Power equipment operation state evaluation system and evaluation method

Abstract

The application discloses a kind of power equipment operating state evaluation system and evaluation method, it is related to power equipment state monitoring and evaluation technical field, the system includes data acquisition module, data integration module, data analysis module, data fusion module, state evaluation module and decision support module;The present application integrates a variety of data acquisition means, including traditional sensor data acquisition, image and video information acquisition of equipment appearance and terahertz nondestructive testing technology, realizes the comprehensive, real-time monitoring of power equipment operating state, this diversified data acquisition mode not only improves the accuracy and integrity of data collection, but also provides information basis for subsequent data analysis and state evaluation, the introduction of terahertz detection technology makes it possible to detect the internal insulation materials and structure of the equipment, fills the gap of traditional detection means in this regard, improves the accuracy and reliability of power equipment state evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power equipment state monitoring and evaluation, in particular to a power equipment operation state evaluation system and evaluation method. BACKGROUND

[0002] In the power industry, the operation state monitoring and evaluation of power equipment is a key link to ensure the safe and stable operation of the power grid. With the continuous development of science and technology, power equipment is becoming increasingly complex, and higher requirements are put forward for the monitoring of its operation state. At present, the power industry is actively exploring and applying new technologies and methods to improve the accuracy and efficiency of equipment monitoring. Against this background, a power equipment operation state evaluation system that integrates multiple data collection and analysis methods has emerged, aiming to achieve comprehensive, real-time and accurate monitoring of the operation state of power equipment.

[0003] Traditional power equipment operation state monitoring technology mainly relies on single sensor data collection and analysis, such as voltage transformers, current transformers and temperature sensors. Although these technologies can reflect the operation state of the equipment to some extent, they have many shortcomings. On the one hand, a single data source cannot fully reflect the overall operation state of the equipment, and key information may be missed. On the other hand, traditional technology has blind spots in monitoring the internal insulation materials and structure of the equipment, making it difficult to discover potential safety hazards. In addition, traditional technology has limited data analysis and processing capabilities, making it difficult to meet the needs of large-scale and complex power equipment monitoring. SUMMARY

[0004] The purpose of the present application is to overcome the shortcomings of the prior art and provide a power equipment operation state evaluation system and evaluation method. The present application integrates data collection, integration, analysis, fusion, evaluation and decision support, achieving comprehensive, real-time and accurate monitoring of the operation state of power equipment. By using multiple sensors, cameras and terahertz detection equipment, it can obtain multi-dimensional information inside and outside the equipment. By using advanced data analysis algorithms and fusion technology, it can construct a feature vector reflecting the current operation state of the equipment. Combined with historical data and comprehensive evaluation functions, it can accurately judge the operation state of the equipment, providing scientific decision support for operation and maintenance personnel, and effectively improving the safety operation level and maintenance efficiency of power equipment.

[0005] To solve the above technical problems, the present application provides the following technical solutions: on the one hand, a power equipment operation state evaluation system, which includes a data collection module, a data integration module, a data analysis module, a data fusion module, a state evaluation module and a decision support module.

[0006] The data acquisition module: equipped with various sensors for collecting operation data of power equipment; equipped with a camera to obtain image and video information of the appearance of the equipment from different angles; equipped with a terahertz detection device for non-destructive detection of the internal insulation materials and structure of the equipment using terahertz waves; all collection devices transmit the collected raw data to the data integration module in real time through wireless communication;

[0007] The data integration module: receives data from the data acquisition module, performs format checking and cleaning operations on the data, removes invalid data and noise data, synchronizes the time stamp of the cleaned data, and aligns the data of different collection devices on the time axis;

[0008] The data analysis module: uses different algorithms to analyze different types of data. For traditional sensor data, it performs regular numerical analysis and trend judgment. For image data, it identifies whether the equipment has appearance damage, oil leakage signs, or foreign matter attachment. For terahertz data, it performs frequency spectrum analysis based on the propagation characteristics and reflection / refraction laws of terahertz waves in different media to determine the moisture content of internal insulation materials, the location and size of micro-defects, and the aging degree of insulation materials. The analysis results are transmitted to the data fusion module in a structured data format;

[0009] The data fusion module: after receiving various analysis results from the data analysis module, uses information entropy to determine the weight of each type of data in the fusion process, and through data fusion algorithms, it fuses data from different sources, considers various factors, and constructs a feature vector that reflects the current operating state of the transformer, which is output to the state evaluation module;

[0010] The state evaluation module: collects various data samples of past power equipment under different operating conditions, including normal operation, abnormal operation before failure, and failure data, and labels and preprocesses these data. Through a comprehensive evaluation function, it constructs a power equipment operating state evaluation model. Let the device state feature vector output by the data fusion module be where is the number of features, the health baseline vector is defined as , the fault feature vector is defined as , the deviation degree of the device state feature vector from the health baseline vector is calculated as , and the formula is: where is the weight of the i-th feature, and the closeness of the device state feature vector to the fault feature vector is calculated as , and the formula is: where is another weight parameter, e is the base of natural logarithm, the running state of the equipment is judged by comprehensive evaluation function , and the formula is: The model is trained through historical data, when receiving the equipment state feature vector output by the data fusion module, the equipment state feature vector is input to the evaluation model for calculation to output the comprehensive evaluation score of the equipment, the corresponding state level is given according to the comprehensive evaluation score, and the evaluation result is transmitted to the decision support module;

[0011] The decision support module: according to the equipment running state evaluation result given by the state evaluation module, decision support information is provided for operation and maintenance personnel, when the equipment is in normal operation, optimization operation suggestion is provided, when the equipment is in abnormal state, specific troubleshooting steps and maintenance scheme suggestion are given, and the decision support module is interacted with external systems.

[0012] Further, the sensors used in the data acquisition module are: voltage transformers, current transformers and pressure sensors.

[0013] Further, the processing of the received data in the data integration module is:

[0014] For sensor data, check whether the numerical range and data type of the data meet the preset standard, and repair or exclude abnormal data points;

[0015] For image and video data, check file integrity and encoding format, detect image integrity, for video data, check frame rate stability and picture continuity, mark and repair the fragments with frame loss and screen damage;

[0016] For data collected by terahertz detection equipment, verify whether the signal strength and frequency range parameters are within the normal detection range, check the integrity of the phase information of the data, identify abnormal signal fluctuation data points, and correct or exclude data that does not meet the physical law.

[0017] Further, the data analysis module analyzes the sensor data, uses dynamic weighted statistical method to calculate the statistical characteristics, sets the data sequence as , introduces the time decay factor , and , calculates the weighted mean to understand the real-time trend of the data, and the formula is: , calculates the weighted variance to measure the dispersion degree and stability change of the data, and the formula is: , adopts an adaptive fitting algorithm to draw a trend curve of the data changing with time to capture the nonlinear change trend of the data, sets the time sequence as , the corresponding sensor data is , and the formula is: wherein is the start time, m is the polynomial order, is the coefficient to be solved.

[0018] Further, the analysis of the equipment image data in the data analysis module is carried out through the appearance damage detection formula, the pixel gray value matrix of the equipment region in the image is G, the size is m*n, two-dimensional discrete wavelet transform is carried out, the low-frequency coefficient matrix LL and the high-frequency coefficient matrix LH, HL and HH are obtained, the damage feature quantity is defined as D, the formula is: , a threshold value is set , when , it is determined that the equipment has appearance damage; the leakage oil sign detection formula is used for leakage oil sign detection, the image is converted from the RGB color space to the HSV color space, the saturation channel matrix is obtained , the average saturation is calculated , the formula is: , wherein m and n are the number of rows and columns of the image, a threshold value T is set, when , it is determined that there is a leakage oil sign; the foreign matter adhesion detection formula is used for foreign matter adhesion detection, edge detection is carried out on the image to obtain an edge image E, the pixel value is , the edge is 1 and the non-edge is 0, at the same time, the normal contour of the equipment is modeled to obtain a contour template TL, the pixel value is , the contour is 1 and the contour is 0, the foreign matter adhesion feature quantity F is defined, and the calculation formula is: , a threshold value is set, when , it is determined that there is foreign matter adhesion.

[0019] Further, the analysis of the terahertz detection data in the data analysis module is carried out, the moisture content of the insulating material in the equipment is determined through the formula, the propagation speed of the terahertz wave in the insulating material is v, the frequency is f, the phase change after passing through the material is , the standard phase change of the material in the dry state is , and the calculation formula of the moisture content M is: , wherein is a proportional coefficient; the position and size of the micro defect are determined through the formula, the reflection signal intensity of the terahertz wave is , the average reflection intensity of the defect-free region is , the relative intensity deviation function is , the calculation formula is: , when , is a preset threshold value, it is determined that The position has a small defect, the size of the defect is estimated by integrating the area satisfying , wherein and are the spatial sampling intervals; the aging degree of the insulation material is determined by the formula, wherein the energy integral value of the terahertz spectrum in the low frequency band is , the energy integral value in the high frequency band is , the high-to-low frequency energy ratio of the new device is , the high-to-low frequency energy ratio of the current detection is , and the aging degree of the insulation material is , the calculation formula of d is , wherein is the aging coefficient.

[0020] Further, the data fusion module uses information entropy to determine the weight of each type of data in the fusion process, wherein is the probability of the jth data point in the ith type of data, which is obtained by statistical frequency for discrete data, and the information entropy , the calculation formula is , the weight of the ith type of data is , wherein n is the total number of data categories.

[0021] Further, the data fusion module uses the multi-source data fusion formula to organically fuse different types of data after normalization and weight allocation, wherein the data after normalization and weight allocation is and , and the formula is , wherein is the adaptive coefficient of the jth data point in the ith type of data.

[0022] Further, the state evaluation module gives a corresponding state grade evaluation according to the comprehensive evaluation score, when , it is evaluated as a normal state; when , it is evaluated as an abnormal state; and when , it is evaluated as a fault state.

[0023] On the other hand, a power equipment operation state evaluation method, the specific steps of the evaluation method are:

[0024] S100, data acquisition: using a variety of sensors to collect power equipment operation data, using a camera to obtain device appearance images and videos, using a terahertz detection device to detect device internal insulation materials and structure information, and transmitting the raw data collected by the data acquisition device to the data integration module in real time through wireless communication; ​​

[0025] S200, data integration: after receiving the collected data, the format of different types of data is checked and cleaned, including checking the numerical range of sensor data, detecting the integrity of image video data, verifying the signal parameters of terahertz detection data, synchronizing the timestamp alignment data, and transmitting the processed data to the data analysis module;

[0026] S300, data analysis: different algorithms are used to analyze different types of data, and for sensor data, regular numerical analysis and trend judgment are performed; image data is detected to determine whether there is appearance damage, oil leakage, or foreign matter attachment; terahertz detection data is used to determine the moisture content, small defects, and aging degree of insulating materials according to relevant characteristics and laws, and the analysis results are transmitted to the data fusion module;

[0027] S400, data fusion: after receiving the analysis results, the information entropy is used to determine the weight of each type of data, and then different types of data are fused through a multi-source data fusion formula to construct a device operating state feature vector, which is output to the state evaluation module;

[0028] S500, state evaluation: collect device data samples under different operating conditions and pre-process them, construct a power device operating state evaluation model through a comprehensive evaluation function, and set the device state feature vector output by the data fusion module as , wherein is the number of features, the health baseline vector is defined as , the fault feature vector is defined as , the deviation degree of the device state feature vector from the health baseline vector is calculated as , and the formula is: , wherein is the weight of the i-th feature, the closeness of the device state feature vector to the fault feature vector is calculated as , and the formula is: , wherein is another weight parameter, e is the base of the natural logarithm, and the operating state of the device is judged through a comprehensive evaluation function , and the calculation formula is: , the power device operating state evaluation model trained through historical data is used to give a state rating to the received real-time data, and the results are transmitted to the decision support module;

[0029] S600, decision support: based on the state evaluation results, decision support is provided for operation and maintenance personnel, optimization suggestions are given for normal operation, and fault troubleshooting and maintenance scheme suggestions are given for abnormal operation, and information sharing and collaborative work are realized through interaction with external systems.

[0030] Compared with the prior art, the power device operating state evaluation system and evaluation method have the following beneficial effects:

[0031] One, the application realizes comprehensive and real-time monitoring of the running state of power equipment by integrating various data collection means, including traditional sensor data collection, image and video information collection of equipment appearance and terahertz nondestructive testing technology. This diversified data collection method not only improves the accuracy and integrity of data collection, but also provides an information foundation for subsequent data analysis and state evaluation. The introduction of terahertz detection technology makes it possible to detect the internal insulation materials and structure of the equipment, filling the gap in this aspect of traditional detection methods and improving the accuracy and reliability of power equipment state evaluation.

[0032] Two, the application performs format verification, cleaning and time stamp synchronization on the original data through the data integration module, ensuring the accuracy and consistency of the data. The data analysis module uses different algorithms to analyze various types of data and extract key information. This detailed data analysis makes the evaluation of the running state of power equipment more accurate and can provide more specific and targeted decision support for operation and maintenance personnel. At the same time, the data fusion module determines the weight of each type of data in the fusion process using information entropy, and uses a multi-source data fusion algorithm to organically fuse different types of data to construct a feature vector that can reflect the current running state of the equipment, further improving the accuracy and reliability of state evaluation. Other advantages, objectives and features of the application will be described to some extent in the subsequent specification, and will be apparent to those skilled in the art based on the study of the following text or can be taught from the practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1 The flow framework diagram of the state evaluation system of the application;

[0034] Figure 2 The flowchart of the state evaluation method of the application. DETAILED DESCRIPTION

[0035] To further illustrate the technical means and effects adopted by the application to achieve the predetermined invention purpose, the specific embodiments, structures, features and effects according to the application are described in detail as follows in combination with the drawings and preferred embodiments.

[0036] Example 1

[0037] 110kV substation main transformer running state evaluation

[0038] In the 110 kV substation, for the main transformer, the voltage transformer and the current transformer in the data acquisition module are respectively installed on the high-voltage side and the low-voltage side of the transformer to collect voltage and current data in real time; the temperature sensor is installed at the transformer oil tank and the winding part to monitor temperature changes; the pressure sensor is installed inside the oil tank to detect oil pressure; the camera is installed at a suitable position around the transformer to capture the overall appearance, bushing and radiator parts of the transformer, and has the functions of optical zoom, low-illumination shooting and wide dynamic range to obtain image and video information of the appearance of the equipment from different angles; the terahertz detection equipment performs non-destructive detection on the internal insulation materials and structure of the transformer during the transformer power-off maintenance; all the collected data are transmitted to the data integration module in real time through wireless communication. After receiving the data, the data integration module checks the sensor data to see whether the numerical range and data type meet the preset standards, such as the voltage value collected by the voltage transformer should be within the normal operating voltage range, and if there are data points outside the range, they are repaired or removed; for image and video data, the file integrity and encoding format are checked, the integrity of the image is detected, and for video data, the frame rate stability and picture continuity are checked; if it is found that part of the appearance image of the transformer is blurred (similar to the screen damage condition), it is marked and repaired; for the data collected by the terahertz detection equipment, the signal strength and frequency range parameters are verified to see whether they are within the normal detection range, the integrity of the phase information of the data is checked, and abnormal signal fluctuation data points are identified and corrected or excluded; then, through the time stamp synchronization algorithm, the data collected by different devices are aligned on the time axis, and the processed data are transmitted to the data analysis module.

[0039] For sensor data, a dynamic weighted statistical method is used to calculate its statistical characteristics, the data sequence is set as , a time decay factor is introduced, and , the weighted mean value is calculated to understand the real-time trend of the data, and the formula is: , the weighted variance is calculated to measure the dispersion degree and stability change of the data, and the formula is: , an adaptive fitting algorithm is used to draw the trend curve of the data change with time to capture the nonlinear change trend of the data.

[0040] For equipment appearance image data, the appearance damage detection formula is used for detection, the pixel gray value matrix of the equipment region in the image is set as G, the size is m x n, two-dimensional discrete wavelet transform is performed on it to obtain the low-frequency coefficient matrix LL and the high-frequency coefficient matrix LH, HL and HH, the damage feature quantity is defined as D, and the formula is: , a threshold is set, and when When the appearance damage condition is determined to exist, the image is converted from the RGB color space to the HSV color space by a leakage oil trace detection formula, and a saturation channel matrix is obtained , the average saturation is calculated , and the calculation formula is: A threshold T is set, and when , it is judged that there is a leakage oil trace, an edge image E is obtained by performing edge detection on the image by a foreign matter adhesion detection formula, and the pixel value of the edge image E is , the edge is 1, and the non-edge is 0; meanwhile, the normal contour of the transformer is modeled to obtain a contour template TL, and the pixel value of the contour template TL is , the contour is 1, and the contour is 0, a foreign matter adhesion feature quantity F is defined, and the calculation formula is: A threshold value is set, and when , it is determined that there is a foreign matter adhesion condition.

[0041] For the terahertz detection data, the moisture content of the insulating material inside the equipment is determined by a formula, the propagation speed of the terahertz wave in the insulating material is v, the frequency is f, and the phase change after passing through the material is , the standard phase change of the material in a dry state is , and the calculation formula of the moisture content M is: , wherein is a proportional coefficient calibrated by experiment, the position and size of the micro defect are determined by a formula, and the reflection signal intensity of the terahertz wave is , the average reflection intensity of the defect-free area is , the relative intensity deviation function is , and the calculation formula is: When , is a preset threshold value, it is determined that position exists a micro defect, and the size of the defect is estimated by integrating the area satisfying , and the formula is: , wherein and are spatial sampling intervals, the aging degree of the insulating material is determined by a formula, the energy integral value of the terahertz spectrum in the low frequency band is , the energy integral value in the high frequency band is , the high-to-low frequency energy ratio of the new equipment is: , the high-to-low frequency energy ratio of the current detection is: , and the aging degree d of the insulating material is calculated by the formula: , wherein The aging coefficient is then transmitted to the data fusion module.

[0042] The data fusion module receives the analysis results of each type of data, and uses information entropy to determine the weight of each type of data in the fusion process, where is the probability of the jth data point in the ith type of data, which is obtained by counting the frequency for discrete data, and the information entropy is The calculation formula is: The weight of the ith type of data is : where n is the total number of data categories, and then the different types of data that have been normalized and assigned weights are organically fused through the multi-source data fusion formula, which is: where is the adaptive coefficient of the jth data point in the ith type of data, and a feature vector reflecting the current operating state of the transformer is constructed and output to the state evaluation module.

[0043] The state evaluation module collects various data samples of the transformer in the past, such as oil temperature, winding temperature, voltage, and current data during normal operation, abnormal operation before failure, and failure, as well as data such as abnormal increase in oil temperature and partial discharge before failure, and short-circuit current and rapid increase in oil temperature during failure, and labels and preprocesses these data. Through a comprehensive evaluation function, a power equipment operating state evaluation model is constructed, where the equipment state feature vector output by the data fusion module is where is the number of features, the health baseline vector is defined as , the fault feature vector is defined as , the deviation degree of the equipment state feature vector from the health baseline vector is calculated as , and the formula is: where is the weight of the ith feature, the closeness of the equipment state feature vector to the fault feature vector is calculated as , and the formula is: where is another weight parameter, e is the base of natural logarithm, and the operating state of the equipment is determined by the comprehensive evaluation function , and the calculation formula is: The model is trained using historical data, and when the equipment state feature vector output by the data fusion module is received, it is input into the evaluation model for calculation to output the comprehensive evaluation score of the equipment. Let the calculation result be According to the state level evaluation rules, when , it is evaluated as normal state, when , it is evaluated as abnormal state; when When the temperature of the main transformer is higher than the normal temperature range, the main transformer is evaluated as a fault state, the main transformer is evaluated as a normal state, and the evaluation result is transmitted to the decision support module.

[0044] The decision support module gives specific troubleshooting steps according to the equipment operation state evaluation result given by the state evaluation module if the switch cabinet is in an abnormal state, such as checking whether the connection of the abnormal temperature part is loose according to the relevant data of the abnormal temperature part in data analysis, and then checking whether the insulation part has damage signs; giving maintenance scheme suggestions, such as tightening the loose connection and replacing the insulation part with aging signs, and interacting with the external system to share the switch cabinet state information and realize collaborative work, such as notifying the maintenance personnel to timely maintenance, feeding back the switch cabinet state information to the operation and maintenance management platform, so as to comprehensively manage the equipment operation of the entire distribution room.

[0045] In summary, by applying the power equipment operation state evaluation system to the 110kV substation main transformer, the data is comprehensively collected by using various sensors, cameras and terahertz detection equipment, and through the integration, analysis, fusion and evaluation links, the running state of the main transformer can be accurately judged, the decision support for operation and maintenance is provided, the safe and stable operation of the main transformer of the substation is effectively guaranteed, and the efficiency and practicability of the application in the state monitoring and management of large power equipment are embodied.

[0046] Example two:

[0047] 35kV distribution room switch cabinet operation state evaluation

[0048] In the 35kV distribution room, for the switch cabinet, the voltage transformer and the current transformer in the data acquisition module are installed at the inlet and outlet lines of the switch cabinet to collect voltage and current data; the temperature sensor is installed at the bus joint, the circuit breaker contact and the cable connection part in the switch cabinet which are prone to heat, to monitor the temperature change; the pressure sensor is installed in the gas chamber (if it is a gas insulated switch cabinet) of the switch cabinet to detect the gas pressure; the camera is installed at the appropriate position of the front and side of the switch cabinet to obtain the appearance image and video information of the equipment; the terahertz detection equipment performs nondestructive detection on the internal insulation material and structure of the switch cabinet when the switch cabinet is powered off for maintenance, and the equipment transmits the raw data to the data integration module in real time through wireless communication.

[0049] After receiving the data, the data integration module checks the value range and data type of the sensor data. For example, the current value collected by the current transformer should be within the normal working current range. If an abnormality occurs, it will be processed. For image and video data, its integrity and format are checked. If it is found that the image taken by the camera is partially missing (similar to an incomplete file), it will be marked and repaired. For terahertz detection data, the signal strength, frequency range parameters, and phase information are verified, and abnormal data is corrected or excluded. Then, the processed data is transmitted to the data analysis module. For equipment appearance image data, the appearance damage detection formula is used to detect whether the switch cabinet door and cabinet body are damaged, the leakage oil sign detection formula is used to detect whether there is leakage oil (for oil-immersed switch cabinets), and the foreign matter attachment detection formula is used to detect whether there is foreign matter attached to the surface or internal components of the switch cabinet.

[0050] For sensor data, a dynamic weighted statistical method is used. Let the data sequence be , a time decay factor is introduced, and , the weighted mean value is calculated to understand the real-time trend of the data, and the formula is: , the weighted variance is calculated to measure the dispersion degree and stability change of the data, and the formula is: , an adaptive fitting algorithm is used to draw the trend curve of data change over time to capture the nonlinear change trend of the data. Let the time sequence be , and the corresponding sensor data be , the formula is: , where is the starting time, m is the polynomial order, is the coefficient to be solved.

[0051] For equipment appearance image data, the appearance damage detection formula is used for detection. Let the pixel gray value matrix of the switch cabinet body area in the image be G, with a size of m x n. Perform two-dimensional discrete wavelet transform on it to obtain the low-frequency coefficient matrix LL and the high-frequency coefficient matrix LH, HL, and HH. Calculate the damage feature quantity, and the formula is: , set a threshold , when , it is determined that the equipment has appearance damage; the leakage oil sign detection formula is used for leakage oil sign detection. Convert the image from the RGB color space to the HSV color space to obtain the saturation channel matrix , calculate the average saturation , and the formula is: , set the threshold T, when , it is determined that there is a leakage oil sign. For oil-immersed switch cabinets, the foreign matter attachment detection formula is used. Perform edge detection on the image to obtain the edge image E, and the pixel value is , 1 at the edge and 0 at the non-edge, while modeling the normal contour of the switch cabinet to obtain a contour template TL, whose pixel value is , 1 inside the contour and 0 outside the contour, define the foreign matter adhesion feature quantity as F, and the calculation formula is: , set the threshold value , when , it is determined that there is foreign matter adhesion.

[0052] For terahertz detection data, the moisture content of the insulation material inside the equipment is determined by the formula, the propagation speed of terahertz waves in the insulation material is v, the frequency is f, and the phase change after passing through the material is , the standard phase change of the material in the dry state is , and the calculation formula of the moisture content M is: , wherein is a proportional coefficient calibrated by experiment, the position and size of the micro defect are determined by the formula, and the reflection signal intensity of the terahertz wave is , the average reflection intensity of the defect-free area is , the relative intensity deviation function is , and the calculation formula is: , when , is a preset threshold value, it is determined that position exists a micro defect, and the size of the defect is estimated by integrating the area that satisfies , and the formula is: , wherein and are spatial sampling intervals, the aging degree of the insulation material is determined by the formula, the energy integral value of the terahertz spectrum in the low frequency band is , and the energy integral value in the high frequency band is , the high-to-low frequency energy ratio of the new equipment is: , the high-to-low frequency energy ratio of the current detection is: , and the aging degree d of the insulation material is calculated by the formula: , wherein is an aging coefficient, and then the analysis result is transmitted to the data fusion module.

[0053] After receiving various analysis results, the data fusion module determines the weight of each type of data in the fusion process using information entropy, and sets as the probability of the jth data point in the ith type of data, which is obtained by statistical frequency for discrete data, and the information entropy is calculated by the formula: , and the weight of the ith type of data is: wherein n is the total number of categories of data, and then the different types of data that have been normalized and assigned weights are organically fused by a multi-source data fusion formula, which is wherein is the adaptive coefficient of the jth data point in the ith category of data, and a feature vector that can reflect the current operating state of the switch cabinet is constructed and output to the state assessment module.

[0054] The state assessment module collects various data samples of the switch cabinet in the past, such as temperature, voltage, and current data when the switch cabinet is in normal operation, data of increased partial discharge and abnormally high temperature before failure, and data of short-circuit current and insulation breakdown when the switch cabinet fails, and labels and preprocesses these data, constructs a power equipment operating state assessment model through a comprehensive evaluation function, and sets the equipment state feature vector output by the data fusion module as wherein is the number of features, the health baseline vector is defined as , the failure feature vector is defined as , the deviation degree of the equipment state feature vector from the health baseline vector is calculated as , and the formula is wherein is the weight of the ith feature, the proximity of the equipment state feature vector to the failure feature vector is calculated as , and the formula is wherein is another weight parameter, e is the base of the natural logarithm, and the operating state of the equipment is determined by the comprehensive evaluation function , and the calculation formula is The model is trained using historical data, and when the equipment state feature vector output by the data fusion module is received, it is input to the assessment model for calculation to output the comprehensive evaluation score of the equipment. Assuming that the value of S obtained by this calculation is S, according to the state level assessment rules, when , it is assessed as normal state; when , it is assessed as abnormal state; and when , it is assessed as failure state, the state of the switch cabinet is determined, and the assessment result is transmitted to the decision support module.

[0055] The decision support module gives specific troubleshooting steps according to the equipment operation state evaluation result given by the state evaluation module if the switch cabinet is in an abnormal state, such as checking whether the connection of the abnormal temperature part is loose according to the relevant data of the abnormal temperature part in the data analysis, and then checking whether the insulation part has damage signs; a maintenance scheme is suggested, such as tightening the loose connection and replacing the insulation part with aging signs, and the switch cabinet state information is shared with the external system of the power distribution room management system to realize collaborative work, such as notifying the maintenance personnel to timely maintenance, feeding back the switch cabinet state information to the operation and maintenance management platform, so as to comprehensively manage the equipment operation state of the entire power distribution room.

[0056] In summary, in the evaluation of the operation state of the switch cabinet in the 35kV power distribution room, the evaluation system of the present application is used, the modules work collaboratively from data collection to final decision support, the abnormal state can be found in time and targeted measures can be given through processing and analysis of the multi-aspect data of the switch cabinet, and the external system is interacted, which shows that the present application can effectively improve the equipment operation and maintenance level in the operation management of the medium and low voltage power equipment, ensures the reliable operation of the switch cabinet in the power distribution room, and has important practical application value. The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above, any person skilled in the art can make some changes or modifications to the equivalent embodiments without departing from the scope of the technical solution of the present application, as long as the changes or modifications do not depart from the technical solution of the present application. Any simple modification, equivalent change and modification of the above embodiments according to the technical essence of the present application are still within the scope of the technical solution of the present application.

Claims

1. A power equipment operating status evaluation system, characterized in that, The system comprises a data acquisition module, a data integration module, a data analysis module, a data fusion module, a state evaluation module and a decision support module; The data acquisition module is equipped with various sensors for collecting operation data of power equipment, a camera for obtaining image and video information of the appearance of the equipment from different angles, and a terahertz detection device for non-destructive detection of the internal insulation materials and structure of the equipment using terahertz waves; all the collected data are transmitted in real time to the data integration module through wireless communication; The data integration module receives data from the data acquisition module, performs format checking and cleaning operations on the data to remove invalid data and noise data, synchronizes the time stamp of the cleaned data, and aligns the data of different collection devices on the time axis; The data analysis module uses different algorithms to analyze different types of data; for traditional sensor data, it performs regular numerical analysis and trend judgment; for image data, it performs identification to determine whether the equipment has appearance damage, oil leakage signs or foreign matter adhesion; for terahertz data, it performs frequency spectrum analysis based on the propagation characteristics and reflection and refraction rules of terahertz waves in different media to determine the moisture content of the internal insulation materials of the equipment, the position and size of the micro defects, and the aging degree of the insulation materials, and transmits the analysis results in a structured data form to the data fusion module; The data fusion module receives various analysis results from the data analysis module, determines the weight of each type of data in the fusion process using information entropy, fuses data from different sources through a data fusion algorithm, considers various factors comprehensively, constructs a feature vector reflecting the current operating state of the transformer, and outputs it to the state evaluation module; The state evaluation module: collects various data samples of the past power equipment under different operating conditions, including normal operation, abnormal operation before failure and data when failure occurs, and labels and preprocesses these data, constructs a power equipment operating state evaluation model through a comprehensive evaluation function, and sets the equipment state feature vector output by the data fusion module as , wherein is the number of features, defines the health baseline vector as , defines the fault feature vector as , calculates the deviation degree of the equipment state feature vector and the health baseline vector , and the formula is , wherein is the weight of the ith feature, calculates the closeness of the equipment state feature vector and the fault feature vector , and the formula is , wherein is another weight parameter, e is the base of natural logarithm, the operating state of the equipment is judged through the comprehensive evaluation function , and the calculation formula is , and the model is trained through historical data, when the equipment state feature vector output by the data fusion module is received, it is input into the evaluation model for calculation to output the comprehensive evaluation score of the equipment, the corresponding state level is given according to the comprehensive evaluation score, and the evaluation result is transmitted to the decision support module; The decision support module provides decision support information for operation and maintenance personnel according to the equipment operating state evaluation results given by the state evaluation module, provides optimization operation suggestions when the equipment is in normal operation, and gives specific troubleshooting steps and maintenance scheme suggestions when the equipment is in an abnormal state, and simultaneously communicates with external systems.

2. The power equipment operation state evaluation system according to claim 1, characterized by The sensors used in the data acquisition module are voltage transformers, current transformers, temperature sensors and pressure sensors.

3. The power equipment operation state evaluation system according to claim 1, characterized by The processing of the received data in the data integration module includes: For sensor data, check whether the data value range and data type meet the preset standards, and repair or exclude abnormal data points; For image and video data, check file integrity and encoding format, detect image integrity, check frame rate stability and picture continuity for video data, and mark and repair segments with frame loss or screen damage; For data collected by the terahertz detection device, verify whether the signal strength and frequency range parameters are within the normal detection range, check the integrity of the phase information of the data, identify abnormal signal fluctuation data points, and correct or exclude data that does not meet the physical law.

4. The power equipment operation state evaluation system according to claim 1, characterized by The analysis of the data analysis module for sensor data includes: applying a dynamic weighted statistical method to calculate its statistical characteristics, setting the data sequence as , introducing a time decay factor , and , calculating the weighted mean to understand the real-time trend of the data, the formula is: , calculating the weighted variance to measure the dispersion degree and stability change of the data, the formula is: , using an adaptive fitting algorithm to draw a trend curve of data change over time to capture the nonlinear change trend of the data, setting the time sequence as , the corresponding sensor data is , the formula is: , wherein is the starting time, m is the polynomial order, is the coefficient to be solved.

5. The power equipment operation state evaluation system according to claim 1, characterized by, The analysis of the equipment image data in the data analysis module is analyzed by appearance damage detection formula appearance damage detection, assuming that the pixel gray value matrix of the equipment region in the image is G, the size is m x n, the two-dimensional discrete wavelet transform is performed on it, the low-frequency coefficient matrix LL and the high-frequency coefficient matrix LH, HL and HH are obtained, the damage feature quantity is defined as D, and the formula is: , a threshold value is set , when , it is determined that the equipment has appearance damage; leakage oil sign detection formula is used for leakage oil sign detection, the image is converted from RGB color space to HSV color space, and the saturation channel matrix is obtained , the average saturation is calculated, and the formula is: , wherein m and n are the number of rows and columns of the image, a threshold value T is set, when , it is determined that there is a leakage oil sign; foreign matter adhesion detection formula is used for foreign matter adhesion detection, edge detection is performed on the image to obtain an edge image E, the pixel value of which is , the edge is 1, and the non-edge is 0; at the same time, the normal contour of the equipment is modeled to obtain a contour template TL, the pixel value of which is , 1 inside the contour and 0 outside the contour, the foreign matter adhesion feature quantity F is defined, and the calculation formula is: , a threshold value is set , when , it is determined that there is foreign matter adhesion.

6. The power equipment operation state evaluation system according to claim 1, characterized by The analysis of the terahertz detection data in the data analysis module determines the moisture content of the insulation material in the equipment by formula, assuming that the propagation speed of terahertz waves in the insulation material is v, the frequency is f, the phase change after passing through the material is , and the standard phase change of the material in a dry state is , then the calculation formula of the moisture content M is: , wherein is a proportional coefficient; the position and size of the micro defect are determined by formula, and the reflection signal intensity of the terahertz wave is , assuming that the average reflection intensity of the defect-free area is , assuming that the relative intensity deviation function is , the calculation formula is: , when , is a preset threshold, and it is determined that the position exists a micro defect, and the size of the defect is estimated by integrating the area satisfying , and the formula is: , wherein and are spatial sampling intervals; The aging degree of the insulating material is determined by the formula, the energy integral value of the terahertz spectrum in the low frequency band is , the energy integral value in the high frequency band is , the high frequency to low frequency energy ratio of the new device is , the high frequency to low frequency energy ratio of the current detection is , and the aging degree of the insulating material is The calculation formula is , wherein is the aging coefficient.

7. The power equipment operation state evaluation system according to claim 1, characterized by The data fusion module uses information entropy to determine the weight of each type of data in the fusion process, and sets The probability of the jth data point in the ith type of data, for discrete data, is obtained by statistical frequency, and the information entropy The calculation formula is: The weight of the ith type of data is : Where n is the total number of data categories.

8. The power equipment operation state evaluation system according to claim 1, characterized by, The different types of data which have been normalized and assigned weights are organically fused by a multi-source data fusion formula in the data fusion module, and the data after normalization and weight assignment is denoted as and , and the formula is: wherein is an adaptive coefficient of the jth data point in the ith type of data.

9. The power equipment operation state evaluation system according to claim 1, characterized by, The state evaluation module gives corresponding state grade evaluation according to the comprehensive evaluation score, when the evaluation is normal state; when the evaluation is abnormal state; and when the evaluation is fault state.

10. A method for evaluating an operation state of a power device, characterized by, The specific steps of the evaluation method are: S100, data collection: using various sensors to collect power equipment operation data, using cameras to obtain equipment appearance images and videos, using terahertz detection equipment to detect internal insulation materials and structural information of the equipment, and transmitting the raw data collected by the collection equipment to the data integration module in real time through wireless communication; S200, data integration: after receiving the collected data, the different types of data are checked and cleaned, including checking the numerical range of sensor data, detecting the integrity of image and video data, verifying the signal parameters of terahertz detection data, synchronizing and aligning the time stamp data, and transmitting the processed data to the data analysis module; S300, data analysis: different algorithms are used to analyze different types of data, and for sensor data, regular numerical analysis and trend judgment are performed; image data is detected to determine whether there is appearance damage, oil leakage, or foreign matter attachment; terahertz detection data is used to determine the moisture content, small defects, and aging degree of insulation materials according to relevant characteristics and laws, and the analysis results are transmitted to the data fusion module; S400, data fusion: after receiving the analysis results, the information entropy is used to determine the weight of each type of data, and then the different types of data are fused through a multi-source data fusion formula to construct a feature vector reflecting the equipment operation state, and output to the state evaluation module; S500, state evaluation: collect equipment data samples under different operating conditions and pretreat, build power equipment operation state evaluation model through comprehensive evaluation function, set the equipment state feature vector output by the data fusion module as , wherein is the number of characteristics, define the health baseline vector as , define the fault feature vector as , calculate the deviation degree of the equipment state feature vector and the health baseline vector , the formula is , wherein is the weight of the i-th characteristic, calculate the closeness of the equipment state feature vector and the fault feature vector , the formula is , wherein is another weight parameter, e is the base of natural logarithm, the operation state of the equipment is judged through the comprehensive evaluation function , the calculation formula is , the power equipment operation state evaluation model trained through historical data is used to give a state grade assessment to the received real-time data, and the result is transmitted to the decision support module; S600, decision support: based on the state evaluation results, decision support is provided for operation and maintenance personnel, optimization suggestions are given for normal operation, and fault troubleshooting and maintenance scheme suggestions are given for abnormal operation, and information sharing and collaborative work are realized through interaction with external systems.

Images