A smart fault judgment method for power internet of things

Abstract

The application discloses a kind of intelligent fault judgment methods of electric power internet of things, it is related to electric power internet of things technical field, in the intelligent fault judgment process of electric power internet of things, by in-depth analysis to the data processing mode corresponding to various fault data, and calculate data variation smooth coefficient and data discrete mutation coefficient, it is favorable to accurately determine the type of fault data belongs to, different types of data often reflect the different aspects of equipment operating state and fault characteristics, accurate classification can make subsequent fault judgment more targeted, when through each fault data processing model fusion mode to the remaining class fault data data processing, by combining histogram calculation data matching degree evaluation coefficient, and then get comprehensive fault evaluation coefficient, it is favorable to objectively, comprehensively analyze whether each remaining class fault data corresponding power equipment exists fault condition and the severity of fault.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power internet of things, and in particular to an intelligent fault judgment method for power internet of things. BACKGROUND

[0002] With the rapid development of power internet of things, the number of devices increases and the operation environment is complex, faults occur frequently and diagnosis is difficult, and the traditional fault judgment method cannot meet the precise and efficient requirements, prompting the generation of the above intelligent fault judgment method, aiming to improve the accuracy and timeliness of fault diagnosis by using advanced technology.

[0003] The prior art such as the patent application for invention with publication number CN114709923A discloses an intelligent fault judgment method for power internet of things, comprising the following steps: S1: dividing the intelligent fault diagnosis system framework of power internet of things into three parts of sensing layer, communication layer and application layer; S2: establishment of the sensing layer; S3: establishment of the communication layer; S4: establishment of the application layer; S5: construction of the overhaul scheduling module. The present application analyzes the device state information in the internet of things information platform through the cooperation of each module, judges the fault type of the device according to certain criteria, and the fault diagnosis framework will send the diagnosis results and related maintenance decisions to the maintenance personnel and the responsible person in the form of short messages in real time, timely eliminate the device fault hidden danger, maintain the device, reduce some unnecessary accidents, and effectively guarantee the real-time performance of fault diagnosis.

[0004] For the above-mentioned scheme, at least the following technical problems exist: 1. The above-mentioned scheme lacks precise analysis and processing of fault data, which will lead to reduced accuracy and reliability in actual intelligent fault judgment. The above-mentioned scheme lacks how to analyze the data processing mode corresponding to various fault data. In the complex power internet of things environment, different types of fault data have different characteristics and change rules. If accurate classification and targeted processing are not performed, the fault diagnosis system may misjudge or miss the fault condition, cannot timely and effectively find the potential problems of the device, thereby increasing the risk of device failure and reducing the stability and safety of the power system.

[0005] 2. The above-mentioned scheme lacks a way of fusing various fault data processing models to process data and evaluate the fault judgment abnormality of various fault data. In actual application, a single fault data processing model often has limitations and cannot comprehensively and accurately deal with various complex fault scenarios. The lack of model fusion method will make the system unable to fully exert the advantages of different models when facing diversified fault data, reducing the accuracy and adaptability of fault judgment.

[0006] 3. The above-mentioned method of calculating the characteristic fault location index corresponding to various fault data under fault conditions and evaluating the specific fault data in the fault data corresponding to each power equipment under fault conditions will lead to the inability to quickly and accurately locate which specific data reflect the most critical problem after the equipment fault is determined. Maintenance personnel will find it difficult to find the root cause of the fault quickly, prolong the equipment maintenance time, reduce the power system recovery efficiency, and further affect the reliability and stability of power supply. Summary of the Invention

[0007] The purpose of this invention is to provide an intelligent fault diagnosis method for the Internet of Things in the power sector, which solves the problems existing in the background technology.

[0008] To solve the above technical problems, the present invention adopts the following technical solution: The present invention provides an intelligent fault judgment method for the power Internet of Things, including: S1, when various types of fault data generated by each power device are transmitted to a designated power Internet of Things terminal, the data processing method corresponding to each type of fault data is analyzed.

[0009] S2. Based on the data processing methods corresponding to various types of fault data, evaluate the abnormal situations of fault judgment corresponding to various types of fault data.

[0010] S3. After all types of fault data have been processed by the respective fault data processing models, if a fault exists, calculate the characteristic fault location index corresponding to each type of fault data that has a fault.

[0011] S4. Based on the characteristic fault location index corresponding to various fault data of existing fault conditions, the specific fault data in the fault data of each power equipment with fault conditions is evaluated.

[0012] The beneficial effects of the present invention are as follows: 1. The present invention provides an intelligent fault judgment method for the Internet of Things in the power Internet of Things. In the process of intelligent fault judgment in the Internet of Things in the power Internet of Things, by conducting in-depth analysis of the data processing methods corresponding to various types of fault data and calculating the data change smoothing coefficient and data discrete mutation coefficient, it is beneficial to accurately determine the type of fault data. Different types of data often reflect different aspects of the operating status and fault characteristics of the equipment. Accurate classification can make subsequent fault judgment more targeted.

[0013] 2、The embodiment of the present application is in the evaluation of various types of fault data corresponding to the fault judgment abnormal situation, through the remaining class fault data that does not conform to the common data type is processed by using the fusion of each fault data processing model, it is beneficial to give full play to the advantages of different models, thereby improving the accuracy of each type of fault detection data corresponding to the fault judgment, through the fusion of each fault data processing model, make up for the limitations of single model, in the face of complex and variable power equipment fault condition, more comprehensive and accurate identification of fault signal, reduce the risk of misjudgment and omission, guarantee the stable operation of power system.

[0014] 3、The embodiment of the present application is in the calculation of the characteristic fault positioning index corresponding to each type of fault data in the presence of fault condition and the evaluation of the specific fault data in the corresponding fault data of each power equipment in the presence of fault condition, it is beneficial to accurately locate the fault root, when the power equipment fails, it is crucial to quickly and accurately find the key data and components that cause the fault, from numerous fault monitoring data, select the factor that is most likely to cause the fault, so that the maintenance personnel can have a clear target for repair work, greatly shorten the maintenance time, reduce the loss of power failure, improve the availability and reliability of power equipment.

[0015] 4、The embodiment of the present application is in the calculation of the model adaptability index corresponding to each fault data processing model, and the calculation of the data matching degree evaluation coefficient combined with the histogram, and then the comprehensive fault evaluation coefficient is obtained, it is beneficial to objectively and comprehensively analyze whether each remaining class fault data corresponding to the power equipment exists in the fault condition and the severity of the fault, the model adaptability index reflects the performance of each fault data processing model in the current data processing task, the data matching degree evaluation coefficient provides supplementary information from the perspective of data distribution characteristics, the comprehensive fault evaluation coefficient formed by the combination of the two can more accurately measure the possibility of fault occurrence and the fault level, considering multiple factors, improve the scientificity and reliability of fault judgment, provide a strong guarantee for the safe and stable operation of power system. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical scheme in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description, obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0017] Figure 1 The embodiment of the present application is in the evaluation of various types of fault data corresponding to the fault judgment abnormal situation, through the remaining class fault data that does not conform to the common data type is processed by using the fusion of each fault data processing model, it is beneficial to give full play to the advantages of different models, thereby improving the accuracy of each type of fault detection data corresponding to the fault judgment, through the fusion of each fault data processing model, make up for the limitations of single model, in the face of complex and variable power equipment fault condition, more comprehensive and accurate identification of fault signal, reduce the risk of misjudgment and omission, guarantee the stable operation of power system. DETAILED DESCRIPTION

[0018] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0019] Please refer to Figure 1 As shown in the drawings, the present application provides a smart fault judgment method for power Internet of Things, which comprises the following steps: S1, when various types of fault data corresponding to each power device are transmitted to a specified power Internet of Things terminal, analyzing the data processing mode corresponding to each type of fault data.

[0020] In a specific embodiment, the analyzing the data processing mode corresponding to each type of fault data comprises the following specific analysis process: according to a set data collection time point, through a sensor device set in each power device in the specified power Internet of Things, each type of fault data of each power device at each data collection time point is collected, when the sensor device transmits each type of fault data corresponding to each power device to the specified power Internet of Things terminal, the data change smoothing coefficient and the data discrete mutation coefficient corresponding to each type of fault data are calculated, respectively denoted as DVi and DDMi, wherein i is the number corresponding to each type of fault data, i = 1, 2, …, n, n represents the total number of types corresponding to the fault monitoring data, n is a positive integer, and then the data type to which each type of fault data belongs is analyzed, the data type to which each type of fault data belongs includes continuous data, discrete data and time series data, according to the data type to which each type of fault data belongs.

[0021] The data type processed by each fault data processing model is obtained from the database, thereby determining each fault data processing model corresponding to each type of fault data for fault judgment, and then each type of fault data is processed through each fault data processing model.

[0022] It should be noted that, taking a power transformer as an example, each type of fault data includes continuous data such as winding temperature, discrete data such as tap changer switching times, and time series data such as load current, etc.

[0023] In a specific embodiment, the calculating the data discrete mutation coefficient corresponding to each type of fault data comprises the following specific calculation process: through the calculation formula: The data change smoothing coefficient DVi corresponding to each type of fault data is obtained, t is the number corresponding to the data collection time point, t = 1, 2, …, m, m represents the total number of data collection time points, m is a positive integer,

[0024] X i (t+1)-Xit | represents the absolute value of the difference between the data value corresponding to the i-th type of fault data at the t+1-th data collection time point and the data value corresponding to the t-th data collection time point, denoted as each absolute value of the difference, min(X i (t+1)-X it ) represents the minimum value in each absolute value of the difference.

[0025] It should be noted that the "data value" refers to the specific numerical value obtained by each type of fault data at each data collection time point, for example, for voltage monitoring data of a power equipment, the voltage value collected at the first data collection time point is 220 volts, and the voltage value collected at the second data collection time point is 218 volts, then the two voltage values are the corresponding data values.

[0026] In a specific embodiment, the calculation of the data discrete mutation coefficient corresponding to each type of fault data is as follows: by the calculation formula:

[0027] The data discrete mutation coefficient DDMi corresponding to each type of fault data is obtained, where λi represents the standard deviation corresponding to the i-th type of fault data, represents the number of absolute values of the difference greater than the sum of the mean value and 3 times the standard deviation of the data value corresponding to the i-th type of fault data.

[0028] It should be noted that by the calculation formula: The standard deviation λi corresponding to the i-th type of fault data is obtained, and the standard deviation is a statistical quantity used to measure the dispersion degree of a group of data. In the context of calculating the data discrete mutation coefficient, for example, for temperature data of a power equipment, the standard deviation reflects the dispersion of the temperature value relative to the average value. When calculating the data discrete mutation coefficient, if the equipment temperature data is normally distributed, the standard deviation is small, when the equipment has local overheating and other faults, the dispersion degree of the temperature data increases, the standard deviation becomes larger, and the number of absolute values of the difference greater than the sum of the mean value and 3 times the standard deviation of the data value corresponding to the i-th type of fault data also changes.

[0029] In the evaluation of the fault judgment abnormal situation corresponding to each type of fault data, the remaining type of fault data that does not conform to the common data type is processed by using the fusion of each fault data processing model, which is advantageous to fully exert the advantages of different models, thereby improving the accuracy of fault judgment corresponding to each type of fault detection data, and compensating for the limitations of a single model by fusing each fault data processing model. In the face of complex and variable power equipment fault conditions, the fault signal is more comprehensively and accurately identified, the risk of misjudgment and omission is reduced, and the stable operation of the power system is ensured.

[0030] S2, according to the data processing mode corresponding to each type of fault data, and then evaluate the fault judgment abnormal situation corresponding to each type of fault data.

[0031] In a specific embodiment, the analysis obtains the data type to which each type of fault data corresponds, and the specific process is as follows: obtaining the preset standard data variation smoothing coefficient threshold, standard data discrete mutation coefficient threshold and standard discrete proportion threshold from the database, and denoting them as DV', DDM' and DDM" respectively, and comparing the data variation smoothing coefficient and the data discrete mutation coefficient corresponding to each type of fault data with the standard data variation smoothing coefficient threshold and the standard data discrete mutation coefficient threshold respectively.

[0032] If (DV i ≥ DV') and (DDM i ≤ DDM'), it indicates that the data type to which the i-th type of fault data belongs is continuous data, and the discrete proportion Bi

[0033] If (DV i < DV') and (DDM i > DDM'), and B i ≥ DDM", it indicates that the data type to which the i-th type of fault data belongs is discrete data, and if (DV i < DV') and (DDM i ≤ DDM'), it indicates that the data type to which the i-th type of fault data belongs is time series data.

[0034] It should be noted that the setting process of the discrete proportion Bi is as follows: for example, in monitoring the operation of power equipment of a substation, voltage, current and temperature are collected, and in a set time, 100 data points are collected, among which there are 40 valid data points of voltage data, 30 valid data points of current data and 30 valid data points of temperature data, and the discrete proportion of voltage data is the number of valid data points of voltage data 40 divided by the total number of data points 100, and the discrete proportions of current and temperature are obtained in the same way.

[0035] In a specific embodiment, the evaluation of the fault judgment abnormal situation corresponding to each type of fault data has the following specific process: according to the data type to which each type of fault data corresponds, each type of fault data that does not belong to continuous data, discrete data and time series data is recorded as each remaining type of fault data, and then each remaining type of fault data is processed by fusing each fault data processing model, so as to improve the accuracy of fault judgment corresponding to each type of fault detection data.

[0036] It should be noted that intelligent fault judgment of various types of fault data by various fault data processing models has been involved in the patent document with publication number CN117269641A, and will not be described in more detail here.

[0037] In a specific embodiment, the data processing of the fault data of this type by fusing various fault data processing models is as follows: the fault results corresponding to each remaining type of fault data processed by fusing various fault data processing models are recorded as Q qj , q is the number of each remaining type of fault data, q = 1, 2,..., p, p represents the total number of remaining types of fault data, p is a positive integer, j is the number of each fault data processing model, j = 1, 2,..., y, y represents the total number of fault data processing models, y is a positive integer;

[0038] Through the calculation formula: the model adaptability index α corresponding to each fault data processing model is obtained j , wherein acc j , rec j , sta j respectively represent the accuracy, recall rate, and stability in response to environmental changes, μ1, μ2, μ3 are respectively the set accuracy weight factor, recall rate weight factor, and stability in response to environmental changes weight factor, and the data matching degree evaluation coefficient β is calculated according to the histogram corresponding to each fault data processing model. j Thus, through the calculation formula: the comprehensive fault evaluation coefficient γq corresponding to each remaining type of fault data is obtained, and it is analyzed whether the power equipment corresponding to each remaining type of fault data has a fault condition.

[0039] It should be noted that the accuracy is obtained by the ratio of the number of data correctly judged by the fault data processing model to the total number of judged data, the recall rate is the ratio of the number of fault data correctly judged by the fault data processing model to the actual number of fault data, and the stability in response to environmental changes is obtained by measuring the fluctuation of the output results of the model under different working conditions and environmental changes, such as the output result variance. The specific analysis process is prior art, and will not be described in more detail here.

[0040] It also needs to be explained that, taking a small hydropower station as an example, assuming that there are models A, B and C, for model A, the stable power generation period accounts for 60%, the accuracy rate is calculated by counting the number of correct judgments and the total number of judgments of the fault and normal state, assuming that the accuracy rate is 80%, the recall rate is the ratio of the number of correct judgments to the actual number of faults, assuming that it is 75%, the stability is measured by comparing the output fluctuation under different environments, assuming that the stability score is 100 points, then it is set to 70 points, the accuracy rate weight is determined by the analytic hierarchy process 0.4, the flood season accounts for 25%, the recall rate weight is determined by the fuzzy comprehensive evaluation method 0.35, and the stability weight is determined by the grey correlation analysis method according to the change of the environment 0.25. The analytic hierarchy process, fuzzy comprehensive evaluation method and grey correlation analysis method are all analysis processes of prior art, and will not be described in detail here.

[0041] In a specific embodiment, the analysis obtains whether the power equipment corresponding to each residual class fault data exists a fault condition, and the specific process is as follows: obtaining the range interval corresponding to each standard comprehensive fault evaluation coefficient threshold preset for each power equipment corresponding to each residual class fault data from the database, and obtaining each fault grade corresponding to the range interval corresponding to each standard comprehensive fault evaluation coefficient threshold, comparing the comprehensive fault evaluation coefficient corresponding to each residual class fault data with the preset standard comprehensive fault evaluation coefficient threshold, if the comprehensive fault evaluation coefficient corresponding to a residual class fault data belongs to the range interval corresponding to a preset standard comprehensive fault evaluation coefficient threshold, it indicates that the power equipment corresponding to the residual class fault data exists a fault condition, and it indicates that the fault grade of the power equipment corresponding to the residual class fault data is the fault grade corresponding to the range interval corresponding to the standard comprehensive fault evaluation coefficient threshold, otherwise, it indicates that each power equipment corresponding to the residual class fault data does not exist a fault condition.

[0042] In the process of calculating the feature fault positioning index corresponding to each fault data of the power equipment existing a fault condition and evaluating the specific fault data in the fault data of the power equipment existing a fault condition, the embodiment of the application is beneficial to accurately positioning the fault root cause, when the power equipment fails, it is crucial to quickly and accurately find the key data and components causing the failure, from numerous fault monitoring data, the factors most likely to cause the failure are screened out, so that the maintenance personnel can carry out maintenance work with a clear purpose, greatly shorten the maintenance time, reduce the power loss, and improve the availability and reliability of the power equipment.

[0043] S3、When each fault data is processed by each fault data processing model, if there is a fault condition, the feature fault positioning index corresponding to each fault data existing a fault condition is calculated.

[0044] In one specific embodiment, the feature fault positioning index corresponding to each type of fault data in the presence of a fault condition is calculated as follows: the type of fault data corresponding to a certain power equipment at a certain data collection time point in the presence of a fault condition is recorded as a feature vector E, E = [e1, e2,..., eg], g represents each data in the type of fault data corresponding to a certain power equipment at a certain data collection time point, g is a positive integer, E is input into the fault data processing model to obtain the corresponding fault probability value F, and the feature disturbance value Δe g is set h , then a new feature vector E' is obtained, E' = [e1, e2,..., eg h + Δe h ,..., eg g ], E' is input into the fault data processing model to obtain the corresponding fault probability value F', and then the feature fault positioning index E h corresponding to the hth data in the type of fault detection data corresponding to a certain power equipment in the presence of a fault condition is obtained through the calculation formula: .

[0045] It should be noted that h takes any integer in the interval [1, g], and the setting process of Δe h is as follows: for example, when monitoring the fault of a power equipment, the input feature vector E includes the voltage e1, the current e2 and the temperature e3 of the power equipment, for the voltage feature e1, if the current voltage value is 220 volts, then Δe h = 0.01 x e1 = 2.2 volts, and a small voltage disturbance value Δe h is obtained, and Δe h is used to observe the influence on the fault judgment result output by the fault data processing model when the feature changes.

[0046] It should be further noted that, taking a fault data processing model for fault diagnosis of a power equipment as an example, the feature vector E = [220, 10, 50] composed of the voltage 220 volts, the current 10 amperes and the temperature 50°C of the power equipment collected at a certain data collection time point is input into the fault data processing model, the fault data processing model outputs the fault probability value F = 0.2 according to the internal neural network structure, the weight parameters and the trained algorithm through a complex calculation process, which means that according to the judgment of the fault data processing model, the probability of the fault of the power equipment at this time is 20%, and the fault data processing model obtains the fault probability value through a complex calculation process, which is the prior art and will not be described in detail here.

[0047] The embodiment of the present application is beneficial to objectively and comprehensively analyze whether the power equipment corresponding to each residual class fault data exists in a fault condition and the severity of the fault by calculating the model adaptability index corresponding to each fault data processing model, combining the histogram to calculate the data matching degree evaluation coefficient, and then obtaining the comprehensive fault evaluation coefficient, the model adaptability index reflects the performance of each fault data processing model in the current data processing task, the data matching degree evaluation coefficient provides supplementary information from the perspective of data distribution characteristics, and the comprehensive fault evaluation coefficient formed by the combination of the two can more accurately measure the possibility of fault occurrence and the fault level, comprehensively considers multiple factors, improves the scientificity and reliability of fault judgment, and provides a strong guarantee for the safe and stable operation of the power system.

[0048] S4, according to the characteristic fault positioning index corresponding to each type of fault data in the fault condition, further evaluating the specific fault data in the fault data corresponding to each power equipment in the fault condition.

[0049] In a specific embodiment, the specific process of evaluating the specific fault data in the fault data corresponding to each power equipment in the fault condition is as follows: the characteristic fault positioning index E h According to the descending order, the data corresponding to the first ranked characteristic fault positioning index is the first probability of existing faults, the data corresponding to the second ranked characteristic fault positioning index is the second probability of existing faults, and thus the probability of existing faults of the specific fault data in the fault data corresponding to each power equipment is obtained, and thus the specific fault data in the fault data corresponding to each power equipment in the fault condition is obtained.

[0050] It should be noted that, for example, for a certain power transformer, the fault detection data includes winding temperature, load current, insulation resistance and the like, and the hth data herein refers to a certain one of the numerous fault detection data, and when h=1, it represents the winding temperature.

[0051] The intelligent fault judgment method of the power Internet of Things provided by the embodiment of the present application is beneficial to accurately determine the type of fault data by deeply analyzing the data processing mode corresponding to each type of fault data and calculating the data change smoothing coefficient and the data discrete mutation coefficient in the intelligent fault judgment process of the power Internet of Things, and different types of data often reflect different aspects of the running state and fault characteristics of the equipment. Accurate classification can make the subsequent fault judgment more targeted.

[0052] The above merely illustrates and describes the concept of the present application, and those skilled in the art can make various modifications or supplements to the described specific embodiments or adopt similar ways to replace, as long as the modifications or supplements do not deviate from the concept of the present application or exceed the scope defined by the specification, and the modifications or supplements shall belong to the protection scope of the present application.

Claims

1. A smart fault judgment method for power Internet of Things, characterized in that, Comprise: S1, when each power equipment corresponding generated each type of fault data is transmitted to the designated power internet of things terminal, analyze the data processing mode corresponding to each type of fault data; S2, according to the data processing mode corresponding to each type of fault data, further evaluate the fault judgment abnormal situation corresponding to each type of fault data; According to the set data collection time point, by specifying the sensor device set in each power device in the power internet of things, and then collecting each type of fault data of each power device at each data collection time point, when the sensor device transmits each type of fault data corresponding to each power device to the specified power internet of things terminal, by calculating the data change smoothing coefficient and the data discrete mutation coefficient corresponding to each type of fault data, respectively denoted as and , wherein is the number corresponding to each type of fault data, , , indicates the total number of types corresponding to the fault monitoring data, is a positive integer, and then the data type to which each type of fault data belongs is analyzed, including continuous data, discrete data and time series data, according to the data type to which each type of fault data belongs; The data change smoothing coefficient is calculated by the absolute value of the difference value of the data value corresponding to the adjacent data collection time point and the data value corresponding to the data collection time point; The data discrete mutation coefficient is calculated by the number of absolute values of each difference greater than the first The number of data acquisition time points is calculated by adding the mean value of the corresponding data values of the class fault data and 3 times the standard deviation. Obtain the data type processed by each fault data processing model from the database, thereby determine the corresponding each fault data processing model of each type of fault data for fault judgment, and further process each type of fault data through each fault data processing model respectively; S3, when each type of fault data is processed through each fault data processing model, if there is a fault condition, calculate the characteristic fault positioning index corresponding to each type of fault data with fault condition; The class fault data of the power equipment in the fault condition at a certain data collection time point is recorded as a feature vector , , Each data in the class fault data of the power equipment at a certain data collection time point is represented is a positive integer, and is input into the fault data processing model to obtain a corresponding fault probability value , a feature disturbance value is set, and a new feature vector , is obtained is input into the fault data processing model to obtain a corresponding fault probability value , and a feature fault positioning index corresponding to the first data in the type fault detection data of the power equipment in the fault condition is obtained through a calculation formula , The feature fault positioning indexes corresponding to various types of fault data in the fault condition are obtained through the calculation. S4, according to the characteristic fault positioning index corresponding to each type of fault data with fault condition, further evaluate the specific fault data in the fault data corresponding to each power equipment with fault condition. 2.The smart fault judgment method of the power Internet of Things according to claim 1, characterized in that, The calculation of the data change smoothing coefficient corresponding to each type of fault data is as follows: By the formula: , the data change smoothing coefficients corresponding to various fault data are obtained , is the number corresponding to the data acquisition time point, , represents the total number of data acquisition time points, is a positive integer, represents the first fault data at the data acquisition time point corresponding to the data value and the data acquisition time point corresponding to the data value, recorded as the absolute value of each difference, represents the minimum value in the absolute value of each difference. 3.The smart fault judgment method of the power Internet of Things according to claim 2, characterized in that, The calculation of the data change smoothing coefficient corresponding to each type of fault data is as follows: By the calculation formula: , the data discrete mutation coefficients corresponding to various fault data are obtained , wherein represents the standard deviation corresponding to the first fault data, represents the number of absolute values greater than the mean value of the data value corresponding to the first fault data and 3 times the standard deviation. 4.The smart fault judgment method of the power Internet of Things according to claim 3, characterized in that, The analysis of the data type corresponding to each type of fault data is as follows: Obtain preset standard data variation smoothing coefficient threshold, standard data discrete mutation coefficient threshold and standard discrete proportion threshold from the database, and mark them as , and Compare the data variation smoothing coefficient and the data discrete mutation coefficient corresponding to each type of fault data with the standard data variation smoothing coefficient threshold and the standard data discrete mutation coefficient threshold respectively. If , it indicates that the first class fault data belongs to continuous data type, and the discrete proportion corresponding to each class of fault data is set ; If and , it indicates that the data type of the first class fault data is discrete data, if , it indicates that the data type of the first class fault data is time series data. 5.The smart fault judgment method of the power Internet of Things according to claim 1, characterized in that, The evaluation of the fault judgment abnormal situation corresponding to each type of fault data is as follows: According to the data type corresponding to each type of fault data, each remaining type of fault data which does not belong to continuous data, discrete data and time series data is recorded, and each remaining type of fault data is processed by fusing each fault data processing model, so as to improve the accuracy of fault judgment of each type of fault detection data. 6.The smart fault judgment method of the power Internet of Things according to claim 5, characterized in that, The data processing of this type of fault data by fusing each fault data processing model is as follows: The fault results corresponding to each residual class fault data processed by the fusion of each fault data processing model are denoted as , is the number of each residual class fault data, , is the total number of residual class fault data, is a positive integer, is the number of each fault data processing model, , is the total number of fault data processing models, is a positive integer; Through the calculation formula: , the model adaptability index corresponding to each fault data processing model is obtained , wherein , , respectively represent the accuracy, recall rate, and stability in coping with environmental changes, , , are respectively the set accuracy weight factor, recall rate weight factor, and stability weight factor in coping with environmental changes, and the data matching degree evaluation coefficient is calculated according to the histogram corresponding to each fault data processing model , thus through the calculation formula: , the comprehensive fault evaluation coefficient corresponding to each residual class fault data is obtained , and it is analyzed whether the power equipment corresponding to each residual class fault data exists a fault condition. 7.The smart fault judgment method of the power Internet of Things according to claim 6, characterized in that, The analysis of whether the power equipment corresponding to each remaining type of fault data has a fault condition is as follows: Obtain the range interval corresponding to each standard comprehensive fault evaluation coefficient threshold of each power equipment corresponding to each remaining type of fault data from the database, and obtain the fault grade corresponding to each range interval of each standard comprehensive fault evaluation coefficient threshold. Compare the comprehensive fault evaluation coefficient corresponding to each remaining type of fault data with the preset standard comprehensive fault evaluation coefficient threshold. If the comprehensive fault evaluation coefficient corresponding to a certain remaining type of fault data belongs to the range interval corresponding to a certain standard comprehensive fault evaluation coefficient threshold, it indicates that the power equipment corresponding to the remaining type of fault data has a fault condition, and the fault grade of the power equipment corresponding to the remaining type of fault data is the fault grade corresponding to the range interval of the standard comprehensive fault evaluation coefficient threshold. Otherwise, it indicates that the power equipment corresponding to the remaining type of fault data does not have a fault condition. 8.The smart fault judgment method of the power Internet of Things according to claim 1, characterized in that, The evaluation of the specific fault data in the fault data corresponding to each power equipment with fault condition is as follows: The fault detection data corresponding to the type of fault in a certain power equipment that has a fault condition. The characteristic fault location index corresponding to each data point The fault location index is sorted from largest to smallest. The probability of a fault is the first one, and the probability of a fault is the second one. This gives the probability of a specific fault in the fault data of each power device. Thus, the specific fault data of each power device with a fault is obtained.

Images