Power equipment defect detection method and system based on multi-modal AI large model

Abstract

The invention relates to the field of defect detection, in particular to an electrical equipment defect detection method and system based on a multi-modal AI large model. The method comprises the following steps: acquiring electrical variable information and insulator information, and determining the position of an abnormal insulator according to the electrical variable information and the insulator information; obtaining an insulator image and working information of the transformer substation, and judging whether the abnormality of the insulator is generated by the working information or not according to the insulator position and the insulator image; if the abnormality of the insulator is not generated by the working information, acquiring environment information, and determining surface abnormality information of the insulator according to the environment information and the insulator image; and determining surface crack information according to the surface abnormal information and the electrical variable information. According to the method, the electrical variable information and the image of the insulator and the working information of the substation are acquired and analyzed, so that a comprehensive and accurate reference basis is provided for determining the crack information of the surface of the insulator, and the accuracy of determining the surface defect of the insulator is remarkably improved.

Description

Technical Field

[0001] The present application relates to the field of defect detection, and in particular to a method and system for defect detection of power equipment based on a multimodal AI large model. Background Art

[0002] With the rapid development of the economy and society, the demand for electricity in daily production and life is increasing day by day. In the power conversion process of the substation, insulators are needed to isolate the power equipment from other objects to ensure safety. In order to prevent cracks on the surface of the insulator from affecting its normal work, it is necessary to conduct timely detection of the electrical data on the surface of the insulator to determine whether there are cracks on the surface of the insulator.

[0003] However, since traditional electrical variable measurement technology mainly determines insulator defects by analyzing data obtained by detection equipment, the accuracy of this detection method will be affected by the detection accuracy of the detection equipment, making it difficult for the detection equipment to detect defects in the insulators of the substation, resulting in increased leakage current and partial discharge in the substation, affecting the working efficiency of the substation. Summary of the invention

[0004] The present application provides a method and system for detecting defects in power equipment based on a multimodal AI large model to solve the above-mentioned problems.

[0005] In a first aspect, the present application provides a method for detecting defects in electric power equipment based on a multimodal AI large model, wherein the method for detecting defects in electric power equipment based on a multimodal AI large model is applied to a device for detecting defects in electric power equipment based on a multimodal AI large model, wherein the device for detecting defects in electric power equipment based on a multimodal AI large model comprises an insulator monitoring device, an electric variable monitoring device, a substation management and control center, an environmental monitoring device, and a server, wherein the method for detecting defects in electric power equipment based on a multimodal AI large model is applied to the server, comprising: The electrical variable information collected by the electrical variable monitoring device and the insulator information of the substation management and control center are obtained, and the position of the insulator where the abnormality occurs is determined based on the electrical variable information and the insulator information; the insulator image of the insulator monitoring device and the working information of the substation management and control center are obtained, and whether the insulator abnormality is caused by the working information is determined based on the insulator position and the insulator image; if the insulator abnormality is not caused by the working information, the environmental information collected by the environmental monitoring device is obtained, and the surface abnormality information of the insulator is determined based on the environmental information and the insulator image; the surface crack information is determined based on the surface abnormality information and the electrical variable information.

[0006] Through this solution, the position of the insulator with abnormality can be accurately located by comprehensively analyzing the electrical variable information and insulator information. Combining the insulator image with the substation working information, it is intelligently determined whether the abnormality is related to the working operation, and effectively distinguishes between faults caused by human factors and natural factors. Once it is confirmed that the abnormality is caused by the working information, the environmental information and the insulator image are further integrated to carefully identify the abnormal features of the insulator surface. Combining the surface abnormality information with the electrical variable information, the surface crack information of the insulator is accurately determined, which provides a detailed and reliable basis for the maintenance and repair of power equipment, greatly improving the detection efficiency and accuracy.

[0007] Optionally, determining the surface crack information according to the surface abnormality information and the electrical variable information includes: determining the dirt type, dirt location and dirt adhesion degree of the surface dirt according to the surface abnormality information; judging whether dirt and cracks exist on the insulator surface at the same time according to the electrical variable information and the insulator image; if dirt and cracks exist on the insulator surface at the same time, determining the surface crack length according to the dirt type, the dirt location, the dirt adhesion degree and the electrical variable information; determining the surface crack width according to the electrical variable information and the surface crack length; determining the surface crack number according to the insulator image and the electrical variable information, and determining the surface crack length, the surface crack width and the surface crack number as the surface crack information.

[0008] Optionally, the electrical variable information includes the electric field strength at the insulator position where the abnormality occurs, the working information includes the working current, and the determining the surface crack length according to the dirt type, the dirt position, the dirt adhesion degree and the working information includes: determining the surface crack depth and the coverage angle of the surface crack by the dirt according to the insulator image; determining the conductivity of the dirt according to the dirt type, the dirt adhesion degree and the environmental information; determining the dirt electric field strength under the influence of the dirt according to the working current, the conductivity, the surface crack depth, the coverage angle and the dirt adhesion degree, and calculating according to the following formula: ; in, represents the dirt electric field strength, Indicates the preset adjustment constant, represents the operating current, represents the surface crack depth, Indicates the preset correction factor, represents the conductivity, Indicates the degree of dirt adhesion, represents the coverage angle; The crack electric field strength under the influence of the surface crack is determined according to the electric field strength and the dirt electric field strength; and the surface crack length is determined according to the crack electric field strength, the working current and the insulator position.

[0009] Optionally, the insulator information includes a dielectric constant of the insulator, the working information includes a working voltage, and the determining the surface crack width according to the electrical variable information and the surface crack length includes: determining the working power of the substation according to the working voltage and the working current; determining the crack charge distribution area of ​​the surface crack according to the dielectric constant, the crack electric field strength, the dirt electric field strength and the working power, and calculating according to the following formula: ; in, represents the charge distribution area, represents the operating current, represents the crack electric field strength, represents the dirt electric field strength, represents the dielectric constant, represents the operating power; The surface crack width is determined according to the charge distribution area and the surface crack length.

[0010] Optionally, the electrical variable information includes potential information, and determining the number of surface cracks according to the insulator image and the electrical variable information includes: determining the crack conductivity and dirt conductivity at the insulator position according to the electrical variable information; determining the crack potential of the surface cracks and the dirt potential of the surface dirt according to the potential information; determining the potential difference at the insulator position according to the crack potential and the dirt potential; determining the ambient humidity and ambient temperature at the insulator position according to the environmental information; determining the potential detection area according to the insulator image and the insulator position; obtaining a potential change threshold, and determining the overlapping area of ​​surface cracks and surface dirt according to the potential difference, the crack conductivity, the potential change threshold, the dirt conductivity, the potential detection area, the ambient temperature and the ambient humidity, and calculating according to the following formula: ; in, represents the overlapping area, represents the potential detection area, represents the preset indicator function, represents the crack conductivity, represents the dirt conductivity, represents the ambient temperature, represents the ambient humidity, represents the potential change threshold, Represents a preset spatial function; The number of surface cracks is determined according to the overlapping area, the insulator image and the electric field strength.

[0011] Optionally, the electrical variable information includes a leakage current flowing through the surface of the insulator, and determining the surface abnormality information of the insulator according to the environmental information and the insulator image includes: obtaining an ideal resistance value of the insulator, determining a resistance change value of the insulator according to the degree of dirt adhesion, the ideal resistance value and the environmental humidity, and calculating according to the following formula: ; in, represents the resistance change value, represents the preset dirt weight, Indicates the degree of dirt adhesion, Indicates the preset humidity weight, represents the ambient humidity, represents the ideal resistance value; Judging whether cracks appear on the surface of the insulator according to the resistance change value, the number of insulators and the leakage current; If the crack appears on the surface of the insulator, the crack is included in the surface abnormality information.

[0012] Optionally, judging whether the abnormality of the insulator is caused by the working information according to the insulator position and the insulator image includes: determining the roughness of the insulator surface based on the insulator position and the insulator image; determining the leakage current passing through the insulator according to the working voltage, the roughness and the ambient humidity, and calculating according to the following formula: ; in, represents the leakage current, represents the operating voltage, Represents the preset roughness weight, Indicates the degree of roughness, represents the base impedance, Indicates the preset humidity weight, Indicates the ambient humidity; The normal operating current range of the insulator is obtained, and according to the leakage current and the normal operating current range, it is determined whether the abnormality of the insulator is caused by the operating information.

[0013] Optionally, the insulator information includes the distribution position of the insulator in the substation, the working information includes the average working current of the substation, and after judging whether the abnormality of the insulator is caused by the working information according to the insulator position and the insulator image, the method further includes: if the abnormality of the insulator is caused by the working information, obtaining the actual working current of the substation, determining the abnormal current position according to the normal working current range, the average working current, the actual working current, the current and the distribution position, and calculating according to the following formula: ; in, Indicates the abnormal current position, represents the distribution position, Represents the actual working current, represents the average operating current, Indicates the preset current change coefficient, Indicates the normal operating current range; A first electric field strength at the abnormal current position is obtained, and abnormal operation information is determined according to the first electric field strength and the number of insulators.

[0014] Optionally, determining the crack potential of the surface crack and the fouling potential of the surface fouling according to the potential information includes: determining the comprehensive potential when the surface crack and the surface fouling coexist according to the potential information; determining the crack potential according to the length of the surface crack, the comprehensive potential and the crack electric field strength, and calculating according to the following formula: ; in, represents the crack potential, represents the integrated potential, Indicates the preset potential adjustment constant, represents the crack electric field strength, represents the surface crack length; The fouling potential is determined based on the crack potential and the comprehensive potential.

[0015] In the second aspect, the present application provides an electric power equipment defect detection system based on a multimodal AI large model, the system comprising: an abnormal position determination module, used to obtain electric variable information collected by an electric variable monitoring device and insulator information of a substation management and control center, and determine the insulator position where the abnormality occurs based on the electric variable information and the insulator information; an abnormality judgment module, used to obtain an insulator image of an insulator monitoring device and working information of a substation management and control center, and determine whether the insulator abnormality is caused by the working information based on the insulator position and the insulator image; an abnormal information determination module, used to obtain environmental information collected by an environmental monitoring device if the insulator abnormality is not caused by the working information, and determine the surface abnormality information of the insulator based on the environmental information and the insulator image; a crack information determination module, used to determine the surface crack information based on the surface abnormality information and the electric variable information. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0017] Figure 1 A schematic diagram of an application scenario provided for an embodiment of the present application; Figure 2 A flowchart of a method for detecting defects in electric power equipment based on a multimodal AI large model provided in one embodiment of the present application; Figure 3 A structural schematic diagram of a power equipment defect detection system based on a multimodal AI large model provided in one embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0019] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0020] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0021] Since traditional electrical variable measurement technology mainly determines insulator defects by analyzing data obtained by detection equipment, the accuracy of this detection method will be affected by the detection accuracy of the detection equipment, making it difficult for the detection equipment to detect defects in the insulators of the substation, resulting in increased leakage current and partial discharge in the substation, affecting the working efficiency of the substation.

[0022] Based on this, the present application provides a method and system for detecting defects in power equipment based on a multimodal AI large model, which accurately locates the position of abnormal insulators by comprehensively analyzing electrical variable information and insulator information. Combining the insulator image with the substation working information, it intelligently determines whether the anomaly is related to the working operation, and effectively distinguishes between faults caused by human factors and natural factors. Further integrating environmental information with insulator images, carefully identifying abnormal surface features of insulators, combining surface abnormality information with electrical variable information, and accurately determining insulator surface crack information, provides a detailed and reliable basis for the maintenance and repair of power equipment, greatly improving detection efficiency and accuracy.

[0023] Figure 1 A schematic diagram of an application scenario provided for the present application. When performing defect detection on insulators in a substation, the method provided by the present application is applied to obtain and analyze electrical variable information, insulator information, insulator images, working information and environmental information, so as to determine the surface crack information of the insulator.

[0024] Specifically, the method provided in the present application is applied to a server in a power equipment defect detection device based on a multimodal AI large model. This server can be any server. The server interacts with insulator monitoring equipment, electrical variable monitoring equipment, substation management and control center, environmental monitoring equipment and equipment of substation management personnel. The substation management and control center can be a center for managing and controlling the operation of the substation. The server determines the surface crack information of the insulator by acquiring and analyzing the insulator image of the insulator monitoring equipment, the electrical variable information of the electrical variable monitoring equipment, the insulator information and working information of the substation management and control center, and the environmental information of the environmental monitoring equipment. By collecting the electrical variable information of the insulator during the working process, and combining the image and environmental information for multi-faceted consideration, the abnormality on the surface of the insulator is determined, and the crack information on the surface of the insulator is determined when the abnormality exists, while providing an accurate reference basis for the determination of the crack information on the surface of the insulator, and significantly improving the inspection efficiency of the staff and the work efficiency of the substation. The specific implementation method can refer to the following embodiments.

[0025] Figure 2 This is a flowchart of a method for detecting defects in power equipment based on a multimodal AI large model provided in one embodiment of the present application. The method of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the method includes: S201. Obtain the electric variable information collected by the electric variable monitoring equipment and the insulator information of the substation management and control center, and determine the position of the insulator where the abnormality occurs according to the electric variable information and the insulator information.

[0026] The electric variable monitoring device may be a device that monitors the electric variables on the surface of the insulator when the insulator of the substation is working. The electric variable information may be information about changes in the electric field and circuit characteristics on the surface of the insulator when the insulator of the substation is working. The electric variable information may include inductance and capacitance. The insulator information may be information about the insulator when the insulator of the substation is working. The insulator information may include the number of insulators and the strength of the insulator. The insulator position may be the position corresponding to the insulator whose working state is abnormal when the insulator of the substation is working.

[0027] Specifically, when the insulators of the substation are working, if the changes in the electric field and circuit characteristics on the surface of the insulators are abnormal, it will affect the normal working state of the insulators, thereby causing the normal insulation function of the insulators to be lost, and further affecting the overall normal operation of the substation. Therefore, the electrical variable information on the surface of the insulator is extracted from the electrical variable monitoring device. The inductance and capacitance on the surface of the insulator are extracted from the electrical variable information. The insulator information is obtained from the substation management and control center, and the number of insulators and the strength of the insulators are extracted from the insulator information. The inductance and capacitance obtained in the above steps are analyzed using impedance detection technology to obtain the position of the insulator that generates abnormal impedance. Based on the position of the insulator that generates the abnormality obtained in the above steps, the number of insulators and the strength of the insulator are analyzed using local discharge detection technology to obtain the position corresponding to the insulator that generates the abnormal signal. The position corresponding to the insulator that generates the abnormal signal is determined as the position of the insulator that generates the abnormality.

[0028] S202, obtaining an insulator image of an insulator monitoring device and working information of a substation management and control center, and determining whether the insulator abnormality is caused by the working information based on the insulator position and the insulator image.

[0029] The insulator monitoring device may be a device that monitors the surface of the insulator when the insulator of the substation is working. The working information may be information about the electric field and circuit characteristics corresponding to the operation of the substation itself when the insulator of the substation is working. The insulator image may be an image corresponding to the monitoring of the surface of the insulator when the insulator of the substation is working.

[0030] Specifically, after the insulator of the substation is in an abnormal state, there are many reasons for the abnormal state of the insulator, such as cracks or stains on the insulator itself, and abnormal working state of the insulator due to abnormal working information of the substation. In order to judge and determine the cause of the abnormal working state of the insulator, it is necessary to obtain the insulator image from the insulator monitoring device, obtain the real-time electric field and circuit characteristics of the substation from the substation management and control center, and determine the real-time electric field and circuit characteristics of the substation as the working information of the substation. Extract the working resistance, working inductance and working capacitance from the working information, use the mathematical analysis algorithm to analyze the working resistance, working inductance, working capacitance and insulator position obtained in the above steps, and obtain the working state index of the substation. Obtain the normal working state index range of the substation from the substation management and control center, match the working state index with the normal working state index range, and if the working state index is within the normal working state index range, it is determined that the abnormality of the insulator is not caused by the working information. If the working status index is not within the normal working status range, it is necessary to use an image analysis algorithm to analyze the insulator image to obtain an image analysis result. If the image analysis result does not contain information about abnormalities on the insulator surface, it is determined that the insulator abnormality is caused by working information.

[0031] S203. If the abnormality of the insulator is not caused by the working information, obtain the environmental information collected by the environmental monitoring equipment, and determine the surface abnormality information of the insulator according to the environmental information and the insulator image.

[0032] The environmental monitoring equipment may be a device that monitors the environment within a certain range around the substation in real time when the insulators of the substation are working.

[0033] The environmental information may be information about the environment within a certain range around the substation when the insulator of the substation is working. The environmental information may be environmental humidity and environmental temperature.

[0034] The surface abnormality information may be information corresponding to an abnormality on the surface of the insulator when the insulator of the substation is working. The surface abnormality information may include the location of surface cracks and the location of surface dirt.

[0035] Specifically, when judging the cause of abnormality of the insulator of the substation, if the abnormality of the insulator is not caused by the working information of the substation itself. At this time, the environmental information collected by the environmental monitoring equipment is obtained, and the environmental humidity and ambient temperature are extracted from the environmental information. The insulator image is analyzed using a feature recognition algorithm to obtain the insulator density. The inductance and capacitance obtained in the above steps are analyzed using impedance imaging technology to obtain the preliminary crack position and preliminary dirt position. Based on the preliminary crack position, preliminary dirt position, ambient temperature and ambient humidity, the insulator density is analyzed using ultrasonic density measurement technology to obtain the dielectric loss on the insulator surface. According to the dielectric loss on the insulator surface, the surface crack position and surface dirt position of the insulator are determined. The surface crack position and surface dirt position are included in the surface abnormality information.

[0036] S204: Determine surface crack information based on surface abnormality information and electrical variable information.

[0037] The surface crack information may be information corresponding to cracks appearing on the surface of the insulator when the insulator of the substation is working.

[0038] Specifically, when an abnormality occurs on the surface of the insulator, because cracks and dirt appear on the surface of the insulator at the same time, the cracks and dirt will have different degrees of impact on the electrical variable information. In order to accurately determine the crack information on the surface of the insulator, it is necessary to extract the surface crack position and the surface dirt position from the surface abnormality information. The voltage on the surface of the insulator is extracted from the electrical variable information. The surface crack position, surface dirt position, voltage and inductance obtained in the above steps are analyzed using current monitoring and partial discharge technology to obtain the number and width of cracks on the surface of the insulator. The number and width of cracks are included in the surface crack information.

[0039] Through this solution, the position of the insulator with abnormality can be accurately located by comprehensively analyzing the electrical variable information and insulator information. Combining the insulator image with the substation working information, it is intelligently determined whether the abnormality is related to the working operation, and effectively distinguishes between faults caused by human factors and natural factors. Once it is confirmed that the abnormality is caused by the working information, the environmental information and the insulator image are further integrated to carefully identify the abnormal features of the insulator surface. Combining the surface abnormality information with the electrical variable information, the surface crack information of the insulator is accurately determined, which provides a detailed and reliable basis for the maintenance and repair of power equipment, greatly improving the detection efficiency and accuracy.

[0040] In some embodiments, the dirt type, dirt location and dirt adhesion degree of the surface dirt are determined based on surface abnormality information; whether dirt and cracks exist on the insulator surface at the same time is determined based on the electrical variable information and the insulator image; if dirt and cracks exist on the insulator surface at the same time, the surface crack length is determined based on the dirt type, dirt location, dirt adhesion degree and electrical variable information; the surface crack width is determined based on the electrical variable information and the surface crack length; the surface crack number is determined based on the insulator image and the electrical variable information, and the surface crack length, surface crack width and surface crack number are determined as surface crack information.

[0041] The degree of dirt adhesion may be the degree or strength of dirt adhesion on the surface of the insulator when the insulator of the substation is working. The type of dirt may be the type of dirt that appears on the surface of the insulator when the insulator of the substation is working. The location of the dirt may be the location of the dirt on the surface of the insulator when the insulator of the substation is working. The length of the surface crack may be the length of the crack that exists on the surface of the insulator when the insulator of the substation is working. The width of the surface crack may be the width of the crack that exists on the surface of the insulator when the insulator of the substation is working. The number of surface cracks may be the number of cracks that exist on the surface of the insulator when the insulator of the substation is working.

[0042] Specifically, when an abnormality occurs on the surface of the insulator, the cracks and dirt on the surface of the insulator will have a certain impact on the normal insulation state of the insulator. In order to ensure that the insulator can be quickly repaired when an abnormality occurs, it is necessary to extract the dirt type, dirt location and dirt adhesion degree of the surface dirt from the surface abnormality information. In the case where it is determined that there is dirt on the surface of the insulator, the electric field strength corresponding to the dirt location is extracted from the electrical variable information. The dirt type, dirt location and dirt adhesion degree obtained in the above steps are analyzed using a mathematical analysis algorithm to obtain the dirt electric field strength under the influence of dirt. The electric field strength corresponding to the dirt location is compared with the dirt electric field strength. If the electric field strength corresponding to the dirt location is inconsistent with the dirt electric field strength, the feature recognition algorithm is used to analyze the insulator image to obtain an image result. If the image recognition result contains cracks other than dirt, it is determined that dirt and cracks exist on the insulator surface at the same time. The dirt type, dirt location, resistance and dirt adhesion degree are analyzed using a mathematical analysis algorithm to obtain the surface crack length. The inductance and surface crack length are analyzed using a mathematical analysis algorithm to obtain the surface crack width. The number of insulators, inductance and resistance are analyzed using a mathematical analysis algorithm to obtain the number of surface cracks. The surface crack length, surface crack width and surface crack number are determined as surface crack information.

[0043] Through this solution, the type, location and degree of dirt on the insulator surface can be accurately identified by analyzing surface abnormality information, providing basic data for subsequent processing. Combined with electrical variable information and insulator images, it is comprehensively judged whether there are both dirt and cracks on the insulator surface to ensure the comprehensiveness of the diagnosis. Once the coexistence is found, the dirt characteristics and electrical variable information are used to accurately determine the crack length to effectively avoid dirt interference. Based on the electrical variable information and the measured crack length, the crack width is further inferred to improve the accuracy of crack assessment. Combined with the image and electrical variable information, the number of cracks is comprehensively counted to ensure that there are no omissions. By integrating multi-source information, the length, width and number of surface cracks are determined to form complete surface crack information, providing a scientific basis for the maintenance and management of insulators.

[0044] In some embodiments, the surface crack depth and the coverage angle of the surface crack by the dirt are determined according to the insulator image; the conductivity of the dirt is determined according to the type of dirt, the degree of dirt adhesion and the environmental information; the dirt electric field strength under the influence of the dirt is determined according to the working current, conductivity, surface crack depth, coverage angle and degree of dirt adhesion according to formula (1); the crack electric field strength under the influence of the surface crack is determined according to the electric field strength and the dirt electric field strength; the surface crack length is determined according to the crack electric field strength, the working current and the insulator position: (1); in, represents the dirt electric field strength, Indicates the preset adjustment constant, Indicates the working current, Indicates the surface crack depth, Indicates the preset correction factor, represents the conductivity, Indicates the degree of dirt adhesion. Indicates the coverage angle; The surface crack depth may be the depth corresponding to the crack on the surface of the insulator when the insulator of the substation is working. The preset adjustment constant may be a value for normalizing the calculation result of the dirt electric field strength. The preset correction coefficient may be a value for correcting the conductivity of the dirt, the degree of dirt adhesion and the coverage angle when calculating the dirt electric field strength. The coverage angle may be the geometric angle of the contact or overlapping area between the surface crack and the surface dirt when there are cracks and dirt on the surface of the insulator of the substation, that is, the angle between the crack and the dirt, reflecting their relative position and interaction mode. The conductivity may be a value indicating the current conduction capacity of the crack and dirt on the surface of the insulator when the insulator of the substation is working. The electric field strength may be the force exerted on a unit charge in the electric field when the insulator of the substation is working and there is dirt on the surface of the insulator. The crack electric field strength may be the force exerted on a unit charge in the electric field when the insulator of the substation is working and there is dirt on the surface of the insulator.

[0045] Specifically, when determining the crack length on the surface of the insulator, since there are both dirt and cracks on the surface of the insulator, the dirt on the surface of the insulator will affect the determination of the crack length. Therefore, it is necessary to obtain the acidity and alkalinity corresponding to all possible dirt types from the dirt information website, integrate the acidity and alkalinity corresponding to all possible dirt types, and obtain a preset dirt type set. Use the edge detection algorithm to analyze the insulator image to obtain the crack profile. Use laser scanning technology to analyze the crack profile to obtain three-dimensional data of the surface crack. Use deep learning technology to analyze the three-dimensional data to obtain the surface crack depth. Use the edge detection algorithm to analyze the insulator image to obtain the surface dirt profile. Use the least squares method to fit the surface dirt profile and the crack profile to obtain the angle between the surface dirt profile and the crack profile. The angle between the surface dirt and the crack profile is determined as the coverage angle. The dirt type obtained in the above steps is matched with the preset dirt type set to obtain the acidity and alkalinity corresponding to the dirt type. The acidity and alkalinity, dirt adhesion and environmental humidity corresponding to the dirt type are analyzed using a mathematical analysis algorithm to obtain the conductivity of the dirt. The historical working current, historical conductivity, historical surface crack depth, historical coverage angle, historical dirt electric field strength and historical dirt adhesion degree are obtained from the substation management and control center. The historical working current, historical conductivity, historical surface crack depth, historical coverage angle, historical dirt electric field strength and historical dirt adhesion degree are analyzed using a linear regression analysis method to obtain a preset correction coefficient set preset adjustment constant. The working current, conductivity, surface crack depth, coverage angle and dirt adhesion degree are analyzed using a mathematical analysis method, and the dirt electric field strength is calculated according to formula (1) to obtain the dirt electric field strength.

[0046] In formula (1), It is the basic term of electric field strength, which is mainly determined by the working current and crack depth; The conductivity of dirt , Adhesion coefficient and coverage angle They will work together to increase the concentration of local electric field strength, thereby affecting the electric field distribution of the insulator. The crack depth is the size of the crack in the vertical direction of the insulator surface (usually refers to the incidence depth of the crack). The monitoring equipment can obtain the crack depth through image processing and analysis, which is usually relatively easy because it can be estimated through methods such as grayscale changes and edge detection of the image. The crack length is the distribution range of the crack in the horizontal direction of the insulator surface. The crack length is affected by many factors, including viewing angle, three-dimensional morphology of the crack, and obstruction of dirt, so it is difficult to measure directly and accurately.

[0047] Dirt on the surface of the insulator not only affects the electric field distribution, but also may cover up or change the actual morphology of the crack. The presence of dirt may cause changes in the electric field concentration area, which directly affects the electrical characteristics of the crack: dirt may cause the electric field near the crack to be enhanced, so even if the crack length is short, the local intensity of the electric field may be amplified due to the conductivity of the dirt. By analyzing the electric field strength at the dirt, the size of the crack (including the length) can be indirectly inferred, because the change in electric field strength is related to the location, depth and length of the crack. Subtract the electric field strength from the dirt electric field strength to obtain the crack electric field strength under crack conditions. The insulator conductivity is extracted from the insulator information. The historical crack electric field strength, historical working current, historical insulator conductivity and historical surface crack length are obtained from the substation management and control center. The historical crack electric field strength, historical working current, historical insulator conductivity and historical surface crack length are analyzed using the linear regression analysis method to obtain the adjustment constant. Based on the insulator position obtained in the above steps, the adjustment constant, crack electric field strength, working current and insulator conductivity are analyzed using mathematical analysis methods. The surface crack length is calculated according to formula (2): (2); in, represents the surface crack length, Indicates the working current, represents the insulator conductivity, represents the crack electric field intensity, Represents the adjustment constant.

[0048] Through this solution, the depth of surface cracks and the coverage angle of dirt on the cracks are accurately measured by analyzing the insulator image. The conductivity of the dirt is scientifically determined by combining the dirt type, degree of adhesion and environmental information, which is an important parameter for evaluating the impact of dirt on the electric field. Then, the formula is used to comprehensively consider the working current, conductivity, crack depth, coverage angle and degree of dirt adhesion to accurately calculate the electric field strength under the influence of dirt, accurately reflecting the interference of dirt on the electric field distribution. Furthermore, by comparing the electric field strength with the electric field strength of dirt, the electric field strength under the influence of cracks is obtained. Finally, combined with the working current and the position of the insulator, the length of the surface crack is accurately calculated, providing a reliable basis for the status assessment and maintenance of the insulator.

[0049] In some embodiments, the working power of the substation is determined according to the working voltage and the working current; the crack charge distribution area of ​​the surface crack is determined according to the dielectric constant, the crack electric field strength, the dirt electric field strength and the working power, and the surface crack width is determined according to the charge distribution area and the surface crack length: (3); in, represents the charge distribution area, Indicates the working current, represents the crack electric field intensity, represents the dirt electric field strength, represents the dielectric constant, Indicates working power; The dielectric constant can be the ability of the insulator to store electrical energy in the electric field when the insulator of the substation is working. The working power can be the rate at which the substation itself converts or transfers electrical energy when the insulator of the substation is working.

[0050] Specifically, when surface cracks and surface dirt exist at the same time, the width of the surface cracks will have a certain impact on the electric field strength on the insulator surface, thereby affecting the distribution area of ​​the charge on the insulator surface. Therefore, it is necessary to multiply the working voltage by the working current to obtain the working power of the substation. The dielectric constant, crack electric field strength, dirt electric field strength and working power are analyzed using a mathematical analysis algorithm, and the crack charge distribution area of ​​the surface crack is calculated according to formula (3).

[0051] In formula (3), the numerator is: power Represents the total power supply of the system at a certain voltage and current, and is the key starting point for calculating the surface crack charge distribution. The multiplier factor of 2 indicates that a certain trade-off has been made in the energy calculation, usually for correction considerations. The introduction of the factor 2 will increase the charge distribution area , which can better match the charge accumulation of cracks in reality. Molecular part It provides the basis for the total energy in the substation and is the "energy source" in the calculation of the crack charge distribution area. The greater the power, the more electrical energy the substation supplies, so under certain conditions, the distribution area of ​​the crack charge will be larger.

[0052] Denominator: , the denominator includes the dielectric constant and the square of the electric field strength. They jointly affect the charge distribution characteristics of the crack area. The square of the electric field strength It represents the square of the equivalent electric field strength under the combined effect of cracks and dirt. The introduction of the square term means that the effect of electric field strength on the charge distribution area is nonlinear. As the electric field strength increases, the charge distribution area will decrease significantly. This is because a higher electric field strength means stronger charge aggregation, resulting in a more concentrated charge distribution area. It is the "limiting factor" when calculating the charge distribution area. The stronger the effect of cracks and dirt on the electric field, the larger the value of the denominator, resulting in a smaller charge distribution area. The smaller it is, the more it reflects the improvement of charge concentration due to the strong electric field.

[0053] Through this solution, the working power of the substation is accurately calculated according to the working voltage and working current of the substation, providing basic data for evaluating the crack charge distribution. The charge distribution area of ​​the surface crack is scientifically calculated through a formula using the dielectric constant, crack electric field strength, dirt electric field strength and working power. A variety of factors are comprehensively considered, including the impact of electric field strength on charge distribution and the reflection of the dielectric constant on the working environment of the insulator, to ensure the accuracy of the calculation results. Combining the charge distribution area and the known crack length, the width of the surface crack is accurately calculated, which provides an important reference for the further maintenance of the insulator and significantly improves the accuracy and efficiency of crack evaluation.

[0054] In some embodiments, the crack conductivity and dirt conductivity at the insulator position are determined according to the electrical variable information; the crack potential of the surface crack and the dirt potential of the surface dirt are determined according to the potential information; the potential difference at the insulator position is determined according to the crack potential and the dirt potential; the ambient humidity and ambient temperature at the insulator position are determined according to the environmental information; the potential detection area is determined according to the insulator image and the insulator position; the potential change threshold is obtained, and the overlapping area of ​​the surface cracks and the surface dirt is determined according to the potential difference, the crack conductivity, the potential change threshold, the dirt conductivity, the potential detection area, the ambient temperature and the ambient humidity according to formula (4); the number of surface cracks is determined according to the overlapping area, the insulator image and the electric field strength: (4); in, represents the overlapping area, represents the potential detection area, represents the preset indicator function, represents the crack conductivity, represents the dirt conductivity, Indicates the ambient temperature, Indicates the ambient humidity. represents the potential change threshold, Represents a preset spatial function; The potential information may be information about changes in the potential on the surface of the insulator due to the presence of cracks and dirt on the surface of the insulator when the insulator of the substation is working. The crack potential may be the potential energy possessed by a unit charge at a crack on the surface of the insulator when the insulator of the substation is working. The dirt potential may be the potential energy possessed by a unit charge at a dirt on the surface of the insulator when the insulator of the substation is working. The potential detection area may be the area determined when the potential of an area where the insulator state is abnormal is detected when the insulator of the substation is working. The potential change threshold may be the numerical range corresponding to the normal state of the potential on the surface of the insulator.

[0055] Specifically, when cracks and dirt exist on the surface of the insulator at the same time and overlap, the potential generated by the two will affect the determination of the number of cracks on the surface of the insulator. Therefore, the crack conductivity and dirt conductivity of the insulator position where the abnormality occurs are extracted from the electrical variable information. The crack potential of the surface crack and the dirt potential of the surface dirt are extracted from the potential information. The crack potential is subtracted from the dirt potential to obtain the potential subtraction result, and the absolute value of the potential subtraction result is removed to obtain the potential difference. The ambient humidity and ambient temperature are extracted from the environmental information. Based on the insulator position determined in the above steps, the insulator image is analyzed using image analysis technology to obtain the potential detection area. The potential change threshold is obtained from the substation management and control center. The potential difference, crack conductivity, potential change threshold, dirt conductivity, potential detection area, ambient temperature and ambient humidity are calculated using mathematical analysis technology according to formula (4) to determine the overlapping area of ​​surface cracks and surface dirt.

[0056] In formula (4), the integral domain is is the entire area to be tested (i.e. the surface of the insulator). In this area, for each point Perform analysis, calculate the potential difference gradient at the point and determine whether it belongs to the boundary area between crack and dirt. Indicator function The function is to "mark" whether the potential difference gradient exceeds a certain threshold . When the potential difference gradient is greater than the threshold, this indicates that the potential difference at this point changes greatly, that is, there may be a boundary area between the crack and the dirt. The value of this function is 1 in the boundary area and 0 in the non-boundary area. The boundary area is identified by the change in the potential difference (gradient). When the potential difference changes dramatically, it means that there are obvious electrical characteristics at the boundary between the crack and the dirt at this point, so there may be a physical interface. Based on the overlapping area obtained in the above steps, the finite element analysis technology is used to analyze the electric field strength obtained in the above steps to obtain the electric field distribution position on the surface of the insulator under the simulated environment. The electric field distribution position under the simulated environment is compared with the position of the crack. If the electric field distribution position under the simulated environment is consistent with the position of the crack, the feature extraction algorithm is used to analyze the area corresponding to the crack position in the insulator image to obtain the number of surface cracks.

[0057] Through this solution, the conductivity of cracks and dirt at the insulator position is determined based on the electrical variable information, and the potential information is used to clarify the potential of cracks and dirt, and then the potential difference at the insulator position is calculated. Combined with meteorological data, the ambient humidity and temperature are accurately obtained to provide key parameters for subsequent calculations. The potential detection area is determined by the insulator image, and the potential change threshold is set. The crack conductivity, dirt conductivity, potential detection area, ambient temperature and humidity are comprehensively considered, and the overlapping area of ​​cracks and dirt is accurately calculated using the formula. Combined with the overlapping area, insulator image and electric field strength, the number of surface cracks can be accurately determined, providing a scientific basis for the condition monitoring and maintenance of insulators, and effectively improving the accuracy and efficiency of the assessment.

[0058] In some embodiments, the ideal resistance value of the insulator is obtained, and the resistance change value of the insulator is determined according to the degree of dirt adhesion, the ideal resistance value and the ambient humidity according to formula (5); whether cracks appear on the surface of the insulator is determined according to the resistance change value, the number of insulators and the leakage current; if cracks appear on the surface of the insulator, the cracks are included in the surface abnormality information: (5); in, Indicates the resistance change value, represents the preset dirt weight, Indicates the degree of dirt adhesion. Indicates the preset humidity weight, Indicates the ambient humidity. Indicates the ideal resistance value; The ideal resistance value may be the resistance value of the insulator when the insulator of the substation is working, when there is no dirt on the surface of the insulator and the humidity on the surface of the insulator is zero. The leakage current may be the value corresponding to the current flowing through the surface of the insulator when the insulator of the substation is working. The resistance change value may be the difference between the resistance value corresponding to cracks and dirt on the surface of the insulator and the ideal resistance value. The preset dirt weight may be the influence of the degree of dirt adhesion on the calculation result of the resistance change value when calculating the resistance change value, and the preset dirt weight may be a specific value when calculating the formula. The preset humidity weight may be the influence of the ambient humidity on the calculation result of the resistance change value when calculating the resistance change value, and the preset humidity weight may be a specific value when calculating the formula.

[0059] Specifically, when dirt appears on the surface of the insulator, the resistance change value of the insulator surface will be affected by other abnormal factors besides dirt. In order to accurately determine other abnormal factors besides dirt, it is necessary to obtain the insulator type information of the same type as the current insulator from the insulator information website, and extract the ideal resistance value corresponding to the current insulator type from the insulator type information. The historical dirt adhesion degree, historical ideal resistance value, historical resistance change value and historical ambient humidity are obtained from the substation management and control center, and the historical dirt adhesion degree, historical ideal resistance value, historical resistance change value and historical ambient humidity are analyzed using a linear regression algorithm to obtain a preset dirt weight and a preset humidity weight. A mathematical analysis algorithm is used to calculate the resistance change value of the insulator according to the dirt adhesion degree, the preset dirt weight, the preset humidity weight, the ideal resistance value and the ambient humidity according to formula (5).

[0060] In formula (5), Reflects the The impact of relevant factors (such as certain inputs, demands, etc.), among which is with The proportionality coefficient associated with The increase of How will it increase. The size determines right If Larger, then right The impact will be more significant. is with The relevant proportionality factor, Another possible influence This term indicates that How does the change affect , the degree of its influence is determined by Decide.

[0061] Based on the number of insulators obtained in the above steps, when there is dirt on the insulator surface, a mathematical analysis algorithm is used to analyze the degree of dirt adhesion, ambient humidity and ambient temperature to obtain the leakage current and resistance change value when only dirt exists on the insulator surface. The leakage current when only dirt exists and the resistance change value when only dirt exists are compared with the resistance change value and the leakage current, respectively. If the leakage current when only dirt exists is less than the leakage current, and the resistance change value when only dirt exists is greater than the resistance change value, it is determined that there are cracks on the insulator surface, and the cracks are included in the surface abnormality information.

[0062] Through this solution, by obtaining the ideal resistance value of the insulator, combining the degree of dirt adhesion and environmental humidity, and using a formula to comprehensively consider the dirt weight and humidity weight, the resistance change value of the insulator is accurately calculated, which effectively reflects the comprehensive impact of dirt and environmental humidity on the resistance performance of the insulator. By comparing the resistance change value, the number of insulators and the current, it is scientifically judged whether cracks appear on the surface of the insulator, which improves the accuracy of crack detection. Once a crack is found, it is immediately included in the surface abnormality information, which provides a key basis for subsequent processing and significantly improves the efficiency and accuracy of insulator status monitoring.

[0063] In some embodiments, based on the insulator position and the insulator image, the roughness of the insulator surface is determined; based on the working voltage, the roughness and the ambient humidity, the leakage current passing through the insulator is calculated according to formula (6); the normal working current range of the insulator is obtained, and based on the leakage current and the normal working current range, it is determined whether the abnormality of the insulator is caused by the working information: (6); in, represents the leakage current, Indicates the operating voltage, Represents the preset roughness weight, Indicates the degree of roughness. represents the base impedance, Indicates the preset humidity weight, Indicates ambient humidity; The roughness may be a numerical value indicating the smoothness of the surface of the insulator when the insulator of the substation is working. The preset roughness weight may be a numerical value that affects the calculation result of the leakage current when calculating the leakage current of the insulator. The normal operating current range may be a range of working currents corresponding to the substation when the substation is operating in the field. The basic impedance may be a numerical value indicating the basic degree of resistance of the insulator to the working current when the insulator of the substation is working.

[0064] Specifically, when the insulator of the substation is working, the leakage current on the surface of the insulator will affect the determination of the abnormal cause of the insulator surface to a certain extent. Therefore, when judging the abnormal cause of the insulator surface, it is necessary to analyze the insulator image based on the insulator position using image analysis technology to obtain the roughness of the insulator surface. The historical working voltage, historical roughness, historical leakage current and historical ambient humidity are obtained from the substation management and control center, and the historical working voltage, historical roughness, historical leakage current and historical ambient humidity are analyzed using a linear regression algorithm to obtain a preset roughness weight. The working voltage, preset roughness weight, preset humidity weight, roughness and ambient humidity are calculated using a mathematical analysis algorithm according to formula (6) to determine the leakage current passing through the insulator.

[0065] Among them, in formula (6), the numerator is: , when the surface roughness When increasing, The value of also increases, which causes the numerator to increase, making the current The higher the roughness, the greater the current, because the rough surface is more likely to produce leakage current channels. Denominator: , the larger the base impedance, the smaller the current, which is consistent with the limiting effect of resistance on current. Humidity influence factor :This is the influence coefficient of humidity on insulation performance. In the above items, the higher the humidity, the higher the risk of insulator leakage current. The increase will Reduce, equivalent impedance Reduce, eventually making the leakage current Increase. This reflects the effect of voltage and surface roughness. An increase in voltage or roughness will increase the operating current. It reflects the combined effect of base impedance and humidity. The higher the humidity, the smaller the denominator and the larger the current; the higher the base impedance, the smaller the current.

[0066] The leakage current is matched with the normal working current range. If the leakage current is within the normal working current range, it is determined that the abnormality of the insulator is not caused by the working information. If the leakage current is not within the normal working current range, it is determined that the abnormality of the insulator is caused by the working information.

[0067] Through this solution, the roughness of the surface of the insulator is accurately evaluated through the insulator image based on the insulator position. Combining the working voltage, roughness and ambient humidity, the formula comprehensively considers the roughness weight and humidity weight to accurately calculate the leakage current passing through the insulator, which effectively reflects the influence of the insulator surface state and environmental conditions on the leakage current. By comparing the leakage current with the normal working current range of the insulator, it is scientifically determined whether the abnormality of the insulator is caused by the working information, which provides strong support for fault troubleshooting.

[0068] In some embodiments, if the insulator abnormality is caused by working information, the actual working current of the substation is obtained, and the current abnormal position is determined according to the normal working current range, the average working current, the actual working current, the current and the distribution position according to formula (7); the first electric field strength at the current abnormal position is obtained, and the abnormal working information is determined according to the first electric field strength and the number of insulators: (7); in, Indicates the abnormal current location. Indicates the distribution location, Indicates the actual working current, Represents the average operating current, Indicates the preset current change coefficient, Indicates the normal operating current range; The average working current may be a value corresponding to the median of a numerical range formed between a maximum value and a minimum value of a normal working current range of the substation when the substation is working. The preset current variation coefficient may be a value indicating an influence on a calculation result of a current abnormality position when calculating a current abnormality position.

[0069] Specifically, when the insulator of the substation is working, the abnormality on the surface of the insulator may not be caused by the abnormality of the insulator itself. At this time, if the abnormality of the insulator is caused by working information, the actual working current is obtained from the substation management and control center. The mathematical analysis algorithm is used to analyze the normal working current range, average working current, preset current change coefficient, actual working current, current and distribution position, and the current abnormal position is determined according to formula (7).

[0070] In formula (7), : This is the location The "normal current" estimate on is used to reflect how the current is distributed at different locations under normal circumstances. Usually, the current is not distributed completely uniformly throughout the system, and this term is expressed by This shows the law of current changing with position. For example, if ( >0), indicating that the current varies with position Increase; if ( <0), the current decreases with position. This term provides a position-dependent current reference value. : This is the location The current deviation on the With normal current The greater the difference, the greater the degree to which the current deviates from the normal distribution. When The current deviation exceeds the allowable range, that is, the actual current is far from the expected current. Considered an abnormal position.

[0071] The first electric field strength at the current abnormal position is obtained from the electric variable information. The first electric field strength and the number of insulators are analyzed using finite element analysis technology to obtain the abnormal working current and abnormal working voltage of the substation. The abnormal working current and abnormal working voltage are determined as abnormal working information.

[0072] Through this solution, after confirming that the insulator abnormality is caused by working information, the actual working current of the substation is obtained, and the current abnormality position is accurately calculated using the formula in combination with the normal working current range, average working current, actual working current and its distribution position, effectively locating the source of the fault. By obtaining the first electric field strength at the current abnormality position and comprehensively considering the number of insulators, the abnormal working information is scientifically determined, providing a strong basis for fault troubleshooting and repair. It not only improves the accuracy of fault location, but also effectively shortens the troubleshooting time and improves the operating efficiency and safety of the substation.

[0073] In some embodiments, the comprehensive potential when the surface crack and the surface dirt coexist is determined based on the potential information; the crack potential is determined based on the surface crack length, the comprehensive potential and the crack electric field strength according to formula (8); the dirt potential is determined based on the crack potential and the comprehensive potential: (8); in, represents the crack potential, represents the integrated potential, Indicates the preset potential adjustment constant, represents the crack electric field intensity, Indicates the surface crack length; The comprehensive potential may be a potential corresponding to the presence of dirt and cracks on the surface of the insulator when the insulator of the substation is working. The preset potential adjustment constant may be a value that affects the calculation result of the crack potential when calculating the crack potential.

[0074] Specifically, the comprehensive potential is extracted from the potential information. The historical surface crack length, historical comprehensive potential, historical crack potential and historical crack electric field strength are obtained from the substation management and control center. The historical surface crack length, historical comprehensive potential, historical crack potential and historical crack electric field strength are analyzed using a linear regression analysis algorithm to obtain a preset potential adjustment constant. The surface crack length, comprehensive potential and crack electric field strength are calculated using a mathematical analysis algorithm according to formula (8) to determine the crack potential for analysis and obtain the crack potential. The comprehensive potential is subtracted from the crack potential to obtain the fouling potential.

[0075] Through this solution, the comprehensive potential when surface cracks and surface dirt coexist is accurately calculated based on the potential information, providing key data for subsequent calculations. By using the surface crack length, comprehensive potential and crack electric field strength, the preset potential adjustment constant is comprehensively considered through the formula to accurately derive the crack potential, which effectively reflects the specific impact of the crack on the potential distribution. By comparing the crack potential with the comprehensive potential, the dirt potential is scientifically determined, providing an important basis for comprehensively evaluating the surface state of the insulator. It not only improves the accuracy of potential calculation, but also helps to deeply understand the comprehensive impact of cracks and dirt on insulator performance.

[0076] Figure 3 A schematic diagram of a structure of a power equipment defect detection system based on a multimodal AI large model provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the power equipment defect detection method and system 300 based on the multimodal AI large model of this embodiment includes: an abnormal position determination module 301, an abnormality judgment module 302, an abnormality information determination module 303 and a crack information determination module 304.

[0077] The abnormal position determination module 301 is used to obtain the electric variable information collected by the electric variable monitoring device and the insulator information of the substation management and control center, and determine the insulator position where the abnormality occurs based on the electric variable information and the insulator information; the abnormality judgment module 302 is used to obtain the insulator image of the insulator monitoring device and the working information of the substation management and control center, and determine whether the insulator abnormality is caused by the working information based on the insulator position and the insulator image; the abnormal information determination module 303 is used to obtain the environmental information collected by the environmental monitoring device if the insulator abnormality is not caused by the working information, and determine the surface abnormality information of the insulator based on the environmental information and the insulator image; the crack information determination module 304 is used to determine the surface crack information based on the surface abnormality information and the electric variable information.

[0078] Optionally, the crack information determination module 304 is specifically used to: determine the dirt type, dirt location and dirt adhesion degree of the surface dirt based on the surface abnormality information; determine whether dirt and cracks exist on the insulator surface at the same time based on the electrical variable information and the insulator image; if dirt and cracks exist on the insulator surface at the same time, determine the surface crack length based on the dirt type, dirt location, dirt adhesion degree and electrical variable information; determine the surface crack width based on the electrical variable information and the surface crack length; determine the surface crack number based on the insulator image and the electrical variable information, and determine the surface crack length, surface crack width and surface crack number as surface crack information.

[0079] Optionally, the crack information determination module 304 is specifically used to: determine the surface crack depth and the coverage angle of the surface crack by the dirt according to the insulator image; determine the conductivity of the dirt according to the dirt type, the degree of dirt adhesion and the environmental information; determine the dirt electric field strength under the influence of the dirt according to the working current, conductivity, surface crack depth, coverage angle and degree of dirt adhesion, and calculate according to the following formula: ;in, represents the dirt electric field strength, Indicates the preset adjustment constant, Indicates the working current, Indicates the surface crack depth, Indicates the preset correction factor, represents the conductivity, Indicates the degree of dirt adhesion. Indicates the coverage angle; determines the crack electric field strength under the influence of surface cracks based on the electric field strength and the dirt electric field strength; determines the surface crack length based on the crack electric field strength, working current and insulator position.

[0080] Optionally, the crack information determination module 304 is specifically used to: determine the working power of the substation according to the working voltage and the working current; determine the crack charge distribution area of ​​the surface crack according to the dielectric constant, the crack electric field strength, the dirt electric field strength and the working power, and calculate according to the following formula: ; in, represents the charge distribution area, Indicates the working current, represents the crack electric field intensity, represents the dirt electric field strength, represents the dielectric constant, Indicates working power; determines surface crack width based on charge distribution area and surface crack length.

[0081] Optionally, the crack information determination module 304 is specifically used to: determine the crack conductivity and dirt conductivity at the insulator position according to the electrical variable information; determine the crack potential of the surface crack and the dirt potential of the surface dirt according to the potential information; determine the potential difference at the insulator position according to the crack potential and the dirt potential; determine the ambient humidity and ambient temperature at the insulator position according to the environmental information; determine the potential detection area according to the insulator image and the insulator position; obtain the potential change threshold, and determine the overlapping area of ​​the surface crack and the surface dirt according to the potential difference, the crack conductivity, the potential change threshold, the dirt conductivity, the potential detection area, the ambient temperature and the ambient humidity, and calculate according to the following formula: ;in, represents the overlapping area, represents the potential detection area, represents the preset indicator function, represents the crack conductivity, represents the dirt conductivity, Indicates the ambient temperature, Indicates the ambient temperature, represents the potential change threshold, Represents a preset spatial function; determines the number of surface cracks based on the overlapping area, insulator image and electric field strength.

[0082] Optionally, the abnormal information determination module 303 is specifically used to obtain the ideal resistance value of the insulator, determine the resistance change value of the insulator according to the degree of dirt adhesion, the ideal resistance value and the ambient humidity, and calculate according to the following formula: ;in, Indicates the resistance change value, represents the preset dirt weight, Indicates the degree of dirt adhesion. Indicates the preset humidity weight, Indicates the ambient humidity. Indicates the ideal resistance value; judge whether cracks appear on the surface of the insulator based on the resistance change value, the number of insulators and the leakage current; if cracks appear, the cracks are included in the surface abnormality information.

[0083] Optionally, the abnormality judgment module 302 is specifically used to: determine the roughness of the insulator surface based on the insulator position and the insulator image; determine the leakage current passing through the insulator based on the operating voltage, the roughness and the ambient humidity, and calculate it according to the following formula: ;in, represents the leakage current, Indicates the operating voltage, Represents the preset roughness weight, Indicates the degree of roughness. represents the base impedance, Indicates the preset humidity weight, Indicates the ambient humidity; obtains the normal working current range of the insulator, and determines whether the insulator abnormality is caused by the working information based on the leakage current and the normal working current range.

[0084] Optionally, the power equipment defect detection system 300 based on the multimodal AI large model also includes an abnormal working information determination module 305, which is specifically used to: if the abnormality of the insulator is caused by working information, obtain the actual working current of the substation, and determine the current abnormal position according to the normal working current range, average working current, actual working current, current and distribution position, and calculate according to the following formula: ;in, Indicates the abnormal current location. Indicates the distribution location, Indicates the actual working current, Represents the average operating current, Indicates the preset current change coefficient, Indicates the normal working current range; obtains the first electric field strength at the current abnormal position, and determines the abnormal working information based on the first electric field strength and the number of insulators.

[0085] Optionally, the crack information determination module 304 is specifically used to: determine the comprehensive potential when the surface crack and the surface dirt exist at the same time according to the potential information; determine the crack potential according to the surface crack length, the comprehensive potential and the crack electric field strength, and calculate according to the following formula: ;in, represents the crack potential, represents the integrated potential, Indicates the preset potential adjustment constant, represents the crack electric field intensity, Indicates the length of surface cracks; determines the fouling potential based on the crack potential and comprehensive potential.

[0086] The system of this embodiment can be used to execute the method of any of the above embodiments. The implementation principles and technical effects are similar and will not be described in detail here.

Claims

1. A method for detecting defects in power equipment based on a multimodal AI large model, characterized in that: The electric power equipment defect detection method based on the multimodal AI large model is applied to the electric power equipment defect detection device based on the multimodal AI large model, the electric power equipment defect detection device based on the multimodal AI large model includes an insulator monitoring device, an electric variable monitoring device, a substation management and control center, an environmental monitoring device and a server, and the electric power equipment defect detection method based on the multimodal AI large model is applied to the server, including: Acquire the electric variable information collected by the electric variable monitoring device and the insulator information of the substation management and control center, and determine the insulator position where the abnormality occurs according to the electric variable information and the insulator information; Acquire an insulator image of the insulator monitoring device and working information of the substation management and control center, and determine whether the abnormality of the insulator is caused by the working information according to the insulator position and the insulator image; If the abnormality of the insulator is not caused by the working information, obtaining the environmental information collected by the environmental monitoring device, and determining the surface abnormality information of the insulator according to the environmental information and the insulator image; Surface crack information is determined based on the surface abnormality information and the electrical variable information.

2. The method according to claim 1, characterized in that The determining of surface crack information according to the surface abnormality information and the electrical variable information includes: Determining the type of dirt, the location of dirt, and the degree of dirt adhesion of the surface dirt according to the surface abnormality information; Judging whether there are both dirt and cracks on the surface of the insulator according to the electrical variable information and the insulator image; If dirt and cracks exist on the surface of the insulator at the same time, the length of the surface crack is determined according to the dirt type, the dirt position, the degree of dirt adhesion and the electrical variable information; Determining the surface crack width according to the electrical variable information and the surface crack length; The number of surface cracks is determined according to the insulator image and the electrical variable information, and the surface crack length, the surface crack width and the number of surface cracks are determined as the surface crack information.

3. The method according to claim 2, characterized in that The electrical variable information includes the electric field strength at the insulator position where the abnormality occurs, the working information includes the working current, and the surface crack length is determined according to the dirt type, the dirt position, the dirt adhesion degree and the working information, including: Determining the depth of the surface crack and the coverage angle of the dirt on the surface crack according to the insulator image; Determining the conductivity of the dirt according to the dirt type, the degree of dirt adhesion and the environmental information; According to the working current, the conductivity, the surface crack depth, the coverage angle and the degree of dirt adhesion, the dirt electric field strength under the influence of dirt is determined and calculated according to the following formula: ； in, represents the dirt electric field strength, Indicates the preset adjustment constant, represents the operating current, represents the surface crack depth, Indicates the preset correction factor, represents the conductivity, Indicates the degree of dirt adhesion, represents the coverage angle; Determining the crack electric field strength under the influence of the surface crack according to the electric field strength and the dirt electric field strength; The surface crack length is determined according to the crack electric field strength, the operating current and the insulator position.

4. The method according to claim 3, characterized in that The insulator information includes the dielectric constant of the insulator, the working information includes the working voltage, and the determining the surface crack width according to the electrical variable information and the surface crack length includes: Determining the operating power of the substation according to the operating voltage and the operating current; According to the dielectric constant, the crack electric field strength, the dirt electric field strength and the working power, the crack charge distribution area of ​​the surface crack is determined and calculated according to the following formula: ； in, represents the charge distribution area, represents the operating current, represents the crack electric field strength, represents the dirt electric field strength, represents the dielectric constant, represents the operating power; The surface crack width is determined according to the charge distribution area and the surface crack length.

5. The method according to claim 4, characterized in that The electrical variable information includes potential information, and determining the number of surface cracks according to the insulator image and the electrical variable information includes: Determining the crack conductivity and dirt conductivity at the insulator position according to the electrical variable information; Determining the crack potential of the surface crack and the fouling potential of the surface fouling according to the potential information; Determining a potential difference at the insulator position according to the crack potential and the dirt potential; Determining the ambient humidity and ambient temperature at the insulator location according to the environmental information; Determining a potential detection area according to the insulator image and the insulator position; Obtain a potential change threshold, and determine the overlapping area of ​​surface cracks and surface dirt according to the potential difference, the crack conductivity, the potential change threshold, the dirt conductivity, the potential detection area, the ambient temperature and the ambient humidity, and calculate according to the following formula: ； in, represents the overlapping area, represents the potential detection area, represents the preset indicator function, represents the crack conductivity, represents the dirt conductivity, represents the ambient temperature, represents the ambient humidity, represents the potential change threshold, Represents a preset spatial function; The number of surface cracks is determined according to the overlapping area, the insulator image and the electric field strength.

6. The method according to claim 5, characterized in that The electrical variable information includes leakage current flowing through the surface of the insulator, and determining the surface abnormality information of the insulator according to the environmental information and the insulator image includes: The ideal resistance value of the insulator is obtained, and the resistance change value of the insulator is determined according to the dirt adhesion degree, the ideal resistance value and the ambient humidity, and is calculated according to the following formula: ； in, represents the resistance change value, represents the preset dirt weight, Indicates the degree of dirt adhesion, Indicates the preset humidity weight, represents the ambient humidity, represents the ideal resistance value; Judging whether cracks appear on the surface of the insulator according to the resistance change value, the number of insulators and the leakage current; If the crack appears on the surface of the insulator, the crack is included in the surface abnormality information.

7. The method according to claim 6, characterized in that The determining, based on the insulator position and the insulator image, whether the abnormality of the insulator is caused by the working information includes: Based on the insulator position and according to the insulator image, determining the roughness of the insulator surface; According to the working voltage, the roughness and the ambient humidity, the leakage current passing through the insulator is determined and calculated according to the following formula: ； in, represents the leakage current, represents the operating voltage, Represents the preset roughness weight, Indicates the degree of roughness, represents the base impedance, Indicates the preset humidity weight, Indicates the ambient humidity; The normal operating current range of the insulator is obtained, and according to the leakage current and the normal operating current range, it is determined whether the abnormality of the insulator is caused by the operating information.

8. The method according to claim 7, characterized in that The insulator information includes the distribution position of the insulator in the substation, the working information includes the average working current of the substation, and after judging whether the abnormality of the insulator is caused by the working information according to the insulator position and the insulator image, the method further includes: If the abnormality of the insulator is caused by the working information, the actual working current of the substation is obtained, and the abnormal current position is determined according to the normal working current range, the average working current, the actual working current, the current and the distribution position, and calculated according to the following formula: ； in, Indicates the abnormal current position, represents the distribution position, Represents the actual working current, represents the average operating current, Indicates the preset current change coefficient, Indicates the normal operating current range; A first electric field strength at the abnormal current position is obtained, and abnormal operation information is determined according to the first electric field strength and the number of insulators.

9. The method according to claim 5, characterized in that Determining the crack potential of the surface crack and the fouling potential of the surface fouling according to the potential information includes: Determining, based on the potential information, a comprehensive potential when the surface cracks and the surface dirt coexist; The crack potential is determined according to the surface crack length, the comprehensive potential and the crack electric field strength, and is calculated according to the following formula: ； in, represents the crack potential, represents the integrated potential, Indicates the preset potential adjustment constant, represents the crack electric field strength, represents the surface crack length; The fouling potential is determined based on the crack potential and the comprehensive potential.

10. A power equipment defect detection system based on a multimodal AI large model, characterized in that: include: The abnormal position determination module is used to obtain the electric variable information collected by the electric variable monitoring equipment and the insulator information of the substation management and control center, and determine the insulator position where the abnormality occurs according to the electric variable information and the insulator information; An abnormality judgment module is used to obtain an insulator image of an insulator monitoring device and working information of a substation management and control center, and judge whether the abnormality of the insulator is caused by the working information according to the insulator position and the insulator image; an abnormality information determination module, configured to obtain environmental information collected by an environmental monitoring device if the abnormality of the insulator is not caused by the working information, and determine surface abnormality information of the insulator based on the environmental information and the insulator image; The crack information determination module is used to determine the surface crack information according to the surface abnormality information and the electrical variable information.

Images