An intelligent equipment fault diagnosis system and method based on energy consumption analysis

Abstract

The present invention discloses an intelligent equipment fault diagnosis system and method based on energy consumption analysis, and relates to the technical field of intelligent equipment fault diagnosis. The present invention includes an energy consumption analysis module, an intelligent fault diagnosis analysis module, a diagnosis analysis determination module, and an equipment operation and maintenance processing module; the energy consumption analysis module is used to analyze the real-time data processing capability of a smart meter and construct a data processing energy consumption model of the smart meter; the fault diagnosis analysis module is used to diagnose and analyze the real-time fault conditions of the smart meter. Based on the continuity of the abnormal energy consumption data collection time and the discrete distribution of the abnormal energy consumption data, the present invention determines from multiple dimensions whether each module of the smart meter is in a fault state, thereby avoiding judging the temporary reduction in module performance of the smart meter due to voltage undervoltage and other reasons as a performance fault of the smart meter, and improving the accuracy of the fault diagnosis results.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment fault intelligent diagnosis, and in particular to an equipment fault intelligent diagnosis system and method based on energy consumption analysis. Background Art

[0002] Smart meters are one of the basic devices for data collection in smart grids (especially smart distribution networks). They are responsible for collecting, measuring and transmitting raw electric energy data, and are the basis for information integration, analysis optimization and information display. In addition to the basic electricity consumption metering function of traditional electric energy meters, smart meters also have intelligent functions such as two-way multi-rate metering, user-side control, two-way data communication in multiple data transmission modes, and anti-electricity theft, which makes smart meters suitable for smart grids and new energy fields.

[0003] The complexity of the operating environment of smart meters and the diversity of fault types require the collection of multi-dimensional data to analyze faults when conducting fault diagnosis and analysis on smart meters. This process requires high data processing capabilities and increases the data processing workload. At the same time, traditional fault handling methods are difficult to ensure the accuracy of fault diagnosis results, and thus cannot ensure the safe, stable and reliable operation of smart meters. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent equipment fault diagnosis system and method based on energy consumption analysis to solve the problems raised in the prior art.

[0005] To achieve the above-mentioned object, the present invention provides the following technical solutions: an intelligent equipment fault diagnosis system based on energy consumption analysis, the system comprising an energy consumption analysis module, a fault intelligent diagnosis analysis module, a diagnosis analysis determination module and an equipment operation and maintenance processing module;

[0006] The energy consumption analysis module is used to analyze the real-time data processing capability of the smart meter according to the energy consumption data of each module of the smart meter in the standby state and the running state, and to build a data processing energy consumption model of the smart meter;

[0007] The fault diagnosis and analysis module is used to calculate the real-time deviation capability index of the smart meter according to the constructed data processing energy consumption model, and to diagnose and analyze the real-time fault conditions of the smart meter according to the calculation results, and to retrain the constructed data processing energy consumption model;

[0008] When the energy consumption of the smart meter is abnormal, the diagnosis analysis and determination module calculates the real-time edge density of each module of the smart meter based on the abnormal energy consumption data, and analyzes and determines the real-time fault type of the smart meter according to the calculation result;

[0009] The equipment operation and maintenance processing module is used to transmit the fault type of the smart meter determined by the diagnosis analysis determination module to the equipment maintenance terminal, and the equipment maintenance terminal dispatches operation and maintenance personnel to perform maintenance and management on the smart meter according to the received fault type.

[0010] Furthermore, the energy consumption analysis module includes an energy consumption data acquisition unit, a data processing capability analysis unit and a data processing energy consumption model construction unit;

[0011] The energy consumption data collection unit collects energy consumption data of the display module, communication module, and data processing module of the smart meter in a standby state and an operating state, respectively. The standby state refers to a state when the smart meter does not perform actual power consumption operations. In this state, the display module, communication module, and data processing module of the smart meter are not in a working state. The operating state refers to a state in which at least one of the display module, communication module, and data processing module of the smart meter is in a working state.

[0012] The data processing capability analysis unit receives the energy consumption data transmitted by the energy consumption data collection unit, analyzes the energy consumed by the smart meter under different modules based on the received information, and predicts the real-time data processing capability of the smart meter based on the analysis result;

[0013] The data processing energy consumption model construction unit receives the energy consumption data transmitted by the energy consumption data acquisition unit, and constructs the data processing energy consumption model based on the received information.

[0014] Furthermore, the specific method for the data processing capability analysis unit to predict the real-time data processing capability of the smart meter is:

[0015] The energy consumption value collected by the smart meter in the standby state is recorded as the basic energy consumption value f, the energy consumption value collected by the smart meter when only the display module is in the working state is recorded as the first energy consumption value g, the energy consumption value collected by the smart meter when the display module and the communication module are both in the working state is recorded as the second energy consumption value h, and the energy consumption value collected by the smart meter when the display module, the communication module and the data processing module are all in the working state is recorded as the third energy consumption value k;

[0016] According to the collected first energy consumption value, second energy consumption value, and third energy consumption value, the real-time data processing capability of the smart meter is predicted. The specific prediction formula is:

[0017] ;

[0018] Wherein, j=1,2,…,n, which means that each collection time point of energy consumption data is numbered in chronological order, n represents the total number of times energy consumption data is collected, d represents the collection interval of energy consumption data, u represents the initial collection time of energy consumption data, e represents the natural constant and e=2.72, k (j-1)*d+u represents the third energy consumption value collected by the smart meter at time (j-1)*d+u, h (j-1)*d+u represents the second energy consumption value collected by the smart meter at time (j-1)*d+u, g (j-1)*d+u represents the first energy consumption value collected by the smart meter at time (j-1)*d+u, R (j-1)*d+u Represents the data processing capability coefficient of the smart meter at time (j-1)*d+u.

[0019] The real-time data processing capability of smart meters is predicted based on their real-time energy consumption data. This process does not require the acquisition of the real-time data collection and processing volume of the smart meters, which further improves the data processing efficiency of the system and enables effective diagnosis of smart meter faults.

[0020] Furthermore, the specific method of the data processing energy consumption model construction unit to construct the data processing energy consumption model is:

[0021] Take the dataset {(h u -g u ,k u -h u ),…,(h (j-1)*d+u -g (j-1)*d+u ,k (j-1)*d+u -h (j-1)*d+u )} as the training data set, train the linear model Y=a*X+b, and obtain the data processing energy consumption model Q (j-1)*d+u =a 1 *(h (j-1)*d+u -g (j-1)*d+u )+b 1 ;

[0022] Among them, a 1 represents the weight of the data processing energy consumption model, b 1 represents the bias of the data processing energy consumption model, Q (j-1)*d+u =k (j-1)*d+u -h (j-1)*d+u .

[0023] Further, the fault diagnosis and analysis module includes a deviation capability index calculation unit and a fault status analysis unit;

[0024] The deviation capability index calculation unit receives the energy consumption data transmitted by the energy consumption data acquisition unit and the data processing energy consumption model transmitted by the data processing energy consumption model construction unit, and calculates the real-time deviation capability index of the smart meter based on the received information;

[0025] The fault state analysis unit compares the deviation capability index of the smart meter at the time (j-1)*d+u with the error value M. If the deviation capability index of the smart meter at the time (j-1)*d+u>M, it means that the energy consumption of the smart meter at the time (j-1)*d+u is abnormal. At this time, the energy consumption data of the smart meter at the time (j-1)*d+u is not put into the data set. If the deviation capability index of the smart meter at the time (j-1)*d+u≤M, it means that the energy consumption of the smart meter at the time (j-1)*d+u is normal. At this time, the energy consumption data of the smart meter at the time (j-1)*d+u is put into the data set, and the data processing energy consumption model is retrained. By retraining the constructed data processing energy consumption module, it is beneficial to improve the recognition accuracy of abnormal energy consumption data, thereby improving the fault diagnosis accuracy of the smart meter.

[0026] Furthermore, the specific method for the deviation capability index calculation unit to calculate the real-time deviation capability index of the smart meter is:

[0027] According to the constructed data processing energy consumption model, the theoretical data processing capacity coefficient W of the smart meter at time (j-1)*d+u is (j-1)*d+u Calculate, W (j-1)*d+u =1-exp{-[(a 1 *(h (j-1)*d+u -g (j-1)*d+u )+b 1 ) / (h (j-1)*d+u -g (j-1)*d+u )]}, where exp represents an exponential function with a natural number e as the base;

[0028] According to formula F (j-1)*d+u =[W (j-1)*d+u -R (j-1)*d+u ] / W (j-1)*d+u The deviation capability index of the smart meter at the time (j-1)*d+u is calculated. The real-time deviation capability index of the smart meter is analyzed according to the constructed data processing energy consumption model, which simplifies the analysis process and is conducive to improving the search rate of abnormal energy consumption data.

[0029] Further, the diagnostic analysis and determination module includes an edge density calculation unit and a fault type analysis and determination unit;

[0030] The edge density calculation unit calculates the real-time edge density of each module of the smart meter according to the continuity of the abnormal energy consumption data collection time and the discrete distribution of the abnormal energy consumption data; based on the continuity of the abnormal energy consumption data collection time and the discrete distribution of the abnormal energy consumption data, it is judged from multiple dimensions whether each module of the smart meter is in a fault state, so as to avoid judging the temporary reduction of module performance of the smart meter due to reasons such as undervoltage as a performance fault of the smart meter;

[0031] The fault type analysis and determination unit receives the calculation result transmitted by the edge density calculation unit, randomly selects a module, and if the edge density calculated by the selected module at the collection time corresponding to the energy consumption data obtained for the pth time is ≥ the set threshold, it is determined that the selection module of the smart meter has a fault; otherwise, it is determined that the selection module of the smart meter has not a fault, and based on the judgment result, the real-time fault type of the smart meter is determined.

[0032] Furthermore, the specific method for the edge density calculation unit to calculate the real-time edge density of each module of the smart meter is:

[0033] When it is analyzed that the energy consumption of the smart meter at the time (j-1)*d+u is abnormal, the energy consumption data of the smart meter at the time (j-1)*d+u is obtained, and the real-time edge density of each module of the smart meter is calculated according to the obtained data. The specific calculation formula is:

[0034] For display modules:

[0035] ;

[0036] For the data processing module:

[0037] ;

[0038] For communication modules:

[0039] ;

[0040] Where p=1,2,…,q, which means the energy consumption data are numbered in chronological order, q represents the total number, z 1 、z 2 、z 3 Respectively represent the energy consumption error values ​​of the display module, data processing module, and communication module, , g p 、h p , k p They represent the first energy consumption value, the second energy consumption value, and the third energy consumption value obtained according to the energy consumption data obtained for the pth time, respectively. prepresents the collection time corresponding to the energy consumption data obtained for the pth time, and the if function represents the conditional judgment function. , when A>0, , when A≤0, , C 2 =max{k 1 -h 1 ,…,k v -h v}, C 3 =max{k 1 -g 1 ,…,k v -g v}, max represents the maximum value symbol, v=1,2,…,V, represents the numbering of normal energy consumption data in chronological order, V represents the total number of normal energy consumption data collection, g v 、h v , k v They represent the first energy consumption value, the second energy consumption value, and the third energy consumption value obtained according to the normal energy consumption data numbered v, respectively. 1p represents the edge density of the display module of the smart meter at the collection time corresponding to the energy consumption data obtained for the pth time, P 2p represents the edge density of the data processing module of the smart meter at the collection time corresponding to the energy consumption data obtained for the pth time, P 3p Represents the edge density of the communication module of the smart meter at the collection time corresponding to the energy consumption data obtained for the pth time.

[0041] An intelligent diagnosis method for equipment faults based on energy consumption analysis, the method comprising:

[0042] S10: Analyze the real-time data processing capability of the smart meter based on the energy consumption data of each module of the smart meter in the standby state and the running state, and build a data processing energy consumption model of the smart meter;

[0043] S20: Calculate the real-time deviation capability index of the smart meter according to the constructed data processing energy consumption model, diagnose and analyze the real-time fault condition of the smart meter according to the calculation result, and retrain the constructed data processing energy consumption model;

[0044] S30: When the energy consumption of the smart meter is abnormal, the real-time edge density of each module of the smart meter is calculated based on the abnormal energy consumption data, and the real-time fault type of the smart meter is analyzed and determined according to the calculation result;

[0045] S40: The fault type of the smart meter determined by the diagnosis analysis determination module is transmitted to the equipment maintenance terminal, and the equipment maintenance terminal dispatches operation and maintenance personnel to perform maintenance and management on the smart meter according to the received fault type.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. The present invention judges whether each module of the smart meter is in a faulty state from multiple dimensions based on the continuity of the abnormal energy consumption data collection time and the discrete distribution of the abnormal energy consumption data, thereby avoiding judging the temporary reduction of module performance of the smart meter due to undervoltage and other reasons as a performance failure of the smart meter. In addition, by retraining the constructed data processing energy consumption model multiple times, it is beneficial to improve the recognition accuracy of abnormal energy consumption data, thereby improving the accuracy of fault diagnosis results.

[0048] 2. The present invention analyzes the real-time deviation capability index of the smart meter based on the constructed data processing energy consumption model, which simplifies the analysis process and is conducive to improving the search rate of abnormal energy consumption data.

[0049] 3. The present invention predicts the real-time data processing capability of the smart meter based on the real-time energy consumption data of the smart meter. This process does not require the acquisition of the real-time collected data volume and the real-time processed data volume of the smart meter, thereby further improving the system's data processing efficiency and effectively diagnosing the faults of the smart meter quickly. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A schematic diagram of the working principle and structure of an intelligent equipment fault diagnosis system and method based on energy consumption analysis of the present invention;

[0051] Figure 2 The present invention is a schematic diagram of the working process of an intelligent equipment fault diagnosis system and method based on energy consumption analysis. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of 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.

[0053] Example: Figure 1 and Figure 2 As shown, the present invention provides a technical solution of an intelligent diagnosis system and method for equipment faults based on energy consumption analysis, an intelligent diagnosis system for equipment faults based on energy consumption analysis, the system includes an energy consumption analysis module, a fault intelligent diagnosis analysis module, a diagnosis analysis determination module and an equipment operation and maintenance processing module;

[0054] The energy consumption analysis module is used to analyze the real-time data processing capability of the smart meter based on the energy consumption data of each module of the smart meter in the standby state and the running state, and to build a data processing energy consumption model of the smart meter;

[0055] The energy consumption analysis module includes an energy consumption data acquisition unit, a data processing capability analysis unit, and a data processing energy consumption model construction unit;

[0056] The energy consumption data collection unit collects energy consumption data of the display module, communication module, and data processing module of the smart meter in the standby state and the running state respectively. The standby state refers to the state when the smart meter does not perform actual power consumption operation. In this state, the display module, communication module, and data processing module of the smart meter are not in the working state. The running state refers to the display module, communication module, and data processing module of the smart meter. At least one module is in the working state;

[0057] The data processing capability analysis unit receives the energy consumption data transmitted by the energy consumption data collection unit, and based on the received information, analyzes the energy consumed by the smart meter under different modules. Based on the analysis results, the real-time data processing capability of the smart meter is predicted. The specific method is as follows:

[0058] The energy consumption value collected by the smart meter in the standby state is recorded as the basic energy consumption value f, the energy consumption value collected by the smart meter when only the display module is in the working state is recorded as the first energy consumption value g, the energy consumption value collected by the smart meter when the display module and the communication module are both in the working state is recorded as the second energy consumption value h, and the energy consumption value collected by the smart meter when the display module, the communication module and the data processing module are all in the working state is recorded as the third energy consumption value k;

[0059] According to the collected first energy consumption value, second energy consumption value, and third energy consumption value, the real-time data processing capability of the smart meter is predicted. The specific prediction formula is:

[0060] ;

[0061] Wherein, j=1,2,…,n, which means that each collection time point of energy consumption data is numbered in chronological order, n represents the total number of times energy consumption data is collected, d represents the collection interval of energy consumption data, u represents the initial collection time of energy consumption data, e represents the natural constant and e=2.72, k (j-1)*d+u represents the third energy consumption value collected by the smart meter at time (j-1)*d+u, h (j-1)*d+u represents the second energy consumption value collected by the smart meter at time (j-1)*d+u, g (j-1)*d+u represents the first energy consumption value collected by the smart meter at time (j-1)*d+u, R (j-1)*d+uRepresents the data processing capacity coefficient of the smart meter at time (j-1)*d+u;

[0062] When the smart meter processes and analyzes the same data, it is assumed that the communication energy consumption of the smart meter for processing the data is equal. At this time, the greater the processing energy consumption of the smart meter for data, the stronger the data processing capability of the smart meter;

[0063] The data processing energy consumption model building unit receives the energy consumption data transmitted by the energy consumption data acquisition unit, and builds the data processing energy consumption model based on the received information. The specific method is as follows:

[0064] Take the dataset {(h u -g u ,k u -h u ),…,(h (j-1)*d+u -g (j-1)*d+u ,k (j-1)*d+u -h (j-1)*d+u )} as the training data set, train the linear model Y=a*X+b, and obtain the data processing energy consumption model Q (j-1)*d+u =a 1 *(h (j-1)*d+u -g (j-1)*d+u )+b 1 ;

[0065] Among them, a 1 represents the weight of the data processing energy consumption model, b 1 represents the bias of the data processing energy consumption model, Q (j-1)*d+u =k (j-1)*d+u -h (j-1)*d+u ;

[0066] The fault diagnosis and analysis module is used to calculate the real-time deviation capability index of the smart meter according to the constructed data processing energy consumption model, and to diagnose and analyze the real-time fault conditions of the smart meter according to the calculation results, and to retrain the constructed data processing energy consumption model;

[0067] The fault diagnosis and analysis module includes a deviation capability index calculation unit and a fault state analysis unit;

[0068] The deviation capability index calculation unit receives the energy consumption data transmitted by the energy consumption data acquisition unit and the data processing energy consumption model transmitted by the data processing energy consumption model construction unit, and calculates the real-time deviation capability index of the smart meter based on the received information. The specific method is as follows:

[0069] According to the constructed data processing energy consumption model, the theoretical data processing capacity coefficient W of the smart meter at time (j-1)*d+u is (j-1)*d+u Calculate, W (j-1)*d+u=1-exp{-[(a 1 *(h (j-1)*d+u -g (j-1)*d+u )+b 1 ) / (h (j-1)*d+u -g (j-1)*d+u )]}, where exp represents an exponential function with a natural number e as the base, and the theoretical data processing capacity coefficient refers to the data processing capacity coefficient of the smart meter obtained based on the collected energy consumption data when the smart meter does not fail;

[0070] According to formula F (j-1)*d+u =[W (j-1)*d+u -R (j-1)*d+u ] / W (j-1)*d+u The deviation capability index of the smart meter at time (j-1)*d+u is calculated, where F (j-1)*d+u It represents the deviation capability index of the smart meter at the time (j-1)*d+u;

[0071] The fault state analysis unit compares the deviation capability index of the smart meter at the time (j-1)*d+u with the error value M. If the deviation capability index of the smart meter at the time (j-1)*d+u>M, it means that the energy consumption of the smart meter at the time (j-1)*d+u is abnormal. At this time, the energy consumption data of the smart meter at the time (j-1)*d+u is not put into the data set. If the deviation capability index of the smart meter at the time (j-1)*d+u≤M, it means that the energy consumption of the smart meter at the time (j-1)*d+u is normal. At this time, the energy consumption data of the smart meter at the time (j-1)*d+u is put into the data set, and the data processing energy consumption model is retrained. The retraining process refers to retraining the linear model Y=a*X+b according to the latest data set;

[0072] When the energy consumption of the smart meter is abnormal, the diagnosis analysis and determination module calculates the real-time edge density of each module of the smart meter based on the abnormal energy consumption data, and analyzes and determines the real-time fault type of the smart meter according to the calculation results;

[0073] The diagnostic analysis and determination module includes an edge density calculation unit and a fault type analysis and determination unit;

[0074] The edge density calculation unit calculates the real-time edge density of each module of the smart meter according to the continuity of the abnormal energy consumption data collection time and the discrete distribution of the abnormal energy consumption data. The specific method is as follows:

[0075] When it is analyzed that the energy consumption of the smart meter at the time (j-1)*d+u is abnormal, the energy consumption data of the smart meter at the time (j-1)*d+u is obtained, and the real-time edge density of each module of the smart meter is calculated according to the obtained data. The specific calculation formula is:

[0076] For display modules:

[0077] ;

[0078] For the data processing module:

[0079] ;

[0080] For communication modules:

[0081] ;

[0082] Where p=1,2,…,q, which means the energy consumption data are numbered in chronological order, q represents the total number, z 1 、z 2 、z 3 Respectively represent the energy consumption error values ​​of the display module, data processing module, and communication module, , g p 、h p , k p They represent the first energy consumption value, the second energy consumption value, and the third energy consumption value obtained according to the energy consumption data obtained for the pth time, respectively. p represents the collection time corresponding to the energy consumption data obtained for the pth time, the if function represents the conditional judgment function, and the if function model is: if (judgment condition, return value when the condition is true, return value when the condition is false), let , when A>0, , when A≤0, , C 2 =max{k 1 -h 1 ,…,k v -h v}, C 3 =max{k 1 -g 1 ,…,k v -g v}, max represents the maximum value symbol, v=1,2,…,V, represents the numbering of normal energy consumption data in chronological order, V represents the total number of normal energy consumption data collection, g v 、h v , k v They represent the first energy consumption value, the second energy consumption value, and the third energy consumption value obtained according to the normal energy consumption data numbered v, respectively. 1p represents the edge density of the display module of the smart meter at the collection time corresponding to the energy consumption data obtained for the pth time, P 2prepresents the edge density of the data processing module of the smart meter at the collection time corresponding to the energy consumption data obtained for the pth time, P 3p represents the edge density of the communication module of the smart meter at the collection time corresponding to the energy consumption data obtained for the pth time;

[0083] The fault type analysis and determination unit receives the calculation result transmitted by the edge density calculation unit, randomly selects a module, and if the edge density calculated by the selected module at the collection time corresponding to the energy consumption data obtained for the pth time is ≥ the set threshold, it is determined that the selected module of the smart meter has a fault, otherwise, it is determined that the selected module of the smart meter has not a fault, and based on the judgment result, the real-time fault type of the smart meter is determined;

[0084] The equipment operation and maintenance processing module is used to transmit the fault type of the smart meter determined by the diagnosis analysis determination module to the equipment maintenance terminal, and the equipment maintenance terminal dispatches operation and maintenance personnel to perform maintenance and management on the smart meter according to the received fault type.

[0085] An intelligent diagnosis method for equipment faults based on energy consumption analysis, the method comprising:

[0086] S10: Analyze the real-time data processing capability of the smart meter based on the energy consumption data of each module of the smart meter in the standby state and the running state, and build a data processing energy consumption model of the smart meter;

[0087] S20: Calculate the real-time deviation capability index of the smart meter according to the constructed data processing energy consumption model, diagnose and analyze the real-time fault condition of the smart meter according to the calculation result, and retrain the constructed data processing energy consumption model;

[0088] S30: When the energy consumption of the smart meter is abnormal, the real-time edge density of each module of the smart meter is calculated based on the abnormal energy consumption data, and the real-time fault type of the smart meter is analyzed and determined according to the calculation result;

[0089] S40: The fault type of the smart meter determined by the diagnosis analysis determination module is transmitted to the equipment maintenance terminal, and the equipment maintenance terminal dispatches operation and maintenance personnel to perform maintenance and management on the smart meter according to the received fault type.

[0090] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An intelligent equipment fault diagnosis system based on energy consumption analysis, characterized in that: The system includes an energy consumption analysis module, a fault intelligent diagnosis and analysis module, a diagnosis and analysis determination module, and an equipment operation and maintenance processing module; The energy consumption analysis module is used to analyze the real-time data processing capability of the smart meter according to the energy consumption data of each module of the smart meter in the standby state and the running state, and to build a data processing energy consumption model of the smart meter; The fault intelligent diagnosis and analysis module is used to calculate the real-time deviation capability index of the smart meter according to the constructed data processing energy consumption model, and to diagnose and analyze the real-time fault conditions of the smart meter according to the calculation results, and to retrain the constructed data processing energy consumption model; When the energy consumption of the smart meter is abnormal, the diagnostic analysis and determination module calculates the real-time edge density of each module of the smart meter based on the abnormal energy consumption data. The specific method is as follows: When it is analyzed that the energy consumption of the smart meter at the time (j-1)*d+u is abnormal, the energy consumption data of the smart meter at the time (j-1)*d+u is obtained, and the real-time edge density of each module of the smart meter is calculated according to the obtained data. The specific calculation formula is: For display modules: ； For the data processing module: ； For communication modules: ； Among them, p=1,2,…,q, which means that the acquired energy consumption data are numbered in chronological order, q represents the total number, z1, z2, z3 represent the energy consumption error values ​​of the display module, data processing module, and communication module respectively, , g p 、h p , k p They represent the first energy consumption value, the second energy consumption value, and the third energy consumption value obtained according to the energy consumption data obtained for the pth time, respectively. p represents the collection time corresponding to the energy consumption data obtained for the pth time, and the if function represents the conditional judgment function. , when A>0, , when A≤0, ,C2=max{k1-h1,…,k v -h v }, C3=max{k1-g1,…,k v -g v }, max represents the maximum value symbol, v=1,2,…,V, represents the numbering of normal energy consumption data in chronological order, V represents the total number of normal energy consumption data collection, g v 、h v , k v They represent the first energy consumption value, the second energy consumption value, and the third energy consumption value obtained according to the normal energy consumption data numbered v, respectively. 1p represents the edge density of the display module of the smart meter at the collection time corresponding to the energy consumption data obtained for the pth time, P 2p represents the edge density of the data processing module of the smart meter at the collection time corresponding to the energy consumption data obtained for the pth time, P 3p represents the edge density of the communication module of the smart meter at the collection time corresponding to the energy consumption data obtained for the pth time, j=1,2,…,n, represents the numbering of each collection time point of the energy consumption data in chronological order, n represents the total number of collection times of the energy consumption data, d represents the collection interval time of the energy consumption data, and u represents the initial collection time of the energy consumption data; And according to the calculation results, the real-time fault type of the smart meter is analyzed and determined; The equipment operation and maintenance processing module is used to transmit the fault type of the smart meter determined by the diagnosis analysis determination module to the equipment maintenance terminal, and the equipment maintenance terminal dispatches operation and maintenance personnel to perform maintenance and management on the smart meter according to the received fault type.

2. According to claim 1, the intelligent equipment fault diagnosis system based on energy consumption analysis is characterized in that: The energy consumption analysis module includes an energy consumption data acquisition unit, a data processing capability analysis unit and a data processing energy consumption model construction unit; The energy consumption data collection unit collects energy consumption data of the display module, communication module, and data processing module of the smart meter in a standby state and an operating state, respectively. The standby state refers to a state when the smart meter does not perform actual power consumption operations. In this state, the display module, communication module, and data processing module of the smart meter are not in a working state. The operating state refers to a state in which at least one of the display module, communication module, and data processing module of the smart meter is in a working state. The data processing capability analysis unit receives the energy consumption data transmitted by the energy consumption data collection unit, analyzes the energy consumed by the smart meter under different modules based on the received information, and predicts the real-time data processing capability of the smart meter based on the analysis result; The data processing energy consumption model construction unit receives the energy consumption data transmitted by the energy consumption data acquisition unit, and constructs the data processing energy consumption model based on the received information.

3. According to claim 2, the intelligent equipment fault diagnosis system based on energy consumption analysis is characterized in that: The specific method for the data processing capability analysis unit to predict the real-time data processing capability of the smart meter is: The energy consumption value collected by the smart meter in the standby state is recorded as the basic energy consumption value f, the energy consumption value collected by the smart meter when only the display module is in the working state is recorded as the first energy consumption value g, the energy consumption value collected by the smart meter when the display module and the communication module are both in the working state is recorded as the second energy consumption value h, and the energy consumption value collected by the smart meter when the display module, the communication module and the data processing module are all in the working state is recorded as the third energy consumption value k; According to the collected first energy consumption value, second energy consumption value, and third energy consumption value, the real-time data processing capability of the smart meter is predicted. The specific prediction formula is: ； Wherein, j=1,2,…,n, which means that each collection time point of energy consumption data is numbered in chronological order, n represents the total number of times energy consumption data is collected, d represents the collection interval of energy consumption data, u represents the initial collection time of energy consumption data, e represents the natural constant and e=2.72, k (j-1)*d+u represents the third energy consumption value collected by the smart meter at time (j-1)*d+u, h (j-1)*d+u represents the second energy consumption value collected by the smart meter at time (j-1)*d+u, g (j-1)*d+u represents the first energy consumption value collected by the smart meter at time (j-1)*d+u, R (j-1)*d+u Represents the data processing capability coefficient of the smart meter at time (j-1)*d+u.

4. The intelligent equipment fault diagnosis system based on energy consumption analysis according to claim 3 is characterized by: The specific method for the data processing energy consumption model construction unit to construct the data processing energy consumption model is: Take the dataset {(h u -g u ,k u -h u ),…,(h (j-1)*d+u -g (j-1)*d+u ,k (j-1)*d+u -h (j-1)*d+u )} as the training data set, train the linear model Y=a*X+b, and obtain the data processing energy consumption model Q (j-1)*d+u =a1*(h (j-1)*d+u -g (j-1)*d+u )+b1; Among them, a1 represents the weight of the data processing energy consumption model, b1 represents the bias of the data processing energy consumption model, Q (j-1)*d+u =k (j-1)*d+u -h (j-1)*d+u .

5. The intelligent equipment fault diagnosis system based on energy consumption analysis according to claim 4 is characterized in that: The fault intelligent diagnosis and analysis module includes a deviation capability index calculation unit and a fault state analysis unit; The deviation capability index calculation unit receives the energy consumption data transmitted by the energy consumption data acquisition unit and the data processing energy consumption model transmitted by the data processing energy consumption model construction unit, and calculates the real-time deviation capability index of the smart meter based on the received information; The fault state analysis unit compares the deviation capability index of the smart meter at time (j-1)*d+u with the error value M. If the deviation capability index of the smart meter at time (j-1)*d+u is greater than M, it indicates that the energy consumption of the smart meter at time (j-1)*d+u is abnormal. At this time, the energy consumption data of the smart meter at time (j-1)*d+u is not put into the data set. If the deviation capability index of the smart meter at time (j-1)*d+u is less than or equal to M, it indicates that the energy consumption of the smart meter at time (j-1)*d+u is normal. At this time, the energy consumption data of the smart meter at time (j-1)*d+u is put into the data set, and the data processing energy consumption model is retrained.

6. The intelligent equipment fault diagnosis system based on energy consumption analysis according to claim 5 is characterized by: The specific method for the deviation capability index calculation unit to calculate the real-time deviation capability index of the smart meter is: According to the constructed data processing energy consumption model, the theoretical data processing capacity coefficient W of the smart meter at time (j-1)*d+u is (j-1)*d+u Calculate, W (j-1)*d+u =1-exp{-[(a1*(h (j-1)*d+u -g (j-1)*d+u )+b1) / (h (j-1)*d+u -g (j-1)*d+u )]}, where exp represents an exponential function with a natural number e as the base; According to formula F (j-1)*d+u =[W (j-1)*d+u -R (j-1)*d+u ] / W (j-1)*d+u Calculate the deviation capability index of the smart meter at time (j-1)*d+u.

7. The intelligent equipment fault diagnosis system based on energy consumption analysis according to claim 6 is characterized by: The diagnostic analysis and determination module includes an edge density calculation unit and a fault type analysis and determination unit; The edge density calculation unit calculates the real-time edge density of each module of the smart meter according to the continuity of the abnormal energy consumption data collection time and the discrete distribution of the abnormal energy consumption data; The fault type analysis and determination unit receives the calculation result transmitted by the edge density calculation unit, randomly selects a module, and if the edge density calculated by the selected module at the collection time corresponding to the energy consumption data obtained for the pth time is ≥ the set threshold, it is determined that the selection module of the smart meter has a fault; otherwise, it is determined that the selection module of the smart meter has not a fault, and based on the judgment result, the real-time fault type of the smart meter is determined.

8. An intelligent device fault diagnosis method based on energy consumption analysis applied to the intelligent device fault diagnosis system based on energy consumption analysis according to any one of claims 1 to 7, characterized in that: The method comprises: S10: Analyze the real-time data processing capability of the smart meter based on the energy consumption data of each module of the smart meter in the standby state and the running state, and build a data processing energy consumption model of the smart meter; S20: Calculate the real-time deviation capability index of the smart meter according to the constructed data processing energy consumption model, diagnose and analyze the real-time fault condition of the smart meter according to the calculation result, and retrain the constructed data processing energy consumption model; S30: When the energy consumption of the smart meter is abnormal, the real-time edge density of each module of the smart meter is calculated based on the abnormal energy consumption data, and the real-time fault type of the smart meter is analyzed and determined according to the calculation result; S40: The fault type of the smart meter determined by the diagnosis analysis determination module is transmitted to the equipment maintenance terminal, and the equipment maintenance terminal dispatches operation and maintenance personnel to perform maintenance and management on the smart meter according to the received fault type.

Images