of equipment.

[0012] Preferably, the training step of the classification model includes: Collect a large amount of equipment failure data to extract the frequency, intensity and time characteristics of the sound source as training data; The classification model is trained using supervised learning algorithms to automatically classify different types of faults.

[0013] Preferably, the specific steps of step 6 include: Step 6.1, based on the type and location of the sound source, a fault diagnosis report is generated in real time, the diagnosis report including the fault type, location and recommended maintenance plan; Step 6.2: Push the diagnostic report to the smart device of the inspector via the wireless communication network.

[0014] Preferably, in step 2, a Kalman filter algorithm is also used to determine the location of the sound source, and the specific formula is: ; Among them, x k│k is the state estimate at the current moment, K k is the Kalman gain, z k is the observed value, H k Peripheral observation matrix.

[0015] The present invention also provides an acoustic imaging positioning system for intelligent monitoring of regional faults in electric power engineering projects, and implements the above-mentioned acoustic imaging positioning method for intelligent monitoring of regional faults in electric power engineering projects, including: The sound source acquisition module is used to deploy a microphone array to collect sound wave signals in the power equipment area, and monitor and record sound data in real time; The sound source localization module is used to use ultrasonic sound source localization technology to perform time delay estimation and multi-channel signal processing on sound wave signals to determine the spatial position of the sound source; The sound image generation module is used to construct a sound image according to the spatial position of the sound source to show the distribution of the sound source caused by the equipment failure; Image overlay module, used to overlay the acoustic image with the visible light image of the equipment's surrounding environment to obtain a comprehensive image that accurately displays the relationship between the three-dimensional position of the equipment's fault point and the surrounding environment; Fault classification and diagnosis module, which is used to provide real-time feedback on fault diagnosis results, display the specific location of the fault point, and provide targeted inspection and repair suggestions; The early warning module is used to analyze sound source data in combination with machine learning models, automatically detect the health of equipment, predict potential failures and issue early warnings.

[0016] Technical effects and advantages of the present invention: (1) The present invention uses a high-precision microphone array and ultrasonic sound source positioning technology to display the distribution of sound sources in space in real time through the imaging method of superimposing sound images and visible light images, quickly determine local discharge points, gas leakage points, and abnormal sound locations of equipment, and improve inspection efficiency and maintenance speed; (2) The present invention utilizes time delay estimation and multi-channel signal processing to efficiently extract spatial position information from data collected by multiple sensors. The time delay difference positioning technology enables the three-dimensional spatial position of the sound source to be accurately determined, thereby helping to quickly locate the fault point of the power equipment and reduce the fault response time. The use of the time delay difference of the multi-channel signal for calculation not only enhances the positioning accuracy of the system, but also reduces errors through multi-point data comparison, thereby improving the overall reliability and accuracy of the system. (3) The present invention uses the Kalman filter algorithm to update and correct the system in real time. In sound source positioning, the Kalman filter can continuously optimize the location estimation of the sound source based on historical data and new observation results; the Kalman filter can steadily improve the positioning accuracy under the influence of signal noise and environmental interference. By estimating the state at multiple moments, it can effectively reduce the impact of accidental errors and improve the stability of sound source positioning; in a complex power equipment environment, dynamic changes and noise source interference may occur. The Kalman filter can quickly adapt to these changes, filter out irrelevant signals, and ensure accurate tracking of the location of the sound source through the Kalman gain K. k The filter can dynamically adjust the prediction accuracy according to the current signal quality, so that the system can adaptively optimize the positioning accuracy under different working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The figure is a schematic diagram of the flow structure of the positioning method of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] The present invention provides Figure 1 The acoustic imaging positioning method for intelligent monitoring of regional faults in a power engineering project shown includes the following steps: Step 1: deploy a microphone array to collect sound wave signals in the power equipment area, and monitor and record sound data in real time; In this embodiment, the specific steps of step 1 are: Step 1.1, deploy multiple microphone arrays in the equipment area to collect sound wave signals in the power equipment; Step 1.2: Perform preliminary filtering on the collected sound wave signals to remove background noise and environmental interference.

[0020] It is necessary to deploy multiple microphone arrays in the power equipment area to ensure that all areas of the equipment are covered, and all sound wave signals during the operation of the equipment are monitored and recorded in real time. This method can effectively avoid monitoring blind spots, provide comprehensive acoustic data on the working status of the equipment, and ensure that there are sufficient sound collection points for each important location in the power equipment area, which allows signals to be captured from different angles and positions, providing higher spatial resolution and more accurate signal source information for subsequent positioning and analysis; the deployment of microphone arrays can receive sound wave signals at multiple points simultaneously, providing high-precision data input for subsequent positioning and analysis. These data include sound characteristics at different time and spatial positions, such as frequency, intensity and signal arrival time; the deployed microphone arrays can record sound data in real time, which is crucial for rapid response to power equipment failures. By collecting sound data in real time, abnormal sound signals can be quickly discovered and analyzed, thereby improving fault diagnosis and response speed; the collected sound wave signals can be The sound waves generated by the equipment may contain a lot of background noise and environmental interference (such as mechanical noise and wind noise during normal operation of the equipment). Through preliminary filtering processing, these irrelevant signals can be removed, leaving effective information related to the equipment failure. This can improve signal quality and reduce errors in subsequent processing. The preliminary filtering process reduces the noise level, which helps to more accurately analyze signal characteristics (such as frequency, intensity changes, etc.), and provide clean signal source data for subsequent fault type classification, positioning and other analysis. After removing noise and unnecessary interference signals, the amount of data and processing complexity can be reduced, and the efficiency of subsequent data analysis and sound source positioning can be improved. Through the above steps, the overall effect is to monitor the sound wave signals of the equipment in real time in the power equipment area, and at the same time ensure high-precision data collection through the deployment of multiple microphone arrays, and remove irrelevant interference signals through filtering processing, so as to provide high-quality input data for subsequent sound source positioning, fault diagnosis and data analysis, and ultimately improve the fault response and repair efficiency of the entire intelligent monitoring system.

[0021] Step 2: Using ultrasonic sound source localization technology, the sound wave signal is subjected to time delay estimation and multi-channel signal processing to determine the spatial position of the sound source; Through time delay estimation and multi-channel signal processing, spatial position information can be efficiently extracted from data collected by multiple sensors. The time delay difference positioning technology enables the three-dimensional spatial position of the sound source to be accurately determined, thereby helping to quickly locate the fault point of the power equipment and reduce the fault response time. Using the time delay difference of multi-channel signals for calculation not only enhances the positioning accuracy of the system, but also reduces errors through multi-point data comparison, thereby improving the overall reliability and accuracy of the system.

[0022] In this embodiment, the specific steps of step 2 are: Step 2.1, according to the delay difference of the multi-channel microphone array, the spatial position of the sound source is calculated by ultrasonic sound source localization technology, and the accurate sound source position is obtained by using the delay estimation algorithm; the sound wave signals collected by multiple microphone arrays can be used to calculate the time difference of each sensor receiving the sound wave by using the delay difference algorithm, and the analysis of this delay difference can be accurately located according to the geometric relationship; the location of the sound source is accurately estimated by the delay difference method, making this method particularly suitable for accurate positioning of the fault point of power equipment, ensuring blind spot monitoring and rapid response; using the delay estimation algorithm to process the signal can effectively reduce the impact of environmental noise on the positioning results and improve system stability; Step 2.2, use the sound wave propagation model to optimize the signal delay, and then use the least squares method to calculate the delay; the sound wave propagation model can be used to optimize the delay of the original signal. By considering the impact of environmental factors (such as temperature, humidity, etc.) on the speed of sound wave propagation, the accuracy of delay estimation can be greatly improved; the least squares method is an optimization algorithm that can select the best solution from multiple possible solutions and reduce the impact of measurement errors on positioning results. Therefore, the application of the least squares method can still provide a relatively accurate sound source location estimate in a high noise environment; the least squares method enhances the robustness of the system, can effectively deal with external noise and imperfect data, and ensure that the positioning results are not affected by extreme interference; Step 2.3: Estimate the sound source position according to the time delay difference calculation formula. The specific formula is: ; Among them, x s is the estimated sound source position, d i is the distance between the ith microphone and the sound source, d i (m) is the distance calculated according to the propagation model, and N is the number of microphones. This formula optimizes the estimation of the sound source position by minimizing the distance difference between each microphone and the sound source. By accurately calculating the delay difference, the specific position of the sound source can be effectively located, especially in complex equipment areas, to ensure accurate positioning. By calculating the signal delay differences of all microphones, data can be fused in multiple directions to improve the accuracy of the estimation results, thereby further optimizing the accuracy of sound source positioning.

[0023] In step 2, the Kalman filter algorithm is also used to determine the location of the sound source. The specific formula is: ; Among them, x k│k is the state estimate at the current moment, K k is the Kalman gain, z k is the observed value, H k Peripheral observation matrix; Kalman filter algorithm is a recursive algorithm used to estimate the state of dynamic systems. It can update and correct the system in real time. In sound source positioning, Kalman filter can continuously optimize the location estimation of the sound source based on historical data and new observation results; Kalman filter can steadily improve the positioning accuracy under the influence of signal noise and environmental interference. By estimating the state at multiple moments, it can effectively reduce the impact of accidental errors and improve the stability of sound source positioning; in complex power equipment environments, dynamic changes and noise source interference may occur. Kalman filter can quickly adapt to these changes, filter out irrelevant signals, and ensure accurate tracking of the location of the sound source through the Kalman gain K k The filter can dynamically adjust the prediction accuracy according to the current signal quality, so that the system can adaptively optimize the positioning accuracy under different working conditions.

[0024] These steps can provide high-precision sound source localization by using time delay estimation algorithms, sound wave propagation model optimization, Kalman filtering and other technologies. The effect of each step enhances the robustness and accuracy of the system in complex power equipment environments. Through the combination of these technologies, real-time, stable and accurate sound source localization can be achieved, especially when monitoring power equipment faults, which can greatly improve fault diagnosis efficiency and response speed.

[0025] Step 3: Construct an acoustic image according to the spatial position of the sound source to display the distribution of the sound source caused by the equipment failure; By constructing an acoustic image, the system can visualize the spatial distribution of sound sources, which helps to more intuitively understand the distribution of sound sources in the equipment fault area, thereby quickly determining the area where the problem occurs. The acoustic image can display the distribution of sound sources in space, which not only helps to display the fault point, but also provides more environmental context information, such as whether the sound source is concentrated in a specific device or area, which facilitates judgment on whether the equipment has abnormalities or faults. Through the visualization of the acoustic image, patrol personnel or equipment maintenance personnel can more quickly identify potential fault sources, make effective decisions and troubleshoot, and avoid wasting time on manual positioning and inspection.

[0026] In this embodiment, the specific steps of step 3 are: Step 3.1. Generate an acoustic image according to the spatial position of the sound source to display the distribution of the sound source in the power equipment area, and use a three-dimensional image processing algorithm to draw a three-dimensional distribution map of the sound source; by generating a three-dimensional acoustic image, the spatial position and distribution of the sound source in the power equipment area can be clearly displayed, which makes the detection of equipment faults more directional, especially when multiple sound sources exist at the same time, the three-dimensional image can intuitively show the distribution of these sound sources in space; the three-dimensional image processing method can more accurately display the specific position of the sound source and its spatial relationship in the equipment area. In a complex equipment environment, traditional two-dimensional images may not be able to fully display all key information, and three-dimensional acoustic images can provide more dimensional data to help locate the fault source more accurately; through the three-dimensional image of the sound source distribution, the inspection personnel can quickly determine the specific fault location of the equipment, optimize the inspection path, and reduce positioning errors and missed inspections; Step 3.2: Use the interpolation algorithm to smooth the sound source intensity to obtain a uniformly distributed sound source intensity map, and determine the specific location and intensity of the sound source in space; Using the interpolation algorithm to smooth the sound source intensity can eliminate signal fluctuations and unevenness caused by equipment or environmental factors (such as uneven sensor spacing, noise, etc.), thereby obtaining a smoother and more uniform sound source intensity map. Through smoothing, more accurate sound source intensity information can be provided; during the processing, the interpolation algorithm can help eliminate local errors in the intensity data, making the sound source intensity at different locations more comparable, which is helpful for further analysis of the equipment. The uniformly distributed sound source intensity map obtained after interpolation can more accurately locate the position of the sound source and provide the intensity information of the sound source, which is crucial for further analyzing the severity and scope of the fault and the priority sorting when taking repair measures. The smoothed intensity map can provide more coherent sound source data, which has better support for subsequent intelligent analysis and processing such as machine learning and fault pattern recognition. Combined with the intensity information of the sound source, the system can identify more potential fault characteristics and make corresponding warnings.

[0027] Step 4: Superimpose the acoustic image with the visible light image of the equipment's surrounding environment to obtain a comprehensive imaging map, accurately showing the relationship between the three-dimensional position of the equipment's fault point and the surrounding environment; By superimposing the sound and image with the visible light image, the system can simultaneously present the spatial distribution of the equipment fault source and the visual information of the equipment and the surrounding environment. This not only provides the location information of the sound source, but also shows the specific position of the location relative to the equipment and the surrounding environment, helping personnel to locate the fault point more quickly; by fusing the sound and image with the visible light image, the comprehensive imaging diagram shows the relationship between the equipment's fault point and its surrounding environment, thereby providing equipment maintenance personnel with richer background information, making fault diagnosis more accurate and positioning more intuitive.

[0028] In this embodiment, the specific steps of step 4 include: Step 4.1. Obtain image data of the equipment area in real time through a visible light camera, and record the environmental image around the equipment in real time. By obtaining image data of the equipment area in real time, it can be ensured that the environmental information is up to date and can reflect the actual situation of the equipment and the surrounding environment in real time. In this way, image lag or inaccuracy caused by environmental changes (such as changes in lighting, equipment displacement, etc.) can be avoided. Through high-resolution visible light image recording, the environmental details around the equipment can be captured and clear visual information can be provided, which provides the necessary basis for subsequent image processing, fault analysis and equipment repair. Real-time recording of visible light images of the equipment's surrounding environment can also provide long-term image records for equipment maintenance and management, facilitating comparative analysis of historical data and tracking of equipment changes. Step 4.2: Process the visible light image to align it with the sound source position in the acoustic image, and superimpose the two through image fusion technology to generate a comprehensive image. Process the visible light image to align it with the sound source position in the acoustic image to ensure the spatial consistency between the image and the sound source positioning data. This allows the position information of the visible light image to accurately match the spatial distribution of the sound source, thereby improving the accuracy of the comprehensive image. Through image fusion technology, two different types of data (sound source position and environmental image) are effectively superimposed to create a comprehensive image. Image fusion technology can retain the accuracy of the sound source position information and simultaneously display the details of the surrounding environment to form a multi-dimensional, actionable visualization result. The fused comprehensive image can clearly show the three-dimensional position and environmental relationship of the equipment fault point, making the fault location more intuitive. The staff can immediately understand the equipment and its surrounding environment where the fault point is located through this image, and quickly formulate a maintenance plan.

[0029] Step 5: Classify different types of fault sound sources, and further determine the type and cause of the fault based on the frequency and intensity characteristics of the sound source; By classifying different types of fault sound sources, the system can automatically identify the fault type and reduce errors and time delays in manual judgment. Combined with the frequency and intensity characteristics of the sound source, the system can more accurately determine the fault type, such as partial discharge, gas leakage and abnormal equipment noise, thereby improving the automation level of fault detection and diagnosis; by combining frequency and intensity characteristics, the system can establish an accurate classification model based on the sound source characteristics of different faults, thereby improving the accuracy of diagnosis and avoiding misjudgment; through more refined classification, it can further determine the cause of the fault and distinguish different types of faults, support targeted maintenance and treatment plans, and thus improve the maintenance efficiency and safety of the power system.

[0030] The specific steps of step 5 include: Step 5.1. Extract the frequency characteristics of the sound source through spectrum analysis, and classify it in combination with the intensity change of the fault sound source, and determine the fault type; Through spectrum analysis, the frequency characteristics of the fault sound source can be effectively extracted. The sound waveform generated by each fault type has different frequency characteristics. For example, partial discharge usually produces high-frequency noise, while abnormal equipment noise may appear as low-frequency noise. Through spectrum analysis, these characteristics can be extracted from the collected sound wave signal to provide a basis for subsequent fault judgment; Combined with the intensity change of the sound source for fault classification, the robustness of the classification model can be enhanced. For example, partial discharge may produce large intensity fluctuations in a short period of time, while gas leakage may produce continuous small fluctuations. By analyzing these intensity characteristics, the system can more accurately determine the fault type; Through the combination of spectrum and intensity, the recognition accuracy of complex fault types can be improved, avoiding the misclassification of certain fault types as other types, and reducing false alarms and missed alarms; Step 5.2: Input the frequency and intensity data into the classification model to predict the fault type through the machine learning algorithm. The fault types include but are not limited to partial discharge, gas leakage and abnormal noise of equipment. Inputting the frequency and intensity data into the machine learning algorithm can enable the system to automatically learn from the data and predict the fault type. The machine learning model (such as support vector machine, decision tree, neural network, etc.) can automatically optimize the classification decision and improve the prediction accuracy by learning a large amount of historical fault data. The machine learning algorithm can comprehensively analyze multiple features (such as frequency, intensity, time characteristics, etc.) to provide more accurate fault classification results. Compared with traditional manual classification methods, machine learning methods can handle more complex pattern recognition problems and improve the system's ability to identify complex faults. Through the machine learning model, the system can quickly analyze the data collected in real time and immediately give the prediction results of the fault type, which helps to achieve instant response and early warning of faults and reduce the downtime of power equipment due to faults.

[0031] Among them, the training steps of the classification model include: Collect a large amount of equipment failure data to extract the frequency, intensity and time characteristics of the sound source as training data; use supervised learning algorithms to train classification models and automatically classify different types of failures; by collecting a large amount of equipment failure data, the system can comprehensively cover different types of equipment failure scenarios and extract the frequency, intensity, time and other characteristics of the sound source. These multi-dimensional data provide rich training samples for machine learning models and enhance the generalization ability of the models; through supervised learning algorithms, machine learning models can learn and find patterns of different fault types on labeled training data, so that the model can classify new data more efficiently and accurately, and continuously optimize its own judgment ability; by continuously training and optimizing new fault data, the classification model can adapt to changes in power equipment under different working conditions, so that over time, the model will become more and more accurate and can identify more fault types and subtle differences.

[0032] Step 6: Real-time feedback of fault diagnosis results, displaying the specific location of the fault point, and providing targeted inspection and repair suggestions; by real-time feedback of fault diagnosis results, the system can quickly provide accurate fault information after a fault occurs, thereby shortening the time for fault handling and repair, and improving the stability and operating efficiency of power equipment; by displaying the specific location of the fault point and providing repair suggestions, the system helps inspection personnel and maintenance personnel make decisions quickly, and targeted inspection and repair plans can accurately locate problems, avoid ineffective inspections and repairs, and reduce equipment downtime and repair costs; the generation and push function of real-time diagnostic reports improves the level of intelligence in the equipment maintenance process, reduces human errors and response delays, and further improves operation and maintenance efficiency.

[0033] The specific steps of step 6 include: Step 6.1. Generate a fault diagnosis report in real time based on the type and location of the sound source. The diagnosis report includes the fault type, location and recommended maintenance plan. Based on the type and location of the sound source, the system can quickly generate a diagnosis report to provide real-time fault information to equipment maintenance personnel, which can ensure that maintenance personnel can obtain accurate diagnosis results as soon as possible after the fault occurs, thereby improving response speed. The diagnosis report includes the fault type, location and recommended maintenance plan, which helps inspection personnel clearly understand the nature, location and treatment method of the fault. The detailed information in the report enables the staff to clearly understand the severity of the fault, the parts that need to be processed and the repair steps, avoiding misjudgment and unnecessary repair work. By providing a specific maintenance plan, the system not only provides fault information, but also provides practical operation guidance, which provides maintenance personnel with a clear repair direction and improves repair efficiency and quality. Step 6.2: Pushing the fault diagnosis report to the smart device of the inspector through the wireless communication network can ensure that the inspector can receive the latest fault information in a timely manner no matter where he is, avoiding response delays caused by information delays; by sending the diagnosis report directly to the smart device of the inspector, the time for personnel to transmit information is saved, and the inspector can view the diagnosis report at any time and take immediate action. This process greatly improves the timeliness of equipment fault response and inspection efficiency; with the help of smart devices, the inspector can view the detailed content of the report in real time, and conduct targeted inspections and repairs according to the suggestions in the report. In this way, the inspection work is more accurate and intelligent, avoiding the deviation of manual judgment.

[0034] Step 7: Analyze the sound source data in conjunction with the machine learning model, automatically detect the health status of the equipment, predict potential failures and give early warnings. By analyzing the sound source data in conjunction with the machine learning model, relevant features of the equipment health status are extracted, and potential failures are further predicted and early warnings are issued. This step is based on historical data and real-time collected sound source signal data, and is processed through intelligent algorithms to automatically identify the operating status and potential risks of the equipment.

[0035] The specific implementation of step 7 is: Data collection and feature extraction: The system first collects real-time sound source data from the device, which includes the frequency, intensity, time and other characteristics of the sound wave signal. These data will serve as input for the machine learning model; Machine learning model training: Use historical equipment failure data to train a machine learning model. The training data includes the normal operation data and historical failure data of the equipment. Through the machine learning algorithm, the model can learn the pattern differences between the normal and failure states of the equipment. Real-time data analysis and health monitoring: By inputting the real-time collected sound source data into the trained model, the machine learning model will analyze the current status of the device and determine its health status. If the device has abnormal signals (such as changes in frequency or intensity), the model will identify these abnormalities based on the learned patterns; Potential failure prediction and early warning: Once a potential failure is detected, the machine learning model will predict the type of failure or equipment problem that may occur based on the trends and patterns of historical data, and issue early warnings in a timely manner. This early warning can help equipment operators prepare in advance and avoid serious failures.

[0036] The present invention also provides an acoustic imaging positioning system for intelligent monitoring of regional faults in electric power engineering projects, and implements the above-mentioned acoustic imaging positioning method for intelligent monitoring of regional faults in electric power engineering projects, including: The sound source acquisition module is used to deploy a microphone array to collect sound wave signals in the power equipment area, and monitor and record sound data in real time; The sound source localization module is used to use ultrasonic sound source localization technology to perform time delay estimation and multi-channel signal processing on sound wave signals to determine the spatial position of the sound source; The sound image generation module is used to construct a sound image according to the spatial position of the sound source to show the distribution of the sound source caused by the equipment failure; Image overlay module, used to overlay the acoustic image with the visible light image of the equipment's surrounding environment to obtain a comprehensive image that accurately displays the relationship between the three-dimensional position of the equipment's fault point and the surrounding environment; Fault classification and diagnosis module, which is used to provide real-time feedback on fault diagnosis results, display the specific location of the fault point, and provide targeted inspection and repair suggestions; The early warning module is used to analyze sound source data in combination with machine learning models, automatically detect the health of equipment, predict potential failures and issue early warnings.

[0037] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering projects, characterized in that: The following steps are involved: Step 1: deploy a microphone array to collect sound wave signals in the power equipment area, and monitor and record sound data in real time; Step 2: Using ultrasonic sound source localization technology, the sound wave signal is subjected to time delay estimation and multi-channel signal processing to determine the spatial position of the sound source; Step 3: Construct an acoustic image according to the spatial position of the sound source to display the distribution of the sound source caused by the equipment failure; Step 4: Superimpose the acoustic image with the visible light image of the equipment's surrounding environment to obtain a comprehensive imaging image, accurately showing the relationship between the three-dimensional position of the equipment's fault point and the surrounding environment; Step 5: Classify different types of fault sound sources, and further determine the type and cause of the fault based on the frequency and intensity characteristics of the sound source; Step 6: Real-time feedback of fault diagnosis results, displaying the specific location of the fault point, and providing targeted inspection and repair suggestions; Step 7: Analyze sound source data with machine learning models to automatically detect equipment health, predict potential failures, and issue warnings.

2. The acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering according to claim 1 is characterized in that: The specific steps of step 1 are: Step 1.1, deploy multiple microphone arrays in the equipment area to collect sound wave signals in the power equipment; Step 1.2: Perform preliminary filtering on the collected sound wave signals to remove background noise and environmental interference.

3. The acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering according to claim 1 is characterized in that: The specific steps of step 2 are: Step 2.1, according to the delay difference of the multi-channel microphone array, the spatial position of the sound source is calculated by ultrasonic sound source localization technology, and the accurate sound source position is obtained by using the delay estimation algorithm; Step 2.2, using the sound wave propagation model, optimize the signal delay, and then use the least square method to calculate the delay; Step 2.3: Estimate the sound source position according to the time delay difference calculation formula. The specific formula is: ； Among them, x s is the estimated sound source position, d i is the distance between the ith microphone and the sound source, d i (m) is the distance calculated according to the propagation model, and N is the number of microphones.

4. The acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering according to claim 1 is characterized in that: The specific steps of step 3 are: Step 3.1, according to the spatial position of the sound source, generate an acoustic image to show the distribution of the sound source in the power equipment area, and use a three-dimensional image processing algorithm to draw a three-dimensional distribution map of the sound source; Step 3.2: Use an interpolation algorithm to smooth the sound source intensity to obtain a uniformly distributed sound source intensity map, and determine the specific position and intensity of the sound source in space.

5. The acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering according to claim 1 is characterized in that: The specific steps of step 4 include: Step 4.1, obtain image data of the device area in real time through a visible light camera, and record the environmental image around the device in real time; Step 4.2: Process the visible light image to align it with the sound source position in the sonogram, and superimpose the two through image fusion technology to generate a comprehensive imaging image.

6. The acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering according to claim 1 is characterized in that: The specific steps of step 5 include: Step 5.1, extract the frequency characteristics of the sound source through spectrum analysis, classify it according to the intensity change of the fault sound source, and determine the fault type; Step 5.2: Input the frequency and intensity data into the classification model to predict the fault type through the machine learning algorithm. The fault type includes but is not limited to partial discharge, gas leakage and abnormal noise of equipment.

7. The acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering according to claim 6 is characterized in that: The training steps of the classification model include: Collect a large amount of equipment failure data to extract the frequency, intensity and time characteristics of the sound source as training data; The classification model is trained using supervised learning algorithms to automatically classify different types of faults.

8. The acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering according to claim 1 is characterized in that: The specific steps of step 6 include: Step 6.1, based on the type and location of the sound source, a fault diagnosis report is generated in real time, the diagnosis report including the fault type, location and recommended maintenance plan; Step 6.2: Push the diagnostic report to the smart device of the inspector via the wireless communication network.

9. The acoustic imaging positioning method for intelligent monitoring of regional faults in power engineering according to claim 1 is characterized in that: In step 2, the Kalman filter algorithm is also used to determine the location of the sound source. The specific formula is: ； Among them, x k│k is the state estimate at the current moment, K k is the Kalman gain, z k is the observed value, H k Peripheral observation matrix.

Images