Electrical equipment automatic fault diagnosis and repair system based on intelligent control

Abstract

The invention relates to the technical field of fault diagnosis and repair, in particular to an electrical equipment automatic fault diagnosis and repair system based on intelligent control. The system comprises an abnormality diagnosis unit, a feature calculation unit, an optimal opportunity unit and a fault repair unit. The optimal opportunity unit outputs a weight coefficient according to the acquired running state data through a linear regression model, and performs optimal opportunity calculation according to the electrical equipment characteristic change rate at the time # imgabs0 # moment, a fixed time interval, the number of time points, the mean value of the electrical equipment characteristic change rate, the standard deviation of the electrical equipment characteristic change rate and the weight coefficient; the optimal opportunity for collecting the fault feature change rule during automatic diagnosis of the electrical equipment fault is calculated, the key feature change rule of the electrical equipment fault can be captured by collecting data at the optimal opportunity, the fault diagnosis error is reduced, the corresponding speed of fault diagnosis is accelerated, and the fault diagnosis efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis and repair, and in particular to an automatic fault diagnosis and repair system for electrical equipment based on intelligent control. Background Art

[0002] Automatic fault diagnosis and repair of electrical equipment with intelligent control has a wide range of application scenarios, covering the fields of industrial automation, energy management, building management, transportation, smart home, medical equipment, water treatment facilities and electricity. When monitoring the operating status of electrical equipment for abnormalities, it is impossible to accurately collect the fault characteristic change law at the best time when automatically diagnosing the fault of the electrical equipment, so that the electrical equipment misses the best time to collect the fault characteristic change law during the automatic diagnosis of the fault, increases the fault diagnosis error, and leads to a decrease in fault diagnosis efficiency. At the same time, due to the missed best time to collect the fault characteristic change law, the probability of misdiagnosis and repair errors of electrical equipment increases, resulting in a decrease in the accuracy of electrical equipment fault repair. Therefore, we provide an automatic fault diagnosis and repair system for electrical equipment based on intelligent control. Summary of the invention

[0003] The purpose of the present invention is to provide an automatic fault diagnosis and repair system for electrical equipment based on intelligent control to solve the problems raised in the above background technology.

[0004] To achieve the above-mentioned object, the present invention provides an automatic fault diagnosis and repair system for electrical equipment based on intelligent control, comprising an abnormality diagnosis unit, a feature calculation unit, an optimal timing unit, and a fault repair unit;

[0005] The abnormality diagnosis unit obtains the operating status and historical data of the electrical equipment to calculate the pressure abnormality value of the electrical equipment, uses the calculated pressure abnormality value to determine whether the electrical equipment is abnormal, and extracts historical fault data and corresponding historical fault time from the fault library to predict the fault of the electrical equipment and the fault time of the electrical equipment;

[0006] The characteristic calculation unit is used to receive data from the abnormal diagnosis unit, and the characteristic calculation unit takes data from the abnormal diagnosis unit, and calculates the electrical equipment at time T according to the taken data. f -t k The characteristic change rate at the moment, and then calculate the mean and standard deviation of the characteristic change rate of the electrical equipment according to the characteristic change rate of the electrical equipment;

[0007] The optimal timing unit is used to receive data from the feature calculation unit and the abnormality diagnosis unit, according to the time T f -t kThe electrical equipment characteristic change rate at the moment, the mean value and the standard deviation of the electrical equipment characteristic change rate are calculated to calculate the best time to collect the fault characteristic change law when automatically diagnosing electrical equipment faults;

[0008] The abnormal diagnosis unit is used to receive the best time calculated in the best time unit, calculate the start time of collecting the fault characteristic change law, use the predicted electrical equipment fault time as the end time of collecting the fault characteristic change law, automatically diagnose the electrical equipment fault type according to the extracted historical fault data and the corresponding historical fault time, and collect the fault characteristic change law from the best time between the start time and the end time, and then diagnose the electrical equipment fault type again according to the extracted historical fault data, the corresponding historical fault time and the collected fault characteristic change law;

[0009] The fault repair unit is used to receive data in the abnormal diagnosis unit, formulate an electrical equipment fault repair strategy according to the data in the abnormal diagnosis unit, and then perform fault repair on the diagnosed electrical equipment fault type according to the formulated repair strategy.

[0010] As a further improvement of the technical solution, the abnormality diagnosis unit includes an abnormality judgment module and a prediction diagnosis module.

[0011] The abnormality judgment module obtains the operation status data of the electrical equipment in real time through the intelligent sensor, and obtains the historical data, takes the actual pressure data according to the obtained operation status data, and then obtains the historical pressure data from the historical data, updates the actual pressure data X and the historical pressure data into a complete historical pressure data set WP, and then records the number of data n in the complete historical pressure data set, and uses the Z-score algorithm to calculate the pressure abnormality value of the electrical equipment according to the complete historical pressure data set WP and the number of data n, and then uses the calculated pressure abnormality value to judge the abnormality. When the calculated pressure abnormality value is greater than the set pressure standard value, it indicates that the electrical equipment is abnormal;

[0012] The predictive diagnosis module is used to receive command data indicating that an abnormality exists in the electrical equipment in the abnormality judgment module, and extract historical fault data and corresponding historical fault times from the fault library, and use a convolutional neural network to predict electrical equipment faults and electrical equipment fault times based on the extracted historical fault data and corresponding historical fault times.

[0013] As a further improvement of the technical solution, the feature calculation unit includes a change rate module and a data calculation module;

[0014] The change rate module is used to receive the electrical equipment fault command predicted by the prediction diagnosis module. The change rate module takes the acquired operating status data and the corresponding historical fault time in the prediction diagnosis module to the abnormality judgment module, and extracts the most recent fault occurrence time T from the corresponding historical fault time.f , according to the acquired running status data, extract the time when the most recent fault occurred T f At time t k The electrical equipment characteristic value S(T f -t k ), and record the number of time points m, and then use time T f -t k The electrical equipment characteristic value S(T f -t k ) Calculate the electrical equipment at time T f -t k The characteristic change rate at the moment, and the time T f -t k The characteristic change rate of electrical equipment at the time R(T f -t k ), and simultaneously record the number N of characteristic change rates of electrical equipment;

[0015] Get time T f -t k The characteristic change rate of electrical equipment at the time R(T f -t k )’s specific algorithm formula:

[0016]

[0017] Among them, t k refers to the kth time point, and Δt refers to a fixed time interval, which is used to calculate the length of the change in the characteristics of an electrical device between two different time points.

[0018] As a further improvement of the technical solution, the data calculation module is used to receive the electrical equipment characteristic change rate R and the number of electrical equipment characteristic change rates N in the change rate module, and calculate the mean value of the electrical equipment characteristic change rate according to the electrical equipment characteristic change rate R and the number of electrical equipment characteristic change rates N. Among them, R j Refers to the jth electrical equipment characteristic change rate R, and then according to the electrical equipment characteristic change rate R, the number of electrical equipment characteristic change rates N and the mean value of the electrical equipment characteristic change rate R μ Calculate the standard deviation of the rate of change of electrical equipment characteristics

[0019] As a further improvement of the technical solution, the optimal timing unit is used to receive data from the change rate module and the data calculation module, receive the operating status data obtained in the abnormal judgment module, and output the weight coefficient according to the obtained operating status data through the linear regression model. Use the arg max function according to time T f -tk The characteristic change rate of electrical equipment at the time R(T f -t k ), fixed time interval Δt, number of time points m, mean value R of the characteristic change rate of electrical equipment μ , Standard deviation R of the rate of change of electrical equipment characteristics σ and weight coefficient Calculate the best time to collect the changing rules of fault characteristics when automatically diagnosing electrical equipment faults.

[0020] As a further improvement of this technical solution, the implementation principle of using the arg max function to calculate the best time is:

[0021] According to the time T f -t k The characteristic change rate of electrical equipment at the time R(T f -t k ), fixed time interval Δt, number of time points m, mean value R of the characteristic change rate of electrical equipment μ , Standard deviation R of the rate of change of electrical equipment characteristics σ and weight coefficient Calculate the optimal time period for collecting the changing rules of fault characteristics when automatically diagnosing electrical equipment faults, and obtain the optimal time period Δt best That is the best time to calculate, the specific algorithm formula is:

[0022]

[0023] in, The value is the kth weight coefficient, is a mathematical function that is used to find the fixed time interval Δt that maximizes the value in the formula.

[0024] As a further improvement of the technical solution, the prediction diagnosis module is used to receive the best time Δt calculated in the best time unit. best , calculate the start time of collecting fault feature change rules To = Y s -Δt best , the predicted electrical equipment failure time Y s The start time To and the end time Et are used as the end time Et for collecting the changing law of fault characteristics, and then the convolutional neural network is used to automatically diagnose the fault type of the electrical equipment according to the extracted historical fault data and the corresponding historical fault time. The fault characteristic changing law is collected from the best time between the start time and the end time, and then the convolutional neural network is used to diagnose the fault type of the electrical equipment again according to the extracted historical fault data, the corresponding historical fault time and the collected fault characteristic changing law.

[0025] As a further improvement of the present technical solution, the fault repair unit is used to receive the fault characteristic change law collected in the predictive diagnosis module, the diagnosed electrical equipment fault type and the extracted historical fault data, and use a convolutional neural network to formulate an electrical equipment fault repair strategy based on the collected fault characteristic change law and the extracted historical fault data, and then perform fault repair on the diagnosed electrical equipment fault type according to the formulated repair strategy.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] 1. In the electrical equipment automatic fault diagnosis and repair system based on intelligent control, the optimal timing unit outputs the weight coefficient according to the acquired operating status data through the linear regression model, and f -t k The optimal time is calculated based on the characteristic change rate of electrical equipment at each moment, the fixed time interval, the number of time points, the mean and standard deviation of the characteristic change rate of electrical equipment, and the weight coefficient. The optimal time to collect the changing law of fault characteristics when automatically diagnosing electrical equipment faults is calculated. Collecting data at the optimal time can ensure that the key characteristic changing law of electrical equipment faults is captured, reduce fault diagnosis errors, thereby speeding up the response speed of fault diagnosis and improving fault diagnosis efficiency.

[0028] 2. In the automatic fault diagnosis and repair system for electrical equipment based on intelligent control, the predictive diagnosis module calculates the start time for collecting the fault characteristic change law according to the predicted fault time of the electrical equipment and the calculated optimal time, and uses the predicted fault time of the electrical equipment as the end time for collecting the fault characteristic change law. The start time and the end time are then input into the convolutional neural network. The fault type of the electrical equipment is automatically diagnosed according to the extracted historical fault data and the corresponding historical fault time, and the fault characteristic change law is collected from the optimal time between the start time and the end time. According to the fault characteristic change law, it can be determined when it is most effective to repair the fault, thereby reducing the downtime of the electrical equipment. The fault type of the electrical equipment is re-diagnosed according to the extracted historical fault data, the corresponding historical fault time and the collected fault characteristic change law. The re-diagnosis combined with the historical fault data and the characteristic change law can more comprehensively understand the fault mode and cause, thereby improving the accuracy of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A unit block diagram of the present invention;

[0030] Figure 2 It is a block diagram of a unit module of the present invention;

[0031] Figure 3 is a block diagram of an abnormality diagnosis unit of the present invention;

[0032] Figure 4 It is a block diagram of the feature calculation unit of the present invention.

[0033] The meaning of each number in the figure is:

[0034] 1. Abnormal diagnosis unit; 11. Abnormal judgment module; 12. Prediction diagnosis module;

[0035] 2. Feature calculation unit; 21. Change rate module; 22. Data calculation module;

[0036] 3. Best time unit; 4. Fault repair unit. DETAILED DESCRIPTION

[0037] The following will be combined with the accompanying drawings in the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Example 1

[0039] The present invention provides an automatic fault diagnosis and repair system for electrical equipment based on intelligent control. Figure 1 , including an abnormality diagnosis unit 1, a feature calculation unit 2, an optimal timing unit 3, and a fault repair unit 4;

[0040] The abnormality diagnosis unit 1 obtains the operating status and historical data of the electrical equipment to calculate the pressure abnormality value of the electrical equipment, uses the calculated pressure abnormality value to determine whether the electrical equipment is abnormal, and extracts historical fault data and corresponding historical fault time from the fault library to predict the electrical equipment fault and the electrical equipment fault time;

[0041] The feature calculation unit 2 is used to receive the data in the abnormal diagnosis unit 1, and the feature calculation unit 2 takes the data from the abnormal diagnosis unit 1, and calculates the electrical equipment at time T according to the taken data. f -t k The characteristic change rate at the moment, and then calculate the mean and standard deviation of the characteristic change rate of the electrical equipment according to the characteristic change rate of the electrical equipment;

[0042] The best timing unit 3 is used to receive the data from the feature calculation unit 2 and the abnormality diagnosis unit 1, and use the arg max function according to the time T f -t k The electrical equipment characteristic change rate at the moment, the mean value and the standard deviation of the electrical equipment characteristic change rate are calculated to calculate the best time to collect the fault characteristic change law when automatically diagnosing electrical equipment faults;

[0043] The abnormality diagnosis unit 1 is used to receive the best time calculated by the best time unit 3, calculate the start time of collecting the fault characteristic change law, use the predicted electrical equipment fault time as the end time of collecting the fault characteristic change law, automatically diagnose the electrical equipment fault type according to the extracted historical fault data and the corresponding historical fault time, and collect the fault characteristic change law from the best time between the start time and the end time, and then diagnose the electrical equipment fault type again according to the extracted historical fault data, the corresponding historical fault time and the collected fault characteristic change law;

[0044] The fault repair unit 4 is used to receive data in the abnormal diagnosis unit 1, formulate an electrical equipment fault repair strategy according to the data in the abnormal diagnosis unit 1, and then perform fault repair on the diagnosed electrical equipment fault type according to the formulated repair strategy.

[0045] The following is a refinement of the above units, see Figure 2-Figure 4

[0046] The abnormality diagnosis unit 1 includes an abnormality judgment module 11 and a prediction diagnosis module 12

[0047] The abnormality judgment module 11 obtains the operation status data of the electrical equipment in real time through the intelligent sensor, and obtains the historical data, takes the actual pressure data according to the obtained operation status data, and then obtains the historical pressure data from the historical data, updates the actual pressure data X and the historical pressure data into a complete historical pressure data set WP, and then records the number of data n in the complete historical pressure data set, and uses the Z-score algorithm to calculate the abnormal pressure value of the electrical equipment according to the complete historical pressure data set WP and the number of data n. The complete historical pressure data set provides a sufficient sample size to facilitate more accurate identification of normal and abnormal pressure values, reduce missed detection and false alarms, and then use the calculated pressure abnormal value and the set pressure standard value to judge the abnormality. When the calculated pressure abnormal value is greater than the set pressure standard value, it indicates that the electrical equipment is abnormal. When the calculated pressure abnormal value is less than the set pressure standard value, the actual pressure of the electrical equipment is controlled and adjusted to adjust the actual pressure data X and the set pressure standard value;

[0048] Historical data includes historical pressure data;

[0049] The implementation steps of calculating outliers using the Z-score algorithm are:

[0050] Step ①, collect the complete historical pressure data set WP and the number of data n to calculate the pressure mean, and obtain the calculated pressure mean μ. The specific algorithm formula is:

[0051]

[0052] Among them, WP i Refers to the i-th pressure data in the complete historical pressure data set WP;

[0053] Step ②: According to the number of data n, the i-th pressure data WP in the complete historical pressure data set WP i The pressure standard deviation is calculated by adding the calculated pressure mean μ to obtain the calculated pressure standard deviation σ. The specific algorithm formula is:

[0054]

[0055] Step ③, use the actual pressure data x, the calculated pressure mean μ and the calculated pressure standard deviation σ to calculate the outlier value, and obtain the calculated pressure outlier value z. The specific algorithm formula is:

[0056]

[0057] Among them, this formula is used to calculate abnormal values. By calculating abnormal values, long-term or potential abnormal trends can be identified, and possible problems of equipment can be discovered early;

[0058] The prediction diagnosis module 12 is used to receive the command data indicating that the electrical equipment has an abnormality in the abnormality judgment module 11, and extract the historical fault data and the corresponding historical fault time from the fault library, and use the convolutional neural network to predict the electrical equipment fault and the electrical equipment fault time according to the extracted historical fault data and the corresponding historical fault time. By predicting the historical fault data, the potential mode and trend of the fault can be identified, which helps to accurately predict future faults;

[0059] The feature calculation unit 2 includes a change rate module 21 and a data calculation module 22;

[0060] The change rate module 21 is used to receive the electrical equipment fault command predicted by the prediction and diagnosis module 12. The change rate module 21 takes the acquired operating status data and the corresponding historical fault time in the prediction and diagnosis module 12 to the abnormality judgment module 11, and extracts the most recent fault occurrence time T from the corresponding historical fault time. f , according to the acquired running status data, extract the time when the most recent fault occurred T f At time t k The electrical equipment characteristic value S(T f -t k ) (referring to the operating parameters of current, voltage, and temperature), and record the number of time points m, and then use the time T f -t k The electrical equipment characteristic value S(T f -t k ) Calculate the electrical equipment at time T f -tk The characteristic change rate at the moment, and the time T f -t k The characteristic change rate of electrical equipment at the time R(T f -t k ), and simultaneously record the number N of characteristic change rates of electrical equipment. Calculating characteristic change rates helps to distinguish different types of faults, avoid confusion, and ensure more accurate fault classification and diagnosis;

[0061] Get time T f -t k The characteristic change rate of electrical equipment at the time R(T f -t k )’s specific algorithm formula:

[0062]

[0063] Among them, t k refers to the kth time point, Δt refers to a fixed time interval, which is used to calculate the length of the change in the characteristics of an electrical device between two different time points. This formula is used to calculate the change in the characteristics of an electrical device at time T f -t k The characteristic change rate at each moment is used to facilitate the subsequent calculation of the best time to collect the fault characteristic change law;

[0064] The data calculation module 22 is used to receive the electrical equipment characteristic change rate R and the number of electrical equipment characteristic change rates N in the change rate module 21, and calculate the mean value of the electrical equipment characteristic change rate according to the electrical equipment characteristic change rate R and the number of electrical equipment characteristic change rates N. Among them, R j Refers to the jth electrical equipment characteristic change rate R, and then according to the electrical equipment characteristic change rate R, the number of electrical equipment characteristic change rates N and the mean value of the electrical equipment characteristic change rate R μ Calculate the standard deviation of the rate of change of electrical equipment characteristics

[0065] The best timing unit 3 is used to receive the time T in the change rate module 21 f -t k The characteristic change rate of electrical equipment at the time R(T f -t k ), a fixed time interval Δt, the number of time points m and the mean value R of the rate of change of the electrical equipment characteristics in the data calculation module 22 μ , Standard deviation R of the rate of change of electrical equipment characteristics σ , receiving the operating status data obtained in the abnormality judgment module 11, and outputting the weight coefficient according to the obtained operating status data through the linear regression model Use the arg max function according to time T f-t k The characteristic change rate of electrical equipment at the time R(T f -t k ), fixed time interval Δt, number of time points m, mean value R of the characteristic change rate of electrical equipment μ , Standard deviation R of the rate of change of electrical equipment characteristics σ and weight coefficient Calculate the best time to collect the changing rules of fault characteristics when automatically diagnosing electrical equipment faults. Collecting data at the best time can ensure that the changing rules of key characteristics of electrical equipment faults are captured, reduce fault diagnosis errors, thereby speeding up the response speed of fault diagnosis and improving fault diagnosis efficiency.

[0066] The implementation principle of using the arg max function to calculate the best time:

[0067] According to the time T f -t k The characteristic change rate of electrical equipment at the time R(T f -t k ), fixed time interval Δt, number of time points m, mean value R of the characteristic change rate of electrical equipment μ , Standard deviation R of the rate of change of electrical equipment characteristics σ and weight coefficient Calculate the optimal time period for collecting the changing rules of fault characteristics when automatically diagnosing electrical equipment faults, and obtain the optimal time period Δt best That is the best time to calculate, the specific algorithm formula is:

[0068]

[0069] in, The value is the kth weight coefficient, It is a mathematical function used to find the fixed time interval Δt that maximizes the formula. The formula is used to calculate the optimal time period for collecting the changing rules of fault characteristics when automatically diagnosing electrical equipment faults. Selecting the optimal time period for collection can ensure that the key characteristics of the equipment fault are captured, avoid missing important information, and thus improve the accuracy of fault diagnosis.

[0070] The prediction diagnosis module 12 is used to receive the optimal time Δt calculated by the optimal time unit 3 best , according to the predicted electrical equipment failure time Y s and calculate the optimal time Δt best Calculate the start time of collecting fault feature change rules To=Y s -Δt best , the predicted electrical equipment failure time Y sThe end time Et for collecting the fault characteristic change law is used, and the start time To and the end time Et are input into the convolutional neural network. The convolutional neural network is used to automatically diagnose the fault type of the electrical equipment according to the extracted historical fault data and the corresponding historical fault time, and the fault characteristic change law is collected from the best time between the start time and the end time. According to the fault characteristic change law, it can be determined when it is most effective to repair the fault, thereby reducing the downtime of the electrical equipment. The convolutional neural network is then used to diagnose the fault type of the electrical equipment again according to the extracted historical fault data, the corresponding historical fault time and the collected fault characteristic change law. The re-diagnosis is performed in combination with the historical fault data and the characteristic change law, which can more comprehensively understand the mode and cause of the fault, thereby improving the accuracy of fault diagnosis. When the diagnosed electrical equipment fault type is the same as the re-diagnosed electrical equipment fault type, it means that the fault type at this time causes the electrical equipment to fail.

[0071] The fault repair unit 4 is used to receive the fault characteristic change law, the diagnosed electrical equipment fault type and the extracted historical fault data collected in the predictive diagnosis module 12, and use a convolutional neural network to formulate an electrical equipment fault repair strategy based on the collected fault characteristic change law and the extracted historical fault data. By formulating a repair strategy based on the collected fault characteristic change law, it is possible to more accurately target specific types of faults, reduce the probability of misdiagnosis and repair errors, and improve the accuracy of fault repair. The diagnosed electrical equipment fault type is then repaired according to the formulated repair strategy. With a clear repair strategy, the trial and error behavior in the repair process is reduced, thereby reducing the repair cost and time.

[0072] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. Automatic fault diagnosis and repair system for electrical equipment based on intelligent control, characterized by: It includes an abnormality diagnosis unit (1), a feature calculation unit (2), an optimal timing unit (3), and a fault repair unit (4); The abnormality diagnosis unit (1) obtains the operating status and historical data of the electrical equipment to calculate the pressure abnormality value of the electrical equipment, uses the calculated pressure abnormality value to determine whether the electrical equipment has an abnormality, and extracts historical fault data and corresponding historical fault time from the fault library to predict the fault of the electrical equipment and the fault time of the electrical equipment; The characteristic calculation unit (2) is used to receive data from the abnormality diagnosis unit (1), and the characteristic calculation unit (2) takes data from the abnormality diagnosis unit (1), and calculates the electrical equipment at time T according to the taken data. f -t k The characteristic change rate at the moment, and then calculate the mean and standard deviation of the characteristic change rate of the electrical equipment according to the characteristic change rate of the electrical equipment; The optimal timing unit (3) is used to receive data from the feature calculation unit (2) and the abnormality diagnosis unit (1), and calculate the optimal timing according to the time T f -t k The electrical equipment characteristic change rate at the moment, the mean value and the standard deviation of the electrical equipment characteristic change rate are calculated to calculate the best time to collect the fault characteristic change law when automatically diagnosing electrical equipment faults; The abnormality diagnosis unit (1) is used to receive the best time calculated by the best time unit (3), calculate the start time of collecting the fault characteristic change law, use the predicted electrical equipment fault time as the end time of collecting the fault characteristic change law, automatically diagnose the electrical equipment fault type according to the extracted historical fault data and the corresponding historical fault time, collect the fault characteristic change law from the best time between the start time and the end time, and then diagnose the electrical equipment fault type again according to the extracted historical fault data, the corresponding historical fault time and the collected fault characteristic change law; The fault repair unit (4) is used to receive data in the abnormal diagnosis unit (1), formulate an electrical equipment fault repair strategy based on the data in the abnormal diagnosis unit (1), and then perform fault repair on the diagnosed electrical equipment fault type according to the formulated repair strategy.

2. The automatic fault diagnosis and repair system for electrical equipment based on intelligent control according to claim 1 is characterized in that: The abnormality diagnosis unit (1) comprises an abnormality judgment module (11) and a prediction diagnosis module (12) The abnormality judgment module (11) obtains the running status data of the electrical equipment in real time through the intelligent sensor, and obtains the historical data, takes the actual pressure data according to the obtained running status data, and then obtains the historical pressure data from the historical data, updates the actual pressure data X and the historical pressure data into a complete historical pressure data set WP, and then records the number of data n in the complete historical pressure data set, and calculates the abnormal pressure value of the electrical equipment according to the complete historical pressure data set WP and the number of data n by using the Z-score algorithm, and then judges the abnormality by using the calculated abnormal pressure value, and when the calculated abnormal pressure value is greater than the set pressure standard value, it indicates that the electrical equipment has an abnormality; The prediction diagnosis module (12) is used to receive command data indicating that an abnormality exists in the electrical equipment in the abnormality judgment module (11), and to extract historical fault data and corresponding historical fault times from a fault library, and to use a convolutional neural network to predict electrical equipment faults and electrical equipment fault times based on the extracted historical fault data and corresponding historical fault times.

3. The electrical equipment automatic fault diagnosis and repair system based on intelligent control according to claim 2 is characterized in that: The feature calculation unit (2) comprises a change rate module (21) and a data calculation module (22); The change rate module (21) is used to receive the electrical equipment fault command predicted by the prediction diagnosis module (12), and the change rate module (21) takes the acquired operating status data and the corresponding historical fault time in the prediction diagnosis module (12) into the abnormality judgment module (11), and extracts the most recent fault occurrence time T from the corresponding historical fault time. f , according to the acquired running status data, extract the time when the most recent fault occurred T f At time t k The electrical equipment characteristic value S(T f -t k ), and record the number of time points m, and then use time T f -t k The electrical equipment characteristic value S(T f -t k ) Calculate the electrical equipment at time T f -t k The characteristic change rate at the moment, and the time T f -t k The characteristic change rate of electrical equipment at the time R(T f -t k ), and simultaneously record the number N of characteristic change rates of electrical equipment; Get time T f -t k The characteristic change rate of electrical equipment at the time R(T f -t k )’s specific algorithm formula: Among them, t k refers to the kth time point, and Δt refers to a fixed time interval, which is used to calculate the length of the change in the characteristics of an electrical device between two different time points.

4. The automatic fault diagnosis and repair system for electrical equipment based on intelligent control according to claim 3 is characterized in that: The data calculation module (22) is used to receive the electrical equipment characteristic change rate R and the number N of electrical equipment characteristic change rates in the change rate module (21), and calculate the mean value of the electrical equipment characteristic change rate according to the electrical equipment characteristic change rate R and the number N of electrical equipment characteristic change rates. Among them, R j Refers to the jth electrical equipment characteristic change rate R, and then according to the electrical equipment characteristic change rate R, the number of electrical equipment characteristic change rates N and the mean value of the electrical equipment characteristic change rate R μ Calculate the standard deviation of the rate of change of electrical equipment characteristics 5. The automatic fault diagnosis and repair system for electrical equipment based on intelligent control according to claim 4 is characterized in that: The optimal timing unit (3) is used to receive data from the change rate module (21) and the data calculation module (22), receive the operating status data obtained from the abnormality judgment module (11), and output a weight coefficient according to the obtained operating status data through a linear regression model. Use the arg max function according to time T f -t k The characteristic change rate of electrical equipment at the time R(T f -t k ), fixed time interval Δt, number of time points m, mean value R of the rate of change of electrical equipment characteristics μ , Standard deviation R of the rate of change of electrical equipment characteristics σ and weight coefficient Calculate the best time to collect the changing rules of fault characteristics when automatically diagnosing electrical equipment faults.

6. The electrical equipment automatic fault diagnosis and repair system based on intelligent control according to claim 5 is characterized in that: The implementation principle of using the arg max function to calculate the best time: According to the time T f -t k The characteristic change rate of electrical equipment at the time R(T f -t k ), fixed time interval Δt, number of time points m, mean value R of the rate of change of electrical equipment characteristics μ , Standard deviation R of the rate of change of electrical equipment characteristics σ and weight coefficient Calculate the optimal time period for collecting the changing rules of fault characteristics when automatically diagnosing electrical equipment faults, and obtain the optimal time period Δt best That is the best time to calculate, the specific algorithm formula is: in, The value is the kth weight coefficient, arg is a mathematical function that is used to find the fixed time interval Δt that maximizes the value in the formula.

7. The automatic fault diagnosis and repair system for electrical equipment based on intelligent control according to claim 5 is characterized in that: The prediction diagnosis module (12) is used to receive the optimal timing Δt calculated in the optimal timing unit (3) best , calculate the start time of collecting fault feature change rules To = Y s -Δt best , the predicted electrical equipment failure time Y s The start time To and the end time Et are used as the end time Et for collecting the changing law of fault characteristics, and then the convolutional neural network is used to automatically diagnose the fault type of the electrical equipment according to the extracted historical fault data and the corresponding historical fault time. The fault characteristic changing law is collected from the best time between the start time and the end time, and then the convolutional neural network is used to diagnose the fault type of the electrical equipment again according to the extracted historical fault data, the corresponding historical fault time and the collected fault characteristic changing law.

8. The electrical equipment automatic fault diagnosis and repair system based on intelligent control according to claim 2 is characterized by: The fault repair unit (4) is used to receive the fault feature change law, the diagnosed electrical equipment fault type and the extracted historical fault data collected in the prediction diagnosis module (12), formulate an electrical equipment fault repair strategy based on the collected fault feature change law and the extracted historical fault data using a convolutional neural network, and then perform fault repair on the diagnosed electrical equipment fault type based on the formulated repair strategy.

Images