A method for monitoring faults in electrical control

Abstract

The present invention discloses a fault monitoring method for electrical control, comprising the following steps: collecting statistics on fault types and device electrical data types of the electrical control system; analyzing the monitoring requirements of the fault types and device electrical data types to obtain long-term and short-term monitoring requirements of the fault types for the device electrical data types; constructing a dynamic vector based on the long-term and short-term monitoring requirements of the fault types for the device electrical data types, and generating real-time monitoring requirements of the fault types for the device electrical data types based on the dynamic vector. The present invention constructs a dynamic vector based on the long-term and short-term monitoring requirements of the fault types for the device electrical data types, and generates real-time monitoring requirements of the fault types for the device electrical data types based on the dynamic vector, thereby achieving dynamic monitoring of the device electrical data types, improving the monitoring accuracy of the fault types, and updating the monitoring field of the fault types.

Description

Technical Field

[0001] The present invention relates to the technical field of fault monitoring, and in particular to a fault monitoring method for electrical control. Background Art

[0002] In industrial production, electrical control systems experience varying degrees of wear, fatigue, deformation, or damage after prolonged use. This can lead to failures, impacting product yield and production efficiency. To ensure product yield and efficiency, fault monitoring of production equipment is essential. Traditional electrical control system fault monitoring methods, due to their inherent limitations, suffer from the following issues: First, they generally rely on empirical data, which can affect the results and lead to low accuracy. Second, traditional electrical control system fault monitoring methods generally utilize machine learning algorithms for data processing. These algorithms select characteristic data from system operating data (electrical data from various components) to identify fault types. However, this characteristic data becomes rigid during subsequent fault monitoring, resulting in a fixed monitoring field for each fault type, making it difficult to ensure accurate fault monitoring. Summary of the Invention

[0003] The purpose of the present invention is to provide an electrically controlled fault monitoring method to solve the technical problem in the prior art that the type of characteristic data of multi-layer materials is difficult to solidify during the fault monitoring process, resulting in the solidification of the monitoring field of the fault type, making it difficult to ensure the accuracy of fault monitoring.

[0004] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0005] A method for monitoring faults in an electrical control system comprises the following steps:

[0006] Collect statistics on fault types and device electrical data types of electrical control systems;

[0007] Analyze the monitoring requirements of fault types and device electrical data types to obtain the long-term and short-term monitoring requirements of fault types for device electrical data types;

[0008] A dynamic vector is constructed based on the long-term and short-term monitoring requirements of the fault type for the device electrical data type, and the real-time monitoring requirements of the fault type for the device electrical data type are generated based on the dynamic vector, so as to realize dynamic monitoring of the device electrical data type to improve the monitoring accuracy of the fault type and update the monitoring field of the fault type.

[0009] As a preferred solution of the present invention, the statistical electrical control system fault type and device electrical data type include:

[0010] The fault types of the electrical control system are counted based on the historical operation log of the electrical control system, and the electrical data type of each component in the electrical control system is counted as the component electrical data type.

[0011] As a preferred solution of the present invention, obtaining the long-term monitoring requirement of the device electrical data type for the fault type includes:

[0012] A Gaussian mixture model is used to perform cluster analysis based on fault type and device electrical data type to obtain the long-term monitoring requirements of each fault type for the device electrical data type.

[0013] Among them, the cluster analysis function expression of long-term monitoring needs is:

[0014] ;

[0015] ;

[0016] Where Z i A is the clustering result of the device electrical data type required for long-term monitoring in the i-th fault type, k,i is the mixing weight of the Gaussian density function corresponding to the i-th fault type, representing The significance of , is the Gaussian density function corresponding to the i-th fault type, , is the cluster center corresponding to the i-th fault type, representing the k-th device electrical data type required for long-term monitoring in the i-th fault type. is the degree of aggregation of the electrical data type of the kth device required for long-term monitoring in the i-th fault type, for Parameters, y is the identifier of the device electrical data type, m is the total number of device electrical data types, k and i are both counting variables.

[0017] As a preferred solution of the present invention, obtaining the short-term monitoring requirement of the fault type on the device electrical data type includes:

[0018] Group all device electrical data types according to fault types;

[0019] All device electrical data types of each fault type are used as input items of the CapsulesNet algorithm. The CapsulesNet algorithm uses the negotiation routing algorithm to predict all device electrical data types to obtain the short-term monitoring requirements of each fault type for the device electrical data type.

[0020] Among them, the loss function of the CapsulesNet algorithm is quantified using MSE error.

[0021] As a preferred solution of the present invention, the construction of the dynamic vector includes:

[0022] The extensiveness of the fault type monitoring requirement is quantified by the long-term monitoring requirement of the device electrical data type based on the fault type. The quantification function of the extensiveness is:

[0023] S i =f s (K i );

[0024] Where S i is the prevalence of the i-th fault type, K i is the total number of device electrical data types required for long-term monitoring in the i-th fault type, f s K i The normalized function operator, f S To K i Limited to (0,lim S ) range, lim S For S i The limiting operator, i is a counting variable;

[0025] The long-term monitoring requirement of the fault type to the device electrical data type is used to quantify the familiarity of the fault type monitoring requirement. The quantification function of the familiarity is:

[0026] F i =f F (|{Z i |i∈[1,p]}|);

[0027] Where, F i is the familiarity of the i-th fault type, Z i is the clustering result of the device electrical data type required for long-term monitoring in the i-th fault type, f F For |{Z i The normalized function operator, f F To Z i Limited to (0,lim F ) range, lim F F iThe limiting operator, |{Z i |i∈[1,p]}| is the modulus of the vector formed by the clustering results of the device electrical data types required for long-term monitoring among all fault types, p is the total number of fault types, and i is a counting variable;

[0028] Based on the extensiveness and familiarity, the vector modulus of the dynamic vector is determined. The function expression of the vector modulus is:

[0029] |E i |=f(S i ,F i );

[0030] Where, |E i | is the vector modulus of the dynamic vector of the i-th fault type, f is the positive correlation operator, S i is the prevalence of the i-th fault type, F i is the familiarity of the i-th fault type, i is a counting variable;

[0031] The vector direction of the dynamic vector is determined by using the long-term monitoring requirement of the fault type on the device electrical data type and the short-term monitoring requirement of the fault type on the device electrical data type;

[0032] The function expression of the vector direction is:

[0033] ;

[0034] Where, E i is the dynamic vector of the i-th fault type, |E i | is the vector modulus length of the dynamic vector of the i-th fault type, is the vector direction of the dynamic vector of the i-th fault type, is a vector of the long-term monitoring requirements for device electrical data types based on the fault type. is a vector of the short-term monitoring requirements of the device electrical data type for each fault type, m is the total number of device electrical data types, n is the total number of device electrical data types that require long-term monitoring in the i-th fault type, and i, j, and k are counting variables.

[0035] As a preferred solution of the present invention, the real-time monitoring requirement of the device electrical data type based on the dynamic vector generation fault type includes:

[0036] Determine the extraction area of ​​the real-time monitoring requirement of the fault type in the fault type versus device electrical data type based on the vector modulus and vector direction of the dynamic vector;

[0037] Determine the sampling rate required for real-time monitoring of the device electrical data type based on the fault type. The functional expression of the sampling rate is:

[0038] ;

[0039] Where, L i is the sampling rate required for real-time monitoring of the i-th fault type, T i is the total number of device electrical data types required for long-term monitoring in the i-th fault type, p is the total number of fault types, and i is a counting variable;

[0040] Directional sampling is performed in the extraction area based on the sampling rate to obtain device electrical data types required for real-time monitoring of fault types.

[0041] As a preferred solution of the present invention, the device electrical data type is normalized before the monitoring demand analysis.

[0042] As a preferred solution of the present invention, the real-time fault monitoring requirement of the fault type is utilized to perform real-time fault monitoring on the electrical control system.

[0043] As a preferred solution of the present invention, the historical operation log of the electrical control system is regularly updated during the fault monitoring process of the electrical control system to achieve dynamic update of the real-time monitoring requirements of the fault type.

[0044] As a preferred solution of the present invention, universality and familiarity are normalized.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] The present invention analyzes the monitoring requirements of fault types and device electrical data types to obtain long-term monitoring requirements and short-term monitoring requirements of fault types for device electrical data types, constructs a dynamic vector based on the long-term monitoring requirements and short-term monitoring requirements of fault types for device electrical data types, and generates real-time monitoring requirements of fault types for device electrical data types based on the dynamic vector, so as to realize dynamic monitoring of device electrical data types to improve the monitoring accuracy of fault types and update the monitoring field of fault types. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other implementation drawings based on the provided drawings without inventive effort.

[0048] Figure 1 This is a flow chart of a fault monitoring method for electrical control provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0050] Traditional electrical control system fault monitoring methods, due to their inherent limitations, suffer from the following problems: First, they generally rely on empirical data during the monitoring process, which results in monitoring results that are subject to subjective factors, resulting in low monitoring accuracy. Second, traditional electrical control system fault monitoring methods generally use machine learning algorithms for data processing during the monitoring process. Feature data used to identify fault types is selected from system operating data (various device electrical data) for machine learning algorithm fault identification. However, during subsequent fault monitoring, the feature data types become fixed, resulting in a fixed monitoring field for each fault type, making it difficult to ensure fault monitoring accuracy. Therefore, the present invention provides an electrical control fault monitoring method that analyzes monitoring requirements based on fault type, deriving long-term and short-term monitoring requirements for each fault type to ensure accuracy and comprehensiveness of fault monitoring, respectively. A dynamic vector is then constructed to generate real-time monitoring requirements for each fault type with respect to the device electrical data type. This allows for dynamic monitoring of the device electrical data type, improving fault type monitoring accuracy and updating the monitoring field of each fault type.

[0051] like Figure 1 As shown, the present invention provides a fault monitoring method for electrical control, comprising the following steps:

[0052] Collect statistics on fault types and device electrical data types of electrical control systems;

[0053] Analyze the monitoring requirements of fault types and device electrical data types to obtain the long-term and short-term monitoring requirements of fault types for device electrical data types;

[0054] A dynamic vector is constructed based on the long-term and short-term monitoring requirements of the fault type for the device electrical data type, and the real-time monitoring requirements of the fault type for the device electrical data type are generated based on the dynamic vector, so as to realize dynamic monitoring of the device electrical data type to improve the monitoring accuracy of the fault type and update the monitoring field of the fault type.

[0055] Statistics on fault types and device electrical data types of electrical control systems, including:

[0056] The fault types of the electrical control system are counted based on the historical operation log of the electrical control system, and the electrical data type of each component in the electrical control system is counted as the component electrical data type.

[0057] Due to the increase in the operating time of the electrical control system, the fault types of the electrical control system will continue to repeat or new fault types will be generated. Therefore, the present invention uses the empirical data of the historical operation log to analyze the fault type monitoring needs, wherein the monitoring needs include long-term needs and short-term needs. The long-term needs are long-term monitoring needs, that is, the electrical data types of various devices that can ensure that the fault types are stably and correctly identified in the long term, and are the electrical data types of various devices that can be used stably for fault type identification in the long term, indicating a long-term fixed monitoring field of view. The short-term needs are short-term monitoring needs, that is, the electrical data types of various devices that can ensure that the fault types are randomly and correctly identified in the short term, and are the electrical data types of various devices that can be used for fault type identification in the short term, indicating a short-term random monitoring field of view.

[0058] Long-term monitoring requirements reflect the stable and accurate nature of fault monitoring, but the monitoring field of view is fixed. Short-term monitoring requirements reflect the extended field of view of fault monitoring, but the stability and accuracy are limited. Therefore, obtaining both long-term and short-term monitoring requirements can complement each other in fault monitoring. The details are as follows:

[0059] The long-term monitoring requirements for device electrical data types based on the fault type include:

[0060] A Gaussian mixture model is used to perform cluster analysis based on fault type and device electrical data type to obtain the long-term monitoring requirements of each fault type for the device electrical data type.

[0061] Among them, the cluster analysis function expression of long-term monitoring needs is:

[0062] ;

[0063] ;

[0064] Where Z i A is the clustering result of the device electrical data type required for long-term monitoring in the i-th fault type, k,i is the mixing weight of the Gaussian density function corresponding to the i-th fault type, representing The significance of , is the Gaussian density function corresponding to the i-th fault type, , is the cluster center corresponding to the i-th fault type, representing the k-th device electrical data type required for long-term monitoring in the i-th fault type. is the degree of aggregation of the electrical data type of the kth device required for long-term monitoring in the i-th fault type, for Parameters, y is the identifier of the device electrical data type, m is the total number of device electrical data types, k and i are both counting variables.

[0065] The short-term monitoring requirements for device electrical data types based on the fault type include:

[0066] Group all device electrical data types according to fault types;

[0067] All device electrical data types of each fault type are used as input items of the CapsulesNet algorithm. The CapsulesNet algorithm uses the negotiation routing algorithm to predict all device electrical data types to obtain the short-term monitoring requirements of each fault type for the device electrical data type.

[0068] Among them, the loss function of the CapsulesNet algorithm is quantified using MSE error.

[0069] Due to the complementary advantages of long-term and short-term monitoring requirements, the present invention constructs a dynamic vector, which can be used to extract the real-time monitoring requirements of the fault type for the device electrical data type in the long-term and short-term monitoring requirements. The real-time monitoring requirements can complement the advantages of long-term and short-term monitoring requirements. While inheriting the advantages of long-term monitoring requirements in fault identification stability and accuracy for repetitive faults during the fault monitoring process, it can broaden the monitoring field of fault identification and further discover new fault types that appear during the fault monitoring process, that is, further improve the fault identification accuracy during the fault monitoring process, as follows:

[0070] The construction of dynamic vector includes:

[0071] The extensiveness of the fault type monitoring demand is quantified by the long-term monitoring demand for device electrical data types based on the fault type. The quantification function of the extensiveness is:

[0072] S i =f s (K i );

[0073] Where S i is the prevalence of the i-th fault type, K i is the total number of device electrical data types required for long-term monitoring in the i-th fault type, f sK i The normalized function operator, f S To K i Limited to (0,lim S ) range, lim S For S i The limiting operator, i is a counting variable;

[0074] The long-term monitoring requirements of device electrical data types based on fault types are used to quantify the familiarity of fault type monitoring requirements. The quantification function of familiarity is:

[0075] F i =f F (|{Z i |i∈[1,p]}|);

[0076] Where, F i is the familiarity of the i-th fault type, Z i is the clustering result of the device electrical data type required for long-term monitoring in the i-th fault type, f F For |{Z i The normalized function operator, f F To Z i Limited to (0,lim F ) range, lim F F i The limiting operator, |{Z i |i∈[1,p]}| is the modulus of the vector formed by the clustering results of the device electrical data types required for long-term monitoring among all fault types, p is the total number of fault types, and i is a counting variable;

[0077] Based on the extensiveness and familiarity, the vector modulus of the dynamic vector is determined. The function expression of the vector modulus is:

[0078] |E i |=f(S i ,F i );

[0079] Where, |E i | is the vector modulus of the dynamic vector of the i-th fault type, f is the positive correlation operator, S i is the prevalence of the i-th fault type, F i is the familiarity of the i-th fault type, i is a counting variable;

[0080] The vector direction of the dynamic vector is determined by using the long-term monitoring requirement of the fault type on the device electrical data type and the short-term monitoring requirement of the fault type on the device electrical data type;

[0081] The function expression of the vector direction is:

[0082] ;

[0083] Where, E i is the dynamic vector of the i-th fault type, |E i | is the vector modulus length of the dynamic vector of the i-th fault type, is the vector direction of the dynamic vector of the i-th fault type, is a vector of the long-term monitoring requirements for device electrical data types based on the fault type. is a vector of the short-term monitoring requirements of the device electrical data type for each fault type, m is the total number of device electrical data types, n is the total number of device electrical data types that require long-term monitoring in the i-th fault type, and i, j, and k are counting variables.

[0084] Based on the extensiveness and familiarity, the vector modulus of the dynamic vector is determined. Extensiveness indicates that the obtained real-time monitoring needs explore the diversity of device electrical data types on the basis of long-term monitoring needs, and can approach the device electrical data types of short-term monitoring needs, broadening the field of vision of monitoring needs. Familiarity indicates that the obtained real-time monitoring needs remain familiar with historical logs on the basis of long-term monitoring needs, that is, they maintain sensitivity to the identification of repeated faults, and maintain the stability and accuracy advantages of fault identification of repeated faults. The long-term monitoring needs of the fault type for the device electrical data type and the short-term monitoring needs of the fault type for the device electrical data type are used to determine the vector direction of the dynamic vector, and the direction of the real-time monitoring needs is directed from the short-term monitoring needs to the long-term monitoring needs. The monitoring field of vision can be broadened on the basis of maintaining long-term stability and accuracy, and the obtained real-time monitoring needs can reasonably integrate the advantages of short-term monitoring needs and long-term monitoring needs, so that the dynamic monitoring of the device electrical data type in the fault monitoring process is rationalized, thereby ensuring the improvement of the monitoring accuracy of the fault type and the rationalization of the monitoring field of vision of the updated fault type.

[0085] Real-time monitoring requirements for device electrical data types based on dynamic vector generation fault types, including:

[0086] Determine the extraction area of ​​the real-time monitoring requirement of the fault type in the fault type versus device electrical data type based on the vector modulus and vector direction of the dynamic vector;

[0087] Determine the sampling rate required for real-time monitoring of the device electrical data type based on the fault type. The function expression of the sampling rate is:

[0088] ;

[0089] Where, L iis the sampling rate required for real-time monitoring of the i-th fault type, T i is the total number of device electrical data types required for long-term monitoring in the i-th fault type, p is the total number of fault types, and i is a counting variable;

[0090] Directed sampling is performed in the extraction area based on the sampling rate to obtain the device electrical data type required for real-time monitoring of the fault type.

[0091] Normalize the device electrical data type before monitoring demand analysis.

[0092] Real-time fault monitoring of electrical control systems is performed using the real-time monitoring requirements of fault types.

[0093] The historical operation log of the electrical control system is regularly updated during the fault monitoring process of the electrical control system to achieve dynamic updates of the real-time monitoring requirements of the fault types.

[0094] Normalize for prevalence and familiarity.

[0095] The present invention analyzes the monitoring requirements of fault types and device electrical data types to obtain long-term monitoring requirements and short-term monitoring requirements of fault types for device electrical data types, constructs a dynamic vector based on the long-term monitoring requirements and short-term monitoring requirements of fault types for device electrical data types, and generates real-time monitoring requirements of fault types for device electrical data types based on the dynamic vector, so as to realize dynamic monitoring of device electrical data types to improve the monitoring accuracy of fault types and update the monitoring field of fault types.

[0096] The above embodiments are merely exemplary embodiments of the present application and are not intended to limit the scope of the present application. The scope of protection of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and scope of protection of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the scope of protection of the present application.

Claims

1. A method for monitoring faults in electrical control, characterized in that: The following steps are involved: Collect statistics on fault types and device electrical data types of electrical control systems; Analyze the monitoring requirements of fault types and device electrical data types to obtain the long-term and short-term monitoring requirements of fault types for device electrical data types; Based on the long-term and short-term monitoring requirements of the fault type for the device electrical data type, a dynamic vector is constructed. Based on the dynamic vector, the real-time monitoring requirements of the fault type for the device electrical data type are generated to achieve dynamic monitoring of the device electrical data type, thereby improving the monitoring accuracy of the fault type and updating the monitoring field of the fault type. Real-time monitoring requirements for device electrical data types based on dynamic vector generation fault types, including: Determine the extraction area of ​​the real-time monitoring requirement of the fault type in the fault type versus device electrical data type based on the vector modulus and vector direction of the dynamic vector; Determine the sampling rate required for real-time monitoring of the device electrical data type based on the fault type. The function expression of the sampling rate is: ； Where, L i is the sampling rate required for real-time monitoring of the i-th fault type, T i is the total number of device electrical data types required for long-term monitoring in the i-th fault type, p is the total number of fault types, and i is a counting variable; Directed sampling is performed in the extraction area based on the sampling rate to obtain the device electrical data type required for real-time monitoring of the fault type.

2. The electrical control fault monitoring method according to claim 1, characterized in that: The statistical electrical control system fault types and device electrical data types include: The fault types of the electrical control system are counted based on the historical operation log of the electrical control system, and the electrical data type of each component in the electrical control system is counted as the component electrical data type.

3. The electrical control fault monitoring method according to claim 1, characterized in that: Obtaining the long-term monitoring requirements of the device electrical data type for the fault type includes: A Gaussian mixture model is used to perform cluster analysis based on fault type and device electrical data type to obtain the long-term monitoring requirements of each fault type for the device electrical data type. Among them, the cluster analysis function expression of long-term monitoring needs is: ； ； Where Z i A is the clustering result of the device electrical data type required for long-term monitoring in the i-th fault type, k,i is the mixing weight of the Gaussian density function corresponding to the i-th fault type, representing The significance of , is the Gaussian density function corresponding to the i-th fault type, , is the cluster center corresponding to the i-th fault type, representing the k-th device electrical data type required for long-term monitoring in the i-th fault type. is the degree of aggregation of the electrical data type of the kth device required for long-term monitoring in the i-th fault type, for Parameters, y is the identifier of the device electrical data type, m is the total number of device electrical data types, k and i are both counting variables.

4. The electrical control fault monitoring method according to claim 3, characterized in that: Obtaining the short-term monitoring requirements of the device electrical data type for the fault type includes: Group all device electrical data types according to fault types; All device electrical data types of each fault type are used as input items of the CapsulesNet algorithm. The CapsulesNet algorithm uses the negotiation routing algorithm to predict all device electrical data types to obtain the short-term monitoring requirements of each fault type for the device electrical data type. Among them, the loss function of the CapsulesNet algorithm is quantified using MSE error.

5. The electrical control fault monitoring method according to claim 4, characterized in that: The construction of the dynamic vector includes: The extensiveness of the fault type monitoring requirement is quantified by the long-term monitoring requirement of the device electrical data type based on the fault type. The quantification function of the extensiveness is: S i =f s (K i ); Where S i is the prevalence of the i-th fault type, K i is the total number of device electrical data types required for long-term monitoring in the i-th fault type, f s K i The normalized function operator, f S To K i Limited to (0,lim S ) range, lim S For S i The limiting operator, i is a counting variable; The long-term monitoring requirement of the fault type to the device electrical data type is used to quantify the familiarity of the fault type monitoring requirement. The quantification function of the familiarity is: F i =f F (|{Z i |i∈[1,p]}|); Where, F i is the familiarity of the i-th fault type, Z i is the clustering result of the device electrical data type required for long-term monitoring in the i-th fault type, f F For |{Z i The normalized function operator, f F To Z i Limited to (0,lim F ) range, lim F F i The limiting operator, |{Z i |i∈[1,p]}| is the modulus of the vector formed by the clustering results of the device electrical data types required for long-term monitoring among all fault types, p is the total number of fault types, and i is a counting variable; Based on the extensiveness and familiarity, the vector modulus of the dynamic vector is determined. The function expression of the vector modulus is: |E i |=f(S i ,F i ); Where, |E i | is the vector modulus of the dynamic vector of the i-th fault type, f is the positive correlation operator, S i is the prevalence of the i-th fault type, F i is the familiarity of the i-th fault type, i is a counting variable; Determine the vector direction of the dynamic vector by using the long-term monitoring requirement of the fault type on the device electrical data type and the short-term monitoring requirement of the fault type on the device electrical data type; The function expression of the vector direction is: ； Where, E i is the dynamic vector of the i-th fault type, |E i | is the vector modulus length of the dynamic vector of the i-th fault type, is the vector direction of the dynamic vector of the i-th fault type, is a vector of the long-term monitoring requirements for device electrical data types based on the fault type. is a vector of the short-term monitoring requirements of the device electrical data type for each fault type, m is the total number of device electrical data types, n is the total number of device electrical data types that require long-term monitoring in the i-th fault type, and i, j, and k are counting variables.

6. The electrical control fault monitoring method according to claim 1, characterized in that: Normalize the device electrical data type before monitoring demand analysis.

7. The electrical control fault monitoring method according to claim 1, characterized in that: The real-time monitoring requirement of the fault type is utilized to perform real-time fault monitoring on the electrical control system.

8. The electrical control fault monitoring method according to claim 1, characterized in that: The historical operation log of the electrical control system is regularly updated during the fault monitoring process of the electrical control system to achieve dynamic update of the real-time monitoring requirements of the fault type.

9. The electrical control fault monitoring method according to claim 5, characterized in that: Normalize for prevalence and familiarity.

Images