Intelligent Fault Diagnosis and Early Warning Method for Distribution Switchgear

Abstract

The present invention discloses an intelligent fault diagnosis and early warning method for distribution switchgear. The present invention collects historical thermal fault cases and times of the distribution switchgear, integrates them into a historical thermal fault data packet, and conducts in-depth evaluation and classification on it. By performing data analysis and processing on the classified data packet, temperature change parameters within a set time period before each fault occurrence in the corresponding thermal fault classification data packet are extracted and analyzed. According to the analysis results, an early warning evaluation digital model and an early warning evaluation index are constructed to form an early warning set. Combining the early warning trigger mechanism and the early warning update mechanism, it can achieve early identification, accurate early warning, and rapid response to thermal faults of the distribution switchgear, thereby improving the reliability and safety of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical system automation, and particularly to an intelligent fault diagnosis and early warning method for distribution switchgear cabinets. Background Art

[0002] A distribution switchgear cabinet is a power equipment mainly used in the power distribution link of a power system. It usually includes multiple switching devices such as circuit breakers, contactors, relays, etc., as well as corresponding control and protection devices. As a key equipment in the power system, the stability of the distribution switchgear cabinet is directly related to the safe operation of the power system.

[0003] However, the existing intelligent fault diagnosis and early warning methods for distribution switchgear cabinets still have the following deficiencies:

[0004] They mostly rely on early warning with fixed thresholds, lack in-depth analysis of the specific types of faults and parameter changes during the fault development process, making it difficult to achieve accurate early warning and resulting in false alarms and missed alarms;

[0005] At the same time, after a fault occurs, the process of updating early warning parameters and models is not timely and intelligent enough, and it is unable to quickly adjust and update the early warning mechanism according to the latest data.

[0006] Therefore, an intelligent fault diagnosis and early warning method for distribution switchgear cabinets is introduced. Summary of the Invention

[0007] In view of this, the present invention provides an intelligent fault diagnosis and early warning method for distribution switchgear cabinets to solve the problems raised in the above background art.

[0008] The object of the present invention can be achieved through the following technical solutions: An intelligent fault diagnosis and early warning method for distribution switchgear cabinets, including:

[0009] Fault data collection: Collect historical thermal fault cases and times of the distribution switchgear cabinet and integrate them into a historical thermal fault data packet of the distribution switchgear cabinet;

[0010] Fault assessment and classification: Analyze the historical thermal fault data packet of the distribution switchgear cabinet. After the analysis is completed, classify it based on the specific reasons for triggering the thermal fault of the distribution switchgear cabinet to obtain a classified thermal fault classification data packet;

[0011] Data analysis and processing: Based on the classified thermal fault classification data packet, extract the temperature change parameters within a set time period before each fault occurrence in the corresponding thermal fault classification data packet and analyze them;

[0012] Establish an early warning mechanism: Construct an early warning assessment digital model and an early warning assessment index for each fault in the corresponding thermal fault classification data packet according to the analysis results , integrate the early warning evaluation digital models and early warning evaluation indices of each fault in the corresponding thermal fault classification data packet into the early warning set of the corresponding thermal fault classification data packet;

[0013] Intelligent early warning trigger: Set the monitoring time interval of the distribution switchgear, and after reaching the set monitoring time interval, analyze the temperature change parameters of each temperature acquisition point in the distribution switchgear within the set time period. If the analysis result of the temperature change parameters of a certain temperature acquisition point matches the early warning set of the corresponding thermal fault type, directly trigger the early warning signal of the corresponding thermal fault type;

[0014] Early warning trigger execution: After triggering the early warning, perform corresponding operations based on the specific thermal fault type that triggers the early warning;

[0015] Early warning update mechanism: Used to analyze the temperature change parameters when the thermal fault appears again in the distribution switchgear, and update the belonging early warning set based on the analysis result according to the specific type of the thermal fault.

[0016] In some embodiments, extract the temperature change parameters within the set time period before each fault occurs in the corresponding thermal fault classification data packet, and perform analysis, specifically:

[0017] Extract the temperature values within the set time period before each fault occurs in the corresponding thermal fault classification data packet, and obtain the temperature values at each time point within the set time period according to the preset division time interval;

[0018] Based on the obtained temperature values at each time point, determine the window size of the moving average. For each time point and the determined window size, calculate the average value of all numerical points within the window size range before and after this time point. As the time series progresses, the sliding window moves forward, and repeat calculating the average value of all numerical points within the window size range before and after each time point until the entire data set is covered;

[0019] Extract the average values calculated for each group, and thus construct a temperature change line graph within the set time period before each fault occurs in the corresponding thermal fault classification data packet; Plot the numerical points corresponding to the average values at each time point within the line graph, and connect adjacent numerical points to obtain the average line;

[0020] Calculate the slope of each average line and the angle between it and the horizontal line. If the angle between a certain average line and the horizontal line is an acute angle, mark the slope of this average line as a decreasing slope. If the angle between a certain average line and the horizontal line is an obtuse angle, mark the slope of this average line as an increasing slope;

[0021] Sum up all the decreasing slopes and increasing slopes respectively to obtain the total decreasing value and the total increasing value, and perform score conversion on the obtained total decreasing value and total increasing value;

[0022] Based on the specific type of the thermal fault classification data packet, preset the value ranges of the total decline value and the total rise value for each group, and set a scoring value corresponding to each value range of the total decline value and the total rise value for each group. Match the total decline value and the total rise value corresponding to each fault with the corresponding value ranges of each group based on the specific type of the thermal fault classification data packet to obtain the scoring values corresponding to the total decline value and the total rise value of each fault; among them, the scoring value corresponding to the total rise value obtained by matching is a positive number, and the scoring value corresponding to the total decline value obtained by matching is a negative number.

[0023] Accumulate the scoring value corresponding to the total rise value and the scoring value corresponding to the total decline value to obtain the trend change index Ea of each fault within the set time period.

[0024] In some embodiments, extract the temperature change parameters within the set time period before each fault occurs in the corresponding thermal fault classification data packet and perform analysis, and further include:

[0025] Based on the temperature values obtained at each time point, take the average value of the temperature values at each time point as the temperature average value of each fault within the set time period. Based on the specific type of the thermal fault classification data packet, preset the reference thresholds of the temperatures corresponding to each type of thermal fault.

[0026] At the same time, extract the maximum value and the minimum value from the temperature values obtained at each time point, calculate the ratios of the temperature average value, the maximum value, and the minimum value to the corresponding preset reference thresholds respectively, and use the three calculated ratios as the average ratio, the maximum ratio, and the minimum ratio of each fault within the set time period.

[0027] Based on the specific type of the thermal fault classification data packet corresponding to the fault, preset the weight coefficients of the average ratio, the maximum ratio, and the minimum ratio, multiply the calculated average ratio, the maximum ratio, and the minimum ratio by the corresponding preset weight coefficients respectively, and then sum to obtain the trend temperature index Eb of each fault within the set time period.

[0028] In some embodiments, construct a warning evaluation index for each fault in the corresponding thermal fault classification data packet according to the analysis results , specifically:

[0029] Extract the trend change index Ea and the trend temperature index Eb of each fault within the set time period, and based on the specific type of the thermal fault classification data packet corresponding to each fault, preset the reference indices of the trend change index Ea and the trend temperature index Eb.

[0030] According to the formula , perform weighted calculation on the trend change index Ea and the trend temperature index Eb of each fault within the set time period to obtain the warning evaluation index ; where and respectively represent the reference indices of the fault trend change index Ea and the trend temperature index Eb for each fault; and are respectively the influence weight factors of the trend change index Ea and the trend temperature index Eb.

[0031] In some embodiments, according to the analysis results, a warning evaluation digital model for each fault in the corresponding thermal fault classification data packet is constructed, specifically:

[0032] Based on each fault in the calculation of the warning evaluation index during the and , extract 's calculation result as the length value of the rectangle, 's calculation result as the width value of the rectangle, construct a rectangle model based on the obtained length value and width value, and use the constructed rectangle model as the warning evaluation digital model for each fault.

[0033] In some embodiments, the warning evaluation digital model and the warning evaluation index for each fault in the corresponding thermal fault classification data packet are integrated into the warning set of the corresponding thermal fault classification data packet, specifically:

[0034] After calculating the warning evaluation index corresponding to each fault, number them based on the specific type of the thermal fault classification data packet corresponding to each fault, and mark the warning evaluation index of each fault in the corresponding thermal fault classification data packet as , where e = a, b, c or d, corresponding to the mechanical thermal fault data packet, the insulation thermal fault data packet, the overload thermal fault data packet, and the environmental thermal fault data packet respectively; m = 1, 2 or k, and k is the total number of faults in the corresponding thermal fault classification data packet;

[0035] After numbering, integrate the warning evaluation digital model and the warning evaluation index of each fault in the corresponding thermal fault classification data packet into the warning set of the corresponding thermal fault classification data packet.

[0036] In some embodiments, perform corresponding operations based on the specific thermal fault type that triggers the warning, specifically:

[0037] S1: If the specific thermal fault type triggering the warning is a mechanical thermal fault, collect the image information of each group in the corresponding mechanical area of the distribution switchgear cabinet, extract and mark the areas with wear characteristics from the collected image information of each group, take the marked areas as the wear areas in the image information of each group of collections, integrate the wear areas into a wear data packet, draw a circle with the location of the distribution switchgear cabinet at the current time point as the center and a set distance as the radius, screen each maintenance personnel within the circle, and send a location feedback signaling to the mobile terminals of each maintenance personnel. After each maintenance personnel confirms the location feedback signaling, obtain the locations of each maintenance personnel within the circle, thereby obtaining the travel distances of each maintenance personnel from the distribution switchgear cabinet at the current time point. Sort the travel distances of each group from small to large, extract the three maintenance personnel with the top-ranked travel distances, and obtain their working hours. Take the maintenance personnel with the longest working hours as the selected personnel for the current triggered warning, and send the wear data packet and the location of the distribution switchgear cabinet to the mobile terminal of this personnel;

[0038] S2: If the specific thermal fault type triggering the warning is an insulation thermal fault, collect the image information of each group in the corresponding insulation area of the distribution switchgear cabinet, extract and mark the areas with crack and aging characteristics from the collected images of each group, take the marked areas as the crack areas and aging areas in the image information of each group, integrate the crack areas and aging areas into a damage data packet, and similarly to step S1, send the damage data packet and the location of the distribution switchgear cabinet to the mobile terminal of the selected personnel.

[0039] In some embodiments, based on the specific thermal fault type triggering the warning, performing corresponding operations further includes:

[0040] S3: If the specific thermal fault type triggering the warning is an overload thermal fault, generate a load adjustment signaling and send it to the mobile terminal of the management personnel, and the management personnel receive the generated load adjustment signaling and reallocate the load;

[0041] S4: If the specific thermal fault type that triggers the warning is an environmental thermal fault, after adjusting the power of the heat dissipation device, collect the image information of the ventilation port area corresponding to the distribution switchgear cabinet. Using image processing technology, extract and mark the areas with dust characteristics from the collected image information of each ventilation port area. Take the marked areas as the polluted areas of the image information of each ventilation port area, and count the number of pixels in the polluted areas. Based on the resolution of the image, convert to the actual size to obtain the pollution area of the ventilation port area corresponding to the distribution switchgear cabinet at the current time point. Preset the maximum allowable area of the pollution area. Compare the pollution area of the ventilation port area corresponding to the distribution switchgear cabinet at the current time point with the preset maximum allowable area. If it is greater than the preset maximum allowable area, trigger a cleaning signal, select the maintenance personnel with the shortest travel distance from the distribution switchgear cabinet at the current time point as the selected personnel, and send the cleaning signal to the mobile terminal of this person.

[0042] Compared with the prior art, the beneficial effects of the present invention are:

[0043] The present invention collects the historical thermal fault cases and times of the distribution switchgear cabinet, integrates them into a historical thermal fault data packet, and conducts in-depth evaluation and classification on it. Through data analysis and processing of the classified data packet, extract the temperature change parameters within the set time period before each fault occurrence in the corresponding thermal fault classification data packet, and conduct analysis. According to the analysis results, construct a warning evaluation digital model and a warning evaluation index, form a warning set, and combine the warning trigger mechanism and the warning update mechanism to achieve early identification, accurate warning, and rapid response to the thermal faults of the distribution switchgear cabinet, thereby improving the reliability and safety of the power system;

[0044] The present invention sets the monitoring time interval of the distribution switchgear cabinet, and after reaching the set monitoring time interval, analyze the temperature change parameters of each temperature acquisition point of the distribution switchgear cabinet within the set time period. If the analysis result of the temperature change parameters of a certain temperature acquisition point matches the warning set of the corresponding thermal fault type, directly trigger the warning signal of the corresponding thermal fault type, and execute the corresponding steps to achieve rapid response and processing of the warning;

[0045] When the distribution switchgear cabinet has a thermal fault again, the present invention analyzes the temperature change parameters, and updates the warning set to which it belongs based on the specific type of the thermal fault according to the analysis results, improving the degree of intelligence. Description of the Drawings

[0046] In the following description of the exemplary embodiments in conjunction with the drawings, more details, features, and advantages of the present application are disclosed. In the drawings:

[0047] Figure 1 is the flowchart of the present invention;

[0048] Figure 2 This is a schematic diagram of the temperature change line graph in the present invention. Specific embodiments

[0049] Several embodiments of the present application will be described in more detail below with reference to the accompanying drawings so that those skilled in the art can implement the present application. The present application can be embodied in many different forms and purposes and should not be limited to the embodiments described herein. These embodiments are provided to make the present application comprehensive and complete and to fully convey the scope of the present application to those skilled in the art. The embodiments do not limit the present application.

[0050] Unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present application belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the relevant art and / or the context of this specification, and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0051] Please refer to Figure 1 As shown, the intelligent fault diagnosis and early warning method for a distribution switchgear cabinet includes:

[0052] Fault data collection: Collect historical thermal fault cases and times of the distribution switchgear cabinet through database retrieval and on-site records, and integrate them into a historical thermal fault data packet of the distribution switchgear cabinet;

[0053] Fault assessment and classification: Analyze the historical thermal fault data packet of the distribution switchgear cabinet. After the analysis is completed, classify it based on the specific reasons for triggering the thermal fault of the distribution switchgear cabinet to obtain a classified thermal fault classification data packet; among them, the thermal fault classification data packet is mainly divided into a mechanical thermal fault data packet, an insulation thermal fault data packet, an overload thermal fault data packet, and an environmental thermal fault data packet;

[0054] Data analysis and processing: Based on the classified thermal fault classification data packet, extract the temperature change parameters within a set time period before each fault occurrence in the corresponding thermal fault classification data packet and analyze them; set that different thermal fault classification data packets respectively correspond to a temperature acquisition point, and the temperature acquisition points include a mechanical thermal fault acquisition point, an insulation thermal fault acquisition point, an overload thermal fault acquisition point, and an environmental thermal fault acquisition point;

[0055] According to the specific type of the thermal fault classification data packet, install temperature sensors at key positions of the corresponding type to monitor temperature changes, for example:

[0056] Mechanical thermal fault:

[0057] Installation location: breaker contacts, mechanical connection components, areas where mechanical wear or corrosion may occur;

[0058] Insulation thermal fault:

[0059] Installation location: insulator surface, cable joints, near bushings, and other areas where insulation degradation or partial discharge may occur;

[0060] Overload thermal fault:

[0061] Installation location: conductor connection points, busbar connection points, circuit parts where overload may occur;

[0062] Ambient thermal fault:

[0063] Installation location: internal space of switchgear, ventilation openings, or areas greatly affected by ambient temperature;

[0064] Specifically:

[0065] Extract the temperature values within the set time period before each fault occurrence in the corresponding thermal fault classification data packet, and obtain the temperature values at each time point within the set time period according to the preset time interval for division;

[0066] Based on the temperature values at each obtained time point, determine the window size of the moving average, that is, the number of points within the considered time period. For each time point and the determined window size, calculate the average value of all numerical points within the window size range before and after this time point. Based on the advancement of the time series, slide the window forward and repeat the calculation of the average value of all numerical points within the window size range before and after each time point until the entire data set is covered;

[0067] It should be noted that at the beginning and end positions of the time series, due to insufficient data points, the complete moving average cannot be calculated, so the moving average of these points is not calculated;

[0068] Extract the average values calculated for each group, and thus construct a temperature change line graph for the set time period before each fault occurrence in the corresponding thermal fault classification data packet; plot the numerical points corresponding to the average values at each time point within the line graph, and connect the adjacent numerical points to obtain the average line;

[0069] Calculate the slope of each average line and the angle between it and the horizontal line. If the angle between an average line and the horizontal line is an acute angle, mark the slope of this average line as a decreasing slope; if the angle between an average line and the horizontal line is an obtuse angle, mark the slope of this average line as an increasing slope;

[0070] Sum up all the decreasing slopes and increasing slopes respectively to obtain the total decreasing value and total increasing value, and perform the conversion of scoring for the obtained total decreasing value and total increasing value;

[0071] Based on the specific type of the thermal fault classification data packet, preset the value ranges of the total decline value and the total rise value for each group, set a scoring value corresponding to each value range of the total decline value and the total rise value, and match the total decline value and the total rise value corresponding to each fault with the corresponding value ranges of each group based on the specific type of the thermal fault classification data packet to obtain the scoring values corresponding to the total decline value and the total rise value of each fault; among them, the scoring value obtained by matching the total rise value is a positive number, and the scoring value obtained by matching the total decline value is a negative number;

[0072] The larger the total rise value and the total decline value, the larger the corresponding matching scoring value, and when the total decline value is 0, the corresponding matching scoring value is -1;

[0073] The scoring value ranges corresponding to the total rise value and the total decline value are respectively set between 1 and 10, -1 and -10, and the specific settings can be adjusted and optimized according to the actual application situation;

[0074] Accumulate the scoring value corresponding to the total rise value and the scoring value corresponding to the total decline value to obtain the trend change index Ea of each fault within the set time period;

[0075] Further based on the temperature values obtained at each time point, take the average value of the temperature values at each time point as the temperature average value of each fault within the set time period, and preset the reference thresholds of the temperatures corresponding to each type of thermal fault based on the specific type of the thermal fault classification data packet;

[0076] At the same time, extract the highest value and the lowest value from the temperature values obtained at each time point, calculate the ratios of the temperature average value, the highest value, and the lowest value to the corresponding preset reference thresholds respectively, and use the three calculated ratios as the average ratio, the highest ratio, and the lowest ratio of each fault within the set time period;

[0077] Based on the specific type of the thermal fault classification data packet corresponding to the fault, preset the weight coefficients of the average ratio, the highest ratio, and the lowest ratio, multiply the calculated average ratio, the highest ratio, and the lowest ratio by the corresponding preset weight coefficients respectively, and then sum to obtain the trend temperature index Eb of each fault within the set time period;

[0078] Establish an early warning mechanism: construct an early warning evaluation digital model and an early warning evaluation index for each fault in the corresponding thermal fault classification data packet according to the analysis results and the early warning evaluation digital model and the early warning evaluation index for each fault in the corresponding thermal fault classification data packet are integrated into the early warning set corresponding to the thermal fault classification data packet;

[0079] Specifically:

[0080] Extract the trend change index Ea and the trend temperature index Eb of each fault within the set time period, and preset the reference indices of the trend change index Ea and the trend temperature index Eb according to the specific type of the thermal fault classification data packet corresponding to each fault;

[0081] According to the formula , perform weighted calculation on the trend change index Ea and the trend temperature index Eb of each fault within the set time period to obtain the warning evaluation index ; where and respectively represent the reference indices of the trend change index Ea and the trend temperature index Eb of each fault, and are valued according to the specific type of the corresponding thermal fault classification data packet; and are respectively the influence weight factors of the trend change index Ea and the trend temperature index Eb, and the values of the influence weight factors are different according to the specific type of the corresponding thermal fault classification data packet;

[0082] Furthermore, based on the above-mentioned each fault in the process of calculating the warning evaluation index in the and , extract 's calculation result as the length value of the rectangle, 's calculation result as the width value of the rectangle, construct a rectangle model according to the obtained length value and width value, and use the constructed rectangle model as the warning evaluation digital model of each fault;

[0083] After calculating the warning evaluation index corresponding to each fault, number them based on the specific type of the thermal fault classification data packet corresponding to each fault, and mark the warning evaluation index of each fault in the corresponding thermal fault classification data packet as , where e = a, b, c or d, corresponding to the mechanical thermal fault data packet, the insulation thermal fault data packet, the overload thermal fault data packet, and the environmental thermal fault data packet respectively; m = 1, 2 or k, and k is the total number of faults in the corresponding thermal fault classification data packet;

[0084] After numbering, integrate the warning evaluation digital model and the warning evaluation index of each fault in the corresponding thermal fault classification data packet into the warning set of the corresponding thermal fault classification data packet;

[0085] Intelligent warning trigger: Set the monitoring time interval of the distribution switchgear cabinet. After the set monitoring time interval is reached, analyze the temperature change parameters of each temperature acquisition point in the distribution switchgear cabinet within the set time period. If the analysis result of the temperature change parameters of a certain temperature acquisition point matches the warning set of the corresponding thermal fault type, directly trigger the warning signal of the corresponding thermal fault type;

[0086] It should be noted that the warning evaluation digital model and warning evaluation index in the warning set , will expand the data according to the preset fluctuation range. For example, the length value and width value of the digital model can be adjusted within the preset fluctuation range, and the warning evaluation index can be adjusted up and down within the preset fluctuation range. The warning evaluation digital model and warning evaluation index with variable fluctuation range are both used as the data in this warning set. If the analysis result of the temperature change parameters of a certain temperature acquisition point does not match the warning set of the corresponding thermal fault type, further match it with the data with variable fluctuation range. If the match is successful, the warning signal of the corresponding thermal fault type is also triggered.

[0087] Warning trigger execution: After the warning is triggered, perform corresponding operations based on the specific thermal fault type that triggers the warning;

[0088] Specifically:

[0089] S1: If the specific thermal fault type that triggers the warning is a mechanical thermal fault, collect each group of image information of the corresponding mechanical area of the distribution switchgear cabinet; collect through the deployed high-resolution cameras; use image processing technology to extract and mark the area with wear characteristics from the collected groups of image information, and regard the marked area as the wear area in each group of collected image information. Integrate the wear areas into a wear data packet. Draw a circle with the location of the distribution switchgear cabinet at the current time point as the center and a set distance as the radius, screen each maintenance personnel within the circle, and send a location feedback signaling to the mobile terminals of each maintenance personnel. After each maintenance personnel confirms the location feedback signaling, obtain the locations of each maintenance personnel within the circle, and thus obtain the distance of each maintenance personnel from the distribution switchgear cabinet at the current time point. Sort the distances of each group from small to large, extract the three maintenance personnel with the top three ranked distances, and obtain their working hours. Regard the maintenance personnel with the longest working hours as the selected personnel for the current triggered warning, and send the wear data packet and the location of the distribution switchgear cabinet to the mobile terminal of this person;

[0090] S2: If the specific thermal fault type triggering the warning is an insulation thermal fault, collect the image information of each group in the corresponding insulation area of the distribution switchgear cabinet, use image processing technology to extract and mark the areas with crack and aging features from the collected images of each group, take the marked areas as the crack areas and aging areas in the image information of each group, integrate the crack areas and aging areas into a damage data packet, and in the same way as step S1, send the damage data packet and the location of the distribution switchgear cabinet to the mobile terminal of the selected personnel;

[0091] S3: If the specific thermal fault type triggering the warning is an overload thermal fault, generate a load adjustment signaling and send it to the mobile terminal of the management personnel, and the management personnel receive the generated load adjustment signaling and reallocate the load;

[0092] S4: If the specific thermal fault type triggering the warning is an environmental thermal fault, after adjusting the power of the heat dissipation device; the specific size of the adjusted power is preset by the technical personnel; after adjusting the power, collect the image information of the corresponding ventilation port area of the distribution switchgear cabinet, use image processing technology to extract and mark the areas with dust features from the collected image information of each ventilation port area, take the marked areas as the pollution areas of the image information of each ventilation port area, and count the number of pixels in the pollution areas, perform the conversion of the actual size based on the resolution of the image to obtain the pollution area of the corresponding ventilation port area of the distribution switchgear cabinet at the current time point, preset the maximum allowable area of the pollution area, compare the pollution area of the corresponding ventilation port area of the distribution switchgear cabinet at the current time point with the preset maximum allowable area, if it is greater than the preset maximum allowable area, trigger a cleaning signaling, select the maintenance personnel with the shortest distance from the current time point to the distribution switchgear cabinet as the selected personnel, and send the cleaning signaling to the mobile terminal of this person;

[0093] Warning update mechanism: used to analyze the temperature change parameters when the thermal fault appears again in the distribution switchgear cabinet, and update the warning set to which it belongs based on the analysis result according to the specific type of the thermal fault;

[0094] The above formulas are all obtained by collecting a large amount of data for software simulation and selecting a formula close to the true value, and the influence weight factors and specific coefficient values in the formulas are set by the technical personnel in the field according to the actual situation and can be adjusted and modified later.

[0095] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only the specific embodiments. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. Intelligent fault diagnosis and early warning method for power distribution switch cabinet, characterized in that: include: Fault data collection: collect historical thermal fault cases and times of the distribution switch cabinet, and integrate them into a historical thermal fault data package of the distribution switch cabinet; Fault assessment classification: Analyze the historical thermal fault data packets of the power distribution switch cabinet. After the analysis is completed, classify them based on the specific causes that trigger the thermal faults of the power distribution switch cabinet to obtain a thermal fault classification data packet after the classification is completed; Data analysis and processing: Based on the thermal fault classification data packet after classification, extract the temperature change parameters within the set time period before each fault occurs in the corresponding thermal fault classification data packet and analyze them; Extract the temperature change parameters within the set time period before each fault occurs in the corresponding thermal fault classification data packet and analyze them, specifically: Extract the temperature values ​​in the set time period before each fault occurs in the corresponding thermal fault classification data packet, and obtain the temperature values ​​at each time point in the set time period according to the preset division time interval; Based on the temperature values ​​obtained at each time point, determine the window size of the moving average. For each time point and the determined window size, calculate the average value of all the value points within the window size range before and after the time point. Based on the advancement of the time series, move the sliding window forward and repeatedly calculate the average value of all the value points within the window size range before and after each time point until the entire data set is covered. Extract the average value calculated for each group, and thereby construct a line graph of temperature changes in a set time period before each fault occurs in the corresponding thermal fault classification data packet; draw the numerical points in the line graph corresponding to the average value of each time point, and connect adjacent numerical points to obtain an average line; Calculate the slope of each average line and the angle between it and the horizontal line. If the angle between a certain average line and the horizontal line is acute, the slope of the average line is marked as a descending slope. If the angle between a certain average line and the horizontal line is obtuse, the slope of the average line is marked as an ascending slope. Sum up all the descending slopes and ascending slopes respectively to obtain the descending total value and the ascending total value, and convert the obtained descending total value and the ascending total value into a score; Based on the specific type of the thermal fault classification data packet, preset the total value ranges of each group of the decreasing total value and the increasing total value, set each group of the total value ranges of the decreasing total value and the increasing total value to correspond to a score value, match the decreasing total value and the increasing total value corresponding to each fault with the corresponding total value ranges of each group based on the specific type of the thermal fault classification data packet, and obtain the score values ​​corresponding to the decreasing total value and the increasing total value of each fault; wherein the score value corresponding to the increasing total value is a positive number, and the score value corresponding to the decreasing total value is a negative number; The score value corresponding to the total increase value and the score value corresponding to the total decrease value are accumulated to obtain the trend change index Ea of each fault within the set time period; Establish an early warning mechanism: Based on the analysis results, construct a digital model and early warning evaluation index for each fault in the corresponding thermal fault classification data package. , the early warning evaluation digital model and early warning evaluation index of each fault in the corresponding thermal fault classification data package Integrate into a warning set corresponding to the thermal fault classification data packet; Intelligent early warning triggering: Set the monitoring time interval of the distribution switch cabinet, and after the set monitoring time interval is reached, analyze the temperature change parameters of each temperature collection point of the distribution switch cabinet within the set time period. If the temperature change parameter analysis result of a certain temperature collection point matches the early warning set of the corresponding thermal fault type, the early warning signal of the corresponding thermal fault type is directly triggered; Warning trigger execution: After the warning is triggered, the corresponding operation is performed based on the specific thermal fault type that triggered the warning; Early warning update mechanism: It is used to analyze the temperature change parameters when a thermal fault occurs again in the distribution switch cabinet, and update the corresponding early warning set based on the specific type of thermal fault.

2. The intelligent fault diagnosis and early warning method for distribution switch cabinet according to claim 1 is characterized in that: Extract the temperature change parameters within the set time period before each fault occurs in the corresponding thermal fault classification data packet and analyze them, including: Based on the temperature values ​​obtained at each time point, the average of the temperature values ​​at each time point is taken as the average temperature of each fault within the set time period, and based on the specific type of the thermal fault classification data packet, the reference threshold value of the temperature corresponding to each type of thermal fault is preset; At the same time, the highest and lowest values ​​are extracted from the temperature values ​​at each time point, and the temperature average, highest and lowest values ​​are respectively calculated with the corresponding preset reference thresholds. The calculated three groups of ratios are respectively used as the average ratio, highest ratio and lowest ratio of each fault in the set time period; Based on the specific type of the thermal fault classification data packet corresponding to the fault, the weight coefficients of the average ratio, the highest ratio and the lowest ratio are preset, and the calculated average ratio, the highest ratio and the lowest ratio are multiplied by the preset weight coefficients respectively, and then the sum is obtained to obtain the trend temperature index Eb of each fault within the set time period.

3. The intelligent fault diagnosis and early warning method for distribution switch cabinet according to claim 2 is characterized in that: According to the analysis results, the early warning evaluation index of each fault in the corresponding thermal fault classification data package is constructed. , specifically: Extract the trend change index Ea and trend temperature index Eb of each fault within a set time period, and preset reference indexes of the trend change index Ea and trend temperature index Eb according to the specific type of thermal fault classification data packet corresponding to each fault; According to the formula , perform weighted calculation on the trend change index Ea and trend temperature index Eb of each fault within the set time period to obtain the early warning evaluation index of each fault ;in and Respectively represent the reference index of each fault trend change index Ea and trend temperature index Eb; and They are the influence weight factors of trend change index Ea and trend temperature index Eb respectively.

4. The intelligent fault diagnosis and early warning method for distribution switch cabinet according to claim 3 is characterized in that: According to the analysis results, a digital model for early warning and evaluation of each fault in the corresponding thermal fault classification data package is constructed, specifically: Based on each failure, the early warning evaluation index is calculated In process and ,extract The calculated result is used as the length of the rectangle. The calculation result is used as the width value of the rectangle, and a rectangular model is constructed according to the obtained length value and width value, and the constructed rectangular model is used as the digital model for early warning evaluation of each fault.

5. The intelligent fault diagnosis and early warning method for distribution switch cabinet according to claim 4 is characterized in that: The early warning evaluation digital model and early warning evaluation index of each fault in the corresponding thermal fault classification data package Integrate into a warning set corresponding to the thermal fault classification data package, specifically: After calculating the early warning evaluation index corresponding to each fault After that, the specific types of thermal fault classification data packets corresponding to each fault are numbered, and the early warning evaluation index of each fault in the corresponding thermal fault classification data packet is Mark as , where e = a, b, c or d, corresponding to the mechanical thermal fault data packet, insulation thermal fault data packet, overload thermal fault data packet and environmental thermal fault data packet respectively; m=1, 2 or k, where k is the total number of faults in the corresponding thermal fault classification data packet; After the numbering is completed, the early warning evaluation digital model and early warning evaluation index of each fault in the corresponding thermal fault classification data package will be Integrate into a warning set corresponding to the thermal fault classification data package.

6. The intelligent fault diagnosis and early warning method for distribution switch cabinet according to claim 5 is characterized in that: Perform corresponding actions based on the specific thermal fault type that triggers the warning, specifically: S1: If the specific thermal fault type that triggers the early warning is a mechanical thermal fault, then collect each group of image information corresponding to the mechanical area of ​​the distribution switch cabinet, extract and mark the area of ​​wear characteristics from each group of collected image information, use the marked area as the wear area in each group of collected image information, integrate the wear area into a wear data packet, take the location of the distribution switch cabinet at the current time point as the center, set the distance as the radius to draw a circle, screen each maintenance personnel within the circle range, and send a position feedback signal to the mobile terminal of each maintenance personnel. After each maintenance personnel confirms the position feedback signal, the location of each maintenance personnel within the circle range is obtained, thereby obtaining the distance of each maintenance personnel from the distribution switch cabinet at the current time point, sorting each group of distances from small to large, extracting the top three maintenance personnel in terms of distance, and obtaining their working hours. The maintenance personnel with the longest working hours are selected as the personnel currently triggering the early warning, and the wear data packet and the location of the distribution switch cabinet are sent to the personnel's mobile terminal; S2: If the specific thermal fault type that triggers the early warning is an insulation thermal fault, then collect each group of image information corresponding to the insulation area of ​​the distribution switch cabinet, extract and mark the areas with crack and aging characteristics from each group of collected images, and use the marked areas as crack areas and aging areas in each group of image information. The crack areas and aging areas are integrated into a damaged data packet, and similarly to step S1, send the damaged data packet and the location of the distribution switch cabinet to the mobile terminal of the selected personnel.

7. The intelligent fault diagnosis and early warning method for distribution switch cabinet according to claim 6 is characterized in that: Execute corresponding actions based on the specific thermal fault type that triggers the warning, including: S3: If the specific thermal fault type that triggers the early warning is an overload thermal fault, a load adjustment signaling is generated and sent to the mobile terminal of the manager, and the manager receives the generated load adjustment signaling and redistributes the load; S4: If the specific thermal fault type that triggers the early warning is an environmental thermal fault, then after adjusting the power of the heat dissipation device, collect image information of the ventilation area corresponding to the distribution switch cabinet, use image processing technology to extract and mark the areas with dust features from the collected image information of each group of ventilation area, and use the marked areas as the contaminated areas of each group of ventilation area image information, and count the number of pixels in the contaminated areas, convert the actual size based on the resolution of the image, and obtain the contaminated area of ​​the ventilation area corresponding to the distribution switch cabinet at the current time point, preset the maximum allowable area of ​​the contaminated area, compare the contaminated area of ​​the ventilation area corresponding to the distribution switch cabinet at the current time point with the preset maximum allowable area, if it is greater than the preset maximum allowable area, trigger the cleaning signaling, select the maintenance personnel who is closest to the distribution switch cabinet at the current time point as the selected personnel, and send the cleaning signaling to the mobile terminal of the personnel.

Images