Power equipment fault diagnosis system based on multi-sensor analysis

Abstract

The invention relates to the technical field of power fault diagnosis, in particular to a power equipment fault diagnosis system based on multi-sensor analysis, which comprises an equipment abnormality diagnosis processor, a preliminary false alarm evaluation unit, a progressive fault division unit, an insulation identification unit, a region positioning unit and a management early warning end, according to the method, analysis is performed preliminarily from the angle of the internal environment of the target power equipment, the false alarm risk of the target power equipment can be reduced, whether the target power equipment breaks down or not is further analyzed in an information feedback mode, and meanwhile the fault classification efficiency is improved along with the in-depth fault classification process. The fault type of the current target power equipment is known to be an insulation fault or a non-insulation fault so that targeted management can be performed according to the fault type, and defect feature identification analysis is performed on the surface feature image in an information progressive mode so that whether the insulation fault is caused by a material fault or a mechanical fault can be known. And thus, targeted management is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of power fault diagnosis, and particularly to a power equipment fault diagnosis system based on multi-sensor analysis. Background Art

[0002] Regarding the research on power system equipment fault diagnosis methods, many beneficial explorations have been made by predecessors. For example, common artificial intelligence technologies in the field of power system equipment fault diagnosis include expert systems, artificial neural networks, etc. In addition, in recent years, technologies such as data mining, fuzzy theory, rough set theory, support vector machines, applications of bionics, and multi-agent systems, as well as comprehensive applications of the above methods, have emerged.

[0003] However, in the prior art, it is impossible to monitor and analyze the internal environment of power equipment, resulting in an increased risk of false alarms for power equipment failures due to excessive changes in the internal environment, increasing the workload of workers, and being unable to manage rationally according to the fault categories of power equipment. At the same time, the factors causing faults cannot be accurately identified, which is not conducive to rational management, and the known and potential fault locations cannot be located, resulting in a decrease in the efficiency of operation and maintenance management and the remaining potential fault risks.

[0004] In view of the above technical deficiencies, a solution is now proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a power equipment fault diagnosis system based on multi-sensor analysis to solve the above-mentioned technical deficiencies. The present invention initially analyzes from the perspective of the internal environment of the target power equipment, which helps to reduce the false alarm risk of the target power equipment. And through the way of information feedback, further analyze whether the target power equipment is faulty. At the same time, along with the in-depth process of fault category division, to understand whether the current fault category of the target power equipment is an insulation fault or a non-insulation fault, so as to carry out targeted management according to the fault category. And through the way of information progression, identify and analyze the defect characteristics of the surface feature image to understand whether the insulation fault is caused by a material fault or a mechanical fault. At the same time, based on infrared technology, further locate the risk of insulation faults, so as to understand the known and potential risk locations in the insulation material, which helps to improve the operation and maintenance efficiency of insulation faults of the target power equipment and reduce the insulation fault risk.

[0006] The purpose of the present invention can be achieved by the following technical solutions: A power equipment fault diagnosis system based on multi-sensor analysis, including an equipment anomaly diagnosis processor, a preliminary false alarm evaluation unit, a progressive fault division unit, an insulation identification unit, a regional positioning unit, and a management warning terminal;

[0007] The device abnormal diagnosis processor is used to retrieve the internal environment information of the target power device, and send the internal environment information to the preliminary false alarm evaluation unit for false alarm potential risk evaluation and analysis to obtain a normal interference signal or a false alarm risk signal. When a normal interference signal is generated, the progressive fault classification unit is used to perform fault category classification and analysis on the collected initial insulation basic information to obtain an insulation fault signal or a non-insulation fault signal;

[0008] When an insulation fault signal is generated, the insulation identification unit is used to perform defect feature identification and analysis on the collected surface feature image to obtain a material fault signal or a mechanical fault signal. The area positioning unit is used to perform abnormal calibration analysis on the collected surface infrared feature image of the insulation area to obtain the spatial coordinates corresponding to the insulation fault point and the potential insulation fault point.

[0009] Preferably, the false alarm potential risk evaluation and analysis process is as follows:

[0010] Collect the energized period of the target power device, set the energized period of the target power device as the time threshold, obtain the internal environment information of the target power device within the time threshold. The internal environment information includes electric field strength, electric field frequency, and internal environment temperature, and perform discrimination processing on the internal environment information of the target power device. Furthermore, obtain the over-electric field strength period, over-electric field frequency period, and over-temperature period of the target power device within the time threshold, obtain the overlapping part of the over-electric field strength period, over-electric field frequency period, and over-temperature period, and set it as the overlapping duration;

[0011] Respectively obtain the field strength inducement duration, field frequency inducement duration, and field temperature inducement duration, obtain the preset field strength interference factor, preset field frequency interference factor, preset field temperature interference factor, and preset superposition interference factor corresponding to the field strength inducement duration, field frequency inducement duration, field temperature inducement duration, and overlapping duration. Furthermore, obtain the sum value of the preset field strength interference factor, preset field frequency interference factor, preset field temperature interference factor, and preset superposition interference factor, and set it as the insulation deviation interference factor, and perform discrimination processing on the insulation deviation interference factor to obtain a normal interference signal or a false alarm risk signal.

[0012] Preferably, the over-electric field strength period represents the time period when the electric field strength exceeds the preset electric field strength threshold, the over-electric field frequency period represents the time period when the electric field frequency exceeds the preset electric field frequency threshold, and the over-temperature period represents the time period when the internal environment temperature exceeds the preset internal environment temperature threshold.

[0013] Preferably, the field strength inducement duration represents the value obtained by subtracting the overlapping duration from the over-electric field strength period, the field frequency inducement duration represents the value obtained by subtracting the overlapping duration from the over-electric field frequency period, and the field temperature inducement duration represents the value obtained by subtracting the overlapping duration from the over-temperature period.

[0014] Preferably, the process of classifying and analyzing fault categories is as follows:

[0015] Obtain the real-time online fault monitoring results of the target power equipment within the time threshold. The real-time online fault monitoring results include normal and abnormal. If the real-time online fault monitoring results of the target power equipment are abnormal, obtain the initial insulation basic information of the target power equipment within the time threshold.

[0016] Obtain the actual dielectric loss tangent value and the actual insulation resistance of the target power equipment within the time threshold. Set the value obtained by subtracting the dielectric loss tangent value from the actual dielectric loss tangent value and the value obtained by subtracting the actual insulation resistance from the insulation resistance as the dielectric loss deviation value and the insulation resistance change value respectively, and perform discriminant processing on the dielectric loss deviation value and the insulation resistance change value to obtain an insulation fault signal or a non-insulation fault signal.

[0017] Preferably, the process of defect feature recognition and analysis is as follows: Obtain the historical surface feature image set of the insulating material in the target power equipment within the time threshold, preprocess the historical surface feature image set, and construct a surface feature recognition model based on the preprocessed historical surface feature image set.

[0018] Obtain the surface feature image corresponding to the insulating material in the target power equipment within the time threshold, input the surface feature image into the surface feature recognition model, and obtain the output result of the surface feature recognition model. If the output result of the surface feature recognition model is a material fault, generate a material fault signal. If the output result of the surface feature recognition model is a mechanical fault, generate a mechanical fault signal.

[0019] Preferably, the process of abnormal calibration analysis of the insulation area is as follows: Obtain the three-dimensional simulation model of the target power equipment based on the space coordinate system, and obtain the defect feature parameters in the surface feature image corresponding to the insulating material in the target power equipment. The defect feature parameters include carbonization and bending of the insulating material, and mark the area corresponding to the defect feature parameters as a red area in the three-dimensional simulation model.

[0020] Preferably, preprocess the surface infrared feature image of the target power equipment, and at the same time obtain the standard surface infrared feature image corresponding to the insulating material in the target power equipment. Compare and analyze the preprocessed surface infrared feature image with the standard surface infrared feature image to obtain the fault thermal points in the preprocessed surface infrared feature image, and obtain the spatial coordinates (Xm, Ym, Zm) of each fault thermal point, where m represents the fault thermal point and m is a natural number greater than zero.

[0021] Compare the spatial coordinates (Xm, Ym, Zm) of each faulty thermal point with the red area. If the spatial coordinates (Xm, Ym, Zm) of the faulty thermal point are within the red area, it is determined that the corresponding faulty thermal point is an insulation fault point. If the spatial coordinates (Xm, Ym, Zm) of the faulty thermal point are outside the red area, it is determined that the corresponding faulty thermal point is a potential insulation fault point, and the spatial coordinates corresponding to the insulation fault point and the potential insulation fault point are obtained.

[0022] The beneficial effects of the present invention are as follows:

[0023] (1) The present invention initially analyzes from the perspective of the internal environment of the target power equipment to determine whether the influence degree of the internal environment factors on the false alarm risk of the target power equipment failure is excessive, so as to manage the internal environment, which helps to reduce the false alarm risk of the target power equipment. And through the way of information feedback, further analyze whether the target power equipment fails, and at the same time, along with the in-depth process of fault category division, to understand whether the current fault category of the target power equipment is an insulation fault or a non-insulation fault, so as to carry out targeted management according to the fault category;

[0024] (2) Through the way of information progression, defect feature recognition and analysis are carried out on the surface feature image to understand whether the insulation fault is caused by a material fault or a mechanical fault, which helps to make targeted management. At the same time, based on the infrared technology, further risk positioning of the insulation fault is carried out to understand the known and potential risk positions in the insulation material, which helps to improve the operation and maintenance efficiency of the insulation fault of the target power equipment and reduce the insulation fault risk. Description of the Drawings

[0025] The present invention will be further described below with reference to the accompanying drawings;

[0026] Figure 1 is the system flow block diagram of the present invention;

[0027] Figure 2 is the local analysis reference diagram of the present invention. Detailed Embodiments

[0028] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0029] References to "embodiments" in this document mean that the specific features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments;

[0030] Embodiment 1:

[0031] Please refer to Figures 1 to 2 As shown, the present invention is a power equipment fault diagnosis system based on multi-sensor analysis, including an equipment anomaly diagnosis processor, a preliminary false alarm evaluation unit, a progressive fault division unit, an insulation identification unit, a regional positioning unit, and a management warning terminal. The equipment anomaly diagnosis processor is in one-way communication connection with the preliminary false alarm evaluation unit. The preliminary false alarm evaluation unit is in one-way communication connection with both the progressive fault division unit and the management warning terminal. The progressive fault division unit is in one-way communication connection with the insulation identification unit, the regional positioning unit, and the management warning terminal. Both the insulation identification unit and the regional positioning unit are in one-way communication connection with the management warning terminal;

[0032] The equipment anomaly diagnosis processor is used to retrieve the internal environment information of the target power equipment and send the internal environment information to the preliminary false alarm evaluation unit for false alarm potential risk evaluation and analysis to determine whether the false alarm risk of the target power equipment fault alarm is too high. The specific false alarm potential risk evaluation and analysis process is as follows:

[0033] Collect the energized period of the target power equipment, set the energized period of the target power equipment as the time threshold, obtain the internal environment information of the target power equipment within the time threshold. The internal environment information includes electric field strength, electric field frequency, and internal environment temperature, and perform discrimination processing on the internal environment information of the target power equipment;

[0034] Obtain the over-electric field strength period, over-electric field frequency period, and over-temperature period of the target power equipment within the time threshold, obtain the overlapping part of the over-electric field strength period, over-electric field frequency period, and over-temperature period, and set it as the overlapping duration. Then, respectively obtain the field strength inducement duration, field frequency inducement duration, and field temperature inducement duration, obtain the preset field strength interference factor, preset field frequency interference factor, preset field temperature interference factor, and preset superposition interference factor corresponding to the field strength inducement duration, field frequency inducement duration, field temperature inducement duration, and overlapping duration. Then, obtain the sum value of the preset field strength interference factor, preset field frequency interference factor, preset field temperature interference factor, and preset superposition interference factor, and set it as the insulation deviation interference factor, and perform discrimination processing on the insulation deviation interference factor:

[0035] If the insulation deviation interference factor is less than the preset insulation deviation interference factor threshold, a normal interference signal is generated; if the insulation deviation interference factor is greater than or equal to the preset insulation deviation interference factor threshold, a false alarm risk signal is generated. The obtained normal interference signal or false alarm risk signal is sent to the management and warning end. After receiving the normal interference signal or false alarm risk signal, the management and warning end immediately displays the preset warning text corresponding to the normal interference signal or false alarm risk signal, so as to intuitively understand whether the impact of the current internal environment on the target power equipment exceeds the normal range, so as to manage the internal environment, which helps to reduce the false alarm risk of the target power equipment;

[0036] The over-electric field intensity period represents the time period when the electric field intensity exceeds the preset electric field intensity threshold, the over-electric field frequency period represents the time period when the electric field frequency exceeds the preset electric field frequency threshold, and the over-temperature period represents the time period when the internal environment temperature exceeds the preset internal environment temperature threshold;

[0037] The field strength inducement duration represents the value obtained by subtracting the overlapping duration from the over-electric field intensity period, the field frequency inducement duration represents the value obtained by subtracting the overlapping duration from the over-electric field frequency period, and the field temperature inducement duration represents the value obtained by subtracting the overlapping duration from the over-temperature period;

[0038] When a normal interference signal is generated, the progressive fault classification unit is used to collect the initial insulation basic information of the target power equipment and perform fault category classification and analysis on the initial insulation basic information to determine whether the target power equipment has an insulation fault. The specific fault category classification and analysis process is as follows:

[0039] Obtain the real-time online fault monitoring results of the target power equipment within the time threshold. The real-time online fault monitoring results include normal and abnormal. If the real-time online fault monitoring results of the target power equipment are abnormal, obtain the initial insulation basic information of the target power equipment within the time threshold. The initial insulation basic information includes the dielectric loss tangent value and the insulation resistance;

[0040] Normal and abnormal: If the target power equipment does not have a fault alarm, it is normal; if the target power equipment has a fault alarm, it is abnormal;

[0041] Obtain the actual dielectric loss tangent value and the actual insulation resistance of the target power equipment within the time threshold. Set the value obtained by subtracting the dielectric loss tangent value from the actual dielectric loss tangent value and the value obtained by subtracting the actual insulation resistance from the insulation resistance as the dielectric loss deviation value and the insulation resistance change value respectively. Then perform discriminant processing on the dielectric loss deviation value and the insulation resistance change value to obtain the output result of the discriminant processing of the dielectric loss deviation value and the insulation resistance change value. The output result includes insulation fault and non-insulation fault. If the output result is an insulation fault, generate an insulation fault signal. If the output result is a non-insulation fault, generate a non-insulation fault signal. Send the insulation fault signal or the non-insulation fault signal to the management warning end. After receiving the insulation fault signal or the non-insulation fault signal, the management warning end immediately displays the preset warning text corresponding to the insulation fault signal or the non-insulation fault signal, so as to intuitively understand whether the current fault category of the target power equipment is an insulation fault or a non-insulation fault, and thus conduct targeted management based on the fault category;

[0042] An insulation fault means that the dielectric loss deviation value is greater than or equal to the preset dielectric loss deviation value threshold, or the insulation resistance change value is greater than or equal to the preset insulation resistance change value threshold;

[0043] A non-insulation fault means that the dielectric loss deviation value is less than the preset dielectric loss deviation value threshold, and the insulation resistance change value is less than the preset insulation resistance change value threshold.

[0044] Embodiment 2:

[0045] When an insulation fault signal is generated, the insulation identification unit is used to collect the surface feature images corresponding to the insulation materials in the target power equipment for defect feature identification and analysis, so as to understand whether the insulation fault is caused by material or mechanical faults, which helps to make targeted management. The specific defect feature identification and analysis process is as follows:

[0046] Obtain the historical surface feature image set of the insulation materials in the target power equipment within the time threshold, preprocess the historical surface feature image set, and the preprocessing includes cleaning, enhancement, etc. Then construct a surface feature recognition model based on the preprocessed historical surface feature image set;

[0047] In the embodiment of the present invention, the historical surface feature image set includes carbonized surface feature images of insulation materials, surface feature images of circuit bending damage, standard surface feature images, etc.;

[0048] Obtain the surface feature image corresponding to the insulating material in the target power equipment within the time threshold, input the surface feature image into the surface feature recognition model, and obtain the output result of the surface feature recognition model. The output result includes material failure and mechanical failure. If the output result of the surface feature recognition model is a material failure, a material failure signal is generated. If the output result of the surface feature recognition model is a mechanical failure, a mechanical failure signal is generated. Send the obtained material failure signal or mechanical failure signal to the management warning end. After receiving the material failure signal or mechanical failure signal, the management warning end immediately displays the preset warning text corresponding to the material failure signal or mechanical failure signal, so as to make reasonable management of the insulating material according to the feedback information, such as replacing the material or bending and restoring, etc., and at the same time helps to reasonably select the insulating material;

[0049] Material failures include carbonization and chemical corrosion of insulating materials; mechanical failures include wear of insulating materials and damage to bent lines;

[0050] When an insulation failure signal is generated, the area positioning unit is used to collect the surface infrared feature image and perform insulation area abnormal calibration analysis on the surface infrared feature image, so as to accurately locate the insulation failure position on the insulating material in the target power equipment, thereby helping to improve the operation and maintenance efficiency of the insulation failure of the target power equipment. The specific insulation area abnormal calibration analysis process is as follows:

[0051] Taking the lower left endpoint of the bottom surface of the target power equipment as the origin, starting from the origin, and using the long side of the bottom surface as the ray to make the X-axis, and at the same time starting from the origin and using the short side of the bottom surface to make the Y-axis, and making an upward ray from the origin as the Z-axis to establish a space coordinate system;

[0052] In the embodiment of the present invention, the bottom surface of the target power equipment is rectangular;

[0053] Based on the space coordinate system, obtain the three-dimensional simulation model of the target power equipment, and obtain the defect feature parameters in the surface feature image corresponding to the insulating material in the target power equipment. The defect feature parameters include carbonization and bending of the insulating material, etc. Mark the area corresponding to the defect feature parameters as a red area in the three-dimensional simulation model;

[0054] The surface infrared characteristic image of the target power equipment is preprocessed. The preprocessing includes cleaning, enhancement, etc. At the same time, the standard surface infrared characteristic image corresponding to the insulating material in the target power equipment is obtained. The preprocessed surface infrared characteristic image is compared and analyzed with the standard surface infrared characteristic image to obtain the fault thermal points in the preprocessed surface infrared characteristic image, and the spatial coordinates (Xm, Ym, Zm) of each fault thermal point are obtained, where m represents the fault thermal point, and m is a natural number greater than zero. For example, when m = 1, it represents the spatial coordinates (X1, Y1, Z1) of the first fault thermal point; when m = 2, it represents the spatial coordinates (X2, Y2, Z2) of the second fault thermal point, and so on.

[0055] The spatial coordinates (Xm, Ym, Zm) of each fault thermal point are compared and analyzed with the red area. If the spatial coordinates (Xm, Ym, Zm) of the fault thermal point are within the red area, it is determined that the corresponding fault thermal point is an insulation fault point; if the spatial coordinates (Xm, Ym, Zm) of the fault thermal point are outside the red area, it is determined that the corresponding fault thermal point is a potential insulation fault point. Then, the spatial coordinates corresponding to the insulation fault point and the potential insulation fault point are sent to the management and warning terminal. After receiving the spatial coordinates corresponding to the insulation fault point and the potential insulation fault point, the management and warning terminal immediately displays the spatial coordinates corresponding to the insulation fault point and the potential insulation fault point, so as to accurately locate the insulation fault position on the insulating material of the target power equipment, which helps to improve the operation and maintenance efficiency of the insulation fault of the target power equipment.

[0056] The fault thermal point means that the temperature value difference between the corresponding points of the surface infrared characteristic image and the standard surface infrared characteristic image exceeds the preset threshold.

[0057] In summary, the present invention initially analyzes from the perspective of the internal environment of the target power equipment to determine whether the influence degree of the internal environment factors on the false alarm risk of the target power equipment failure is excessive, so as to manage the internal environment, which helps to reduce the false alarm risk of the target power equipment. And through the way of information feedback, it further analyzes whether the target power equipment fails. At the same time, along with the in-depth process of fault category division, it is to understand whether the current fault category of the target power equipment is an insulation fault or a non-insulation fault, so as to carry out targeted management according to the fault category. And through the way of information progression, it conducts defect feature recognition and analysis on the surface feature image to understand whether the insulation fault is caused by a material fault or a mechanical fault, which helps to make targeted management. At the same time, based on the infrared technology, it further locates the risk of the insulation fault to understand the known and potential risk positions in the insulating material, which helps to improve the operation and maintenance efficiency of the insulation fault of the target power equipment and reduce the insulation fault risk.

[0058] The setting of the threshold value is for the convenience of comparison. Regarding the size of the threshold value, it depends on the amount of sample data and the base quantity set by those skilled in the art for each group of sample data; as long as the proportional relationship between the parameter and the quantized value is not affected.

[0059] The size of the coefficient is a specific value obtained by quantizing each parameter for the convenience of subsequent comparison. Regarding the size of the coefficient, it depends on the amount of sample data and the corresponding operation coefficient initially set by those skilled in the art for each group of sample data; as long as the proportional relationship between the parameter and the quantized value is not affected.

[0060] The above is only the preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered within the protection scope of the present invention.

Claims

1. A power equipment fault diagnosis system based on multi-sensor analysis, characterized in that, It includes a device anomaly diagnosis processor, a preliminary false alarm evaluation unit, a progressive fault classification unit, an insulation identification unit, a regional positioning unit, and a management warning terminal; The device anomaly diagnosis processor is used to retrieve the internal environment information of the target power device and send the internal environment information to the preliminary false alarm evaluation unit for false alarm potential risk evaluation and analysis to obtain a normal interference signal or a false alarm risk signal. When a normal interference signal is generated, the progressive fault classification unit is used to perform fault category classification and analysis on the collected initial insulation basic information to obtain an insulation fault signal or a non-insulation fault signal; When an insulation fault signal is generated, the insulation identification unit is used to perform defect feature identification and analysis on the collected surface feature image to obtain a material fault signal or a mechanical fault signal, and the regional positioning unit is used to perform insulation region anomaly calibration analysis on the collected surface infrared feature image to obtain the spatial coordinates corresponding to the insulation fault point and the potential insulation fault point.

2. The power equipment fault diagnosis system based on multi-sensor analysis according to claim 1, wherein The process of false alarm potential risk evaluation and analysis is as follows: The energization period of the target power device is collected, and the energization period of the target power device is set as the time threshold. The internal environment information of the target power device within the time threshold is obtained. The internal environment information includes electric field strength, electric field frequency, and internal environment temperature. The internal environment information of the target power device is discriminated and processed, and then the over-electric field strength period, over-electric field frequency period, and over-temperature period of the target power device within the time threshold are obtained. The overlapping part of the over-electric field strength period, over-electric field frequency period, and over-temperature period is obtained and set as the overlapping duration; The field strength inducement duration, field frequency inducement duration, and field temperature inducement duration are obtained respectively. The preset field strength interference factor, preset field frequency interference factor, preset field temperature interference factor, and preset superposition interference factor corresponding to the field strength inducement duration, field frequency inducement duration, field temperature inducement duration, and overlapping duration are obtained. Then, the sum value of the preset field strength interference factor, preset field frequency interference factor, preset field temperature interference factor, and preset superposition interference factor is obtained and set as the insulation deviation interference factor. The insulation deviation interference factor is discriminated and processed to obtain a normal interference signal or a false alarm risk signal.

3. The power equipment fault diagnosis system based on multi-sensor analysis according to claim 2, characterized in that, The over-electric field strength period represents the time period when the electric field strength exceeds the preset electric field strength threshold. The over-electric field frequency period represents the time period when the electric field frequency exceeds the preset electric field frequency threshold. The over-temperature period represents the time period when the internal environment temperature exceeds the preset internal environment temperature threshold.

4. The power equipment fault diagnosis system based on multi-sensor analysis according to claim 2, wherein The field strength inducement duration represents the value obtained by subtracting the overlapping duration from the over-electric field strength period. The field frequency inducement duration represents the value obtained by subtracting the overlapping duration from the over-electric field frequency period. The field temperature inducement duration represents the value obtained by subtracting the overlapping duration from the over-temperature period.

5. The power equipment fault diagnosis system based on multi-sensor analysis according to claim 1, characterized in that The process of fault category classification and analysis is as follows: The real-time online fault monitoring result of the target power device within the time threshold is obtained. The real-time online fault monitoring result includes normal and abnormal. If the real-time online fault monitoring result of the target power device is abnormal, the initial insulation basic information of the target power device within the time threshold is obtained; The actual dielectric loss tangent value and the actual insulation resistance of the target power equipment within the time threshold are obtained. The values obtained by subtracting the dielectric loss tangent value from the actual dielectric loss tangent value and subtracting the actual insulation resistance from the insulation resistance are respectively set as the dielectric loss deviation value and the insulation resistance change value, and the dielectric loss deviation value and the insulation resistance change value are discriminated to obtain an insulation fault signal or a non-insulation fault signal.

6. The power equipment fault diagnosis system based on multi-sensor analysis according to claim 1, characterized in that The defect feature recognition and analysis process is as follows: The historical surface feature image set of the insulating material in the target power equipment within the time threshold is obtained, the historical surface feature image set is preprocessed, and a surface feature recognition model is constructed based on the preprocessed historical surface feature image set; The surface feature image corresponding to the insulating material in the target power equipment within the time threshold is obtained, the surface feature image is input into the surface feature recognition model, and the output result of the surface feature recognition model is obtained. If the output result of the surface feature recognition model is a material fault, a material fault signal is generated. If the output result of the surface feature recognition model is a mechanical fault, a mechanical fault signal is generated.

7. The power equipment fault diagnosis system based on multi-sensor analysis according to claim 1, characterized in that, The abnormal calibration analysis process of the insulation area is as follows: Based on the space coordinate system, a three-dimensional simulation model of the target power equipment is obtained. The defect feature parameters in the surface feature image corresponding to the insulating material in the target power equipment are obtained. The defect feature parameters include carbonization and bending of the insulating material, and the area corresponding to the defect feature parameters is marked as a red area in the three-dimensional simulation model.

8. The power equipment fault diagnosis system based on multi-sensor analysis according to claim 7, characterized in that, The surface infrared feature image of the target power equipment is obtained and preprocessed. At the same time, the standard surface infrared feature image corresponding to the insulating material in the target power equipment is obtained. The preprocessed surface infrared feature image is compared and analyzed with the standard surface infrared feature image to obtain the fault thermal points in the preprocessed surface infrared feature image. The spatial coordinates (Xm, Ym, Zm) of each fault thermal point are obtained, where m represents the fault thermal point and m is a natural number greater than zero; The spatial coordinates (Xm, Ym, Zm) of each fault thermal point are compared and analyzed with the red area. If the spatial coordinates (Xm, Ym, Zm) of the fault thermal point are within the red area, the corresponding fault thermal point is determined to be an insulation fault point. If the spatial coordinates (Xm, Ym, Zm) of the fault thermal point are outside the red area, the corresponding fault thermal point is determined to be a potential insulation fault point, and the spatial coordinates corresponding to the insulation fault point and the potential insulation fault point are obtained.

Images