CN114638170A - Electric energy metering device fault diagnosis method and system based on data mining technology - Google Patents
Electric energy metering device fault diagnosis method and system based on data mining technology Download PDFInfo
- Publication number
- CN114638170A CN114638170A CN202210328343.5A CN202210328343A CN114638170A CN 114638170 A CN114638170 A CN 114638170A CN 202210328343 A CN202210328343 A CN 202210328343A CN 114638170 A CN114638170 A CN 114638170A
- Authority
- CN
- China
- Prior art keywords
- electric energy
- metering device
- energy metering
- gull
- fault diagnosis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 162
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000007418 data mining Methods 0.000 title claims abstract description 20
- 238000005516 engineering process Methods 0.000 title claims abstract description 20
- 241000272168 Laridae Species 0.000 claims abstract description 155
- 238000012706 support-vector machine Methods 0.000 claims abstract description 69
- 238000005457 optimization Methods 0.000 claims abstract description 41
- 238000012549 training Methods 0.000 claims abstract description 34
- 238000012545 processing Methods 0.000 claims abstract description 18
- 230000000739 chaotic effect Effects 0.000 claims description 25
- 238000013507 mapping Methods 0.000 claims description 19
- 230000006399 behavior Effects 0.000 claims description 14
- 230000002159 abnormal effect Effects 0.000 claims description 11
- 241000272161 Charadriiformes Species 0.000 claims description 10
- 238000013508 migration Methods 0.000 claims description 10
- 230000005012 migration Effects 0.000 claims description 10
- 230000005611 electricity Effects 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 9
- 230000003044 adaptive effect Effects 0.000 claims description 7
- 238000011161 development Methods 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 6
- 230000016507 interphase Effects 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 238000004140 cleaning Methods 0.000 claims description 4
- 230000002349 favourable effect Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000013145 classification model Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000001617 migratory effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to a fault diagnosis method and a fault diagnosis system of an electric energy metering device based on a data mining technology, which comprises the following steps: the method comprises the following steps: acquiring different fault state data of the electric energy metering device; step two: processing fault data of the electric energy metering device, extracting key operation parameters and classifying the metering fault types; step three: constructing an electric energy metering device fault diagnosis model based on an improved gull optimization algorithm optimized support vector machine; step four: training a fault diagnosis model of the electric energy metering device; step five: and carrying out fault diagnosis by using the fault diagnosis model of the electric energy metering device. The fault diagnosis model constructed by the invention has higher identification precision and identification efficiency on the running state of the electric energy metering device, can improve the fault diagnosis accuracy and stability of the electric energy metering device, and is favorable for maintaining the fairness and justice of electric energy metering and the safety and stability of power grid running.
Description
Technical Field
The invention belongs to the technical field of electric energy metering device fault diagnosis, and particularly relates to a method and a system for electric energy metering device fault diagnosis based on a data mining technology.
Background
With the development of social economy, domestic electricity consumption is increased, power grid technology is continuously developed, and the construction of an intelligent power grid is a necessary condition for social production and life. The electric energy metering device is an important component of an intelligent power grid as a device for measuring and recording generated energy, power consumption and power consumption. The metering accuracy and the operation reliability of the electric energy metering device play an important role in ensuring the fairness and justice of electric energy metering and the safety and stability of the operation of a power grid. Therefore, the real-time operation data of the electric energy metering device needs to be analyzed, and the fault state of the electric energy metering device needs to be accurately diagnosed.
At present, the fault diagnosis of the electric energy metering device mainly comprises the steps of processing power consumption data of a user, analyzing potential rules between fault characteristics and the power consumption data, establishing a fault diagnosis model, and realizing accurate identification of an abnormal state of the metering device. A Support Vector Machine (SVM) model is a data mining classification model with good performance, can classify electric energy metering faults according to known training data, and is suitable for being used as a fault diagnosis model of an electric energy metering device. However, with the increase of the power consumption, the increase of the data samples leads to overhigh calculation complexity, and meanwhile, the electric energy metering device has a complex structure and a plurality of abnormal working states, so that the model training time is overlong, and the real-time fault diagnosis of the electric energy metering device is influenced. The internal parameter penalty factor C and the kernel function parameter g of the support vector machine directly determine the training efficiency and the application effect of the support vector machine, and the support vector machine model randomly generating the internal parameters has poor classification effect and cannot accurately diagnose faults of the electric energy metering device.
In summary, the complicated structure, the large amount of power consumption data and the various abnormal operating conditions of the electric energy metering system bring difficulties to the fault diagnosis of the electric energy metering device, and how to perform accurate and efficient fault diagnosis on the electric energy metering device becomes a problem which needs to be solved urgently.
Disclosure of Invention
In view of this, the invention provides a fault diagnosis method and system for an electric energy metering device based on a data mining technology. On the premise of giving electricity consumption data of a user, the fault operation state of the electric energy metering device is diagnosed through a data mining technology, and compared with the existing method, the fault diagnosis accuracy rate and efficiency are higher.
The first aspect of the embodiments of the present invention provides a fault diagnosis system for an electric energy metering device based on a data mining technology, including:
the method comprises the following steps: acquiring different fault state data of the electric energy metering device; collecting power consumption data of a user within a period of time through a metering automation system; the power consumption data comprises the abnormal working state of the electric energy metering device and the running parameters under the normal working state.
Step two: processing fault data of the electric energy metering device, extracting key operation parameters and classifying the metering fault types;
step three: constructing an electric energy metering device fault diagnosis model based on an improved gull optimization algorithm optimized support vector machine;
step four: training a fault diagnosis model of the electric energy metering device;
step five: and carrying out fault diagnosis by using the fault diagnosis model of the electric energy metering device.
A second aspect of the embodiments of the present invention provides a method for diagnosing a fault of an electric energy metering device based on a data mining technology, including:
the data acquisition module is used for acquiring different fault data of the electric energy metering device, wherein the different fault data of the electric energy metering device are power consumption data of a user acquired by the metering automation system, part of the data are used as an electric energy metering device training sample set and are used for training an electric energy metering device fault diagnosis model, and the other part of the data are used as an electric energy metering device test sample set and are used for fault diagnosis;
the data processing module is used for processing different fault data of the electric energy metering device, extracting key operation parameters and classifying the metering fault types;
the fault diagnosis model parameter optimization module is used for optimizing a kernel function parameter g and a penalty factor C of the support vector machine by improving the gull algorithm to obtain an electric energy metering device fault diagnosis model based on the improved gull optimization algorithm optimized support vector machine; the improved gull algorithm takes a kernel function parameter g and a penalty factor C of a support vector machine as solutions to be optimized, continuously iterates and optimizes after gull population initialization based on chaotic sine mapping, outputs an optimal gull individual position (C, g) and substitutes the optimal gull individual position into the support vector machine model, and realizes parameter optimization of a fault diagnosis model;
and the fault diagnosis model training module is used for inputting the electric energy metering device training sample set data into the electric energy metering device fault diagnosis model and training the electric energy metering device fault diagnosis model.
And the fault diagnosis module is used for inputting the test sample set data of the electric energy metering device into the fault diagnosis model, outputting the fault type label, obtaining the fault diagnosis result and determining whether the electric energy metering device has a fault and the fault type.
The invention relates to a fault diagnosis method of an electric energy metering device based on a data mining technology, which comprises the following specific steps:
(1) acquiring different fault state data of electric energy metering device
Collecting power consumption data of a user within a period of time through a metering automation system; the power consumption data comprise running parameters of the electric energy metering device in an abnormal working state and a normal working state.
(2) Processing fault data of the electric energy metering device, extracting key operation parameters and classifying the operation state of the electric energy metering device
1) Carrying out data cleaning on the collected power utilization data to remove incomplete data with incomplete collected operation parameters;
2) extracting key operation parameters closely related to the fault of the electric energy metering device, and using the key operation parameters as input data of a support vector machine model; wherein the key operating parameters include: the phase voltage between two phases of the electric energy metering device A, B, the phase current of the a phase, the current and voltage of the metering unit, and the detection signal of the secondary side of the current transformer in the metering device.
3) Classifying the running states of the electric energy metering device according to the collected electricity utilization data, setting corresponding classification labels, namely each running state corresponds to one number, and taking the fault type label as output data of a support vector machine model; wherein the fault operation state of the electric energy metering device comprises the following steps: the short circuit of the primary side of the current transformer, the short circuit of the front end of the secondary side of the current transformer, the short circuit of the rear end of the secondary side of the current transformer, the short circuit of the secondary interphase of the current transformer and the open circuit of the voltage transformer.
(3) Electric energy metering device fault diagnosis model based on improved gull optimization algorithm optimized support vector machine
Optimizing an internal parameter penalty factor C and a kernel function parameter g of a Support Vector Machine (SVM) by adopting an improved gull optimization (ISOA) algorithm, wherein when the internal parameters (C, g) of a fault diagnosis model of the electric energy metering device are optimized, gull individuals represent internal parameter candidate solutions of the SVM model, gull groups represent solution sets generated by the internal parameters of the SVM model, and the individual position with the optimal fitness in the gull groups represents the optimal solution of the internal parameters (C, g) of the support vector machine model;
optimizing the SVM model by using an ISOA algorithm, and setting an objective function Obj of the algorithm as the Accuracy (Accuracy) of fault diagnosis of the electric energy metering device:
in the formula, num.right is the number of data with correct model fault diagnosis; total is the total number of data participating in fault diagnosis.
The method for optimizing the internal parameters (C, g) of the fault diagnosis model of the electric energy metering device by the improved gull optimization algorithm comprises the following specific steps:
1) initializing a gull population based on chaotic sine mapping;
determining the population scale and the maximum iteration times, and initializing the gull population by using chaotic sine mapping to obtain the position of the initialized gull.
The expression of chaotic sine mapping is:
in the formula, gn+1For SVM kernel function parameter to be optimized, delta is control factor with value range of [0, 4 ]]，xnIs a chaotic variable, n is 1, 2.
2) Calculating and evaluating individual fitness values to obtain globally optimal individuals;
and calculating and sequencing the fitness of each gull position in the initial gull group, and finding out the gull position with the optimal fitness. The fitness function Fit of the invention is an objective function Obj as follows:
3) updating the migration position of the individual gull;
the migration behavior of the individual gull represents the global search process in the algorithm, the internal parameters of the fault diagnosis model of the electric energy metering device are different, and the gull individual corresponding to each internal parameter candidate solution updates the current individual position according to the optimal individual position.
The migration position updating formula of the seagull individual is as follows:
Anew＝A·Xs(iter)
A＝fc-(fc·(iter/T))
Dbest＝B·(Xbest(iter)-Xp(iter))
B＝2×A2×rand
Rdis＝|Anew+Dbest|
in the formula, AnewRepresenting updated positions of individual gulls corresponding to internal parameters (C, g) of fault diagnosis model of electric energy metering device, Xp(iter) represents the current position of individual gull, A represents the motion behavior of individual gull in search space, and is used to calculate the update of variables, fcIndicating frequency of use for controlling variable ACoefficient, iter represents the current iteration number, T represents the maximum iteration number, B represents a random number for global search and local development of the balance algorithm, rand is a random number in the range of (0, 1), DbestDirection, X, representing the location of the optimal gull individualbest(iter) represents the current optimal gull individual position, RdisRepresents the distance between the individual gull and the optimal individual gull.
4) Updating the attack position of the individual gull;
the attack behavior of the individual gull represents a local development process in the algorithm, and the attack position of the individual gull in the space is updated according to the formula:
Xp(iter)＝(Rdisu·v·w)+Xbest(iter)
in the formula, u, v and w represent the motion behaviors of the individual gull in a three-dimensional space, r represents the radius of each spiral line when the attack position of the individual gull is updated,
is [0, 2 π ]]The random numbers in the range, k and j, represent constants that define the shape of the helix, and e is the base of the natural logarithm.
5) Introducing wave state adaptive weight to update the individual positions of the gulls;
and introducing wave state self-adaptive weight S into a position updating formula of the individual gull, and updating the position of the individual gull according to the optimal individual position.
The expression for updating the individual positions of the gulls is as follows:
Xp(iter)＝(Rdis×u×v×w)+S×Xbest(iter)
wherein S is the wave state adaptive weight.
6) Updating the global optimal position of the gull group;
calculating and sequencing the fitness of each gull position after position updating, finding out the gull position with the optimal current fitness, and taking the gull position as the global optimal position of the gull group.
7) Judging whether an end condition of the algorithm is met;
and checking whether the algorithm reaches the maximum iteration times or not, and judging whether the end condition of the algorithm is met or not. If the maximum iteration times are reached, the algorithm is ended, and step 8) is executed, if the maximum iteration times are not reached, the next iteration is started, and the steps 3) to 6) are repeatedly executed to continuously search the parameters.
8) Outputting optimal seagull individual positions (C, g) and optimizing support vector machine
And outputting the optimal individual positions (C, g) of the gull after the optimized improved gull algorithm is optimized, and obtaining the optimal kernel function parameter g and the penalty factor C, namely the optimal solution of the internal parameters of the fault diagnosis model of the electric energy metering device.
And substituting the C, g obtained through optimization into the SVM model to obtain an electric energy metering device fault diagnosis model based on an improved gull optimization algorithm optimization support vector machine (ISOA-SVM).
(4) Training fault diagnosis model of electric energy metering device
And taking the different fault state data of the electric energy metering device subjected to data processing as a training sample set, inputting the training sample set data into an ISOA-SVM electric energy metering device fault diagnosis model, and training the electric energy metering device fault diagnosis model.
(5) Fault diagnosis by using fault diagnosis model of electric energy metering device
And performing fault diagnosis by using the ISOA-SVM fault diagnosis model of the electric energy metering device, inputting data of the electric energy metering device needing fault diagnosis, outputting a fault type label, obtaining a fault diagnosis result, and determining whether a fault occurs and the fault type.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
(1) the embodiment of the invention provides an electric energy metering device fault diagnosis method and system based on a data mining technology, wherein a kernel function parameter g and a penalty factor C of a support vector machine are optimized through an improved gull algorithm, an electric energy metering device fault diagnosis model based on an improved gull optimization algorithm optimized support vector machine (ISOA-SVM) is constructed, collected power utilization data are mined and classified through the ISOA-SVM fault diagnosis model, and fault diagnosis of an electric energy metering device is realized. Compared with the existing model, the ISOA-SVM fault diagnosis model has higher identification precision and identification efficiency on the running state of the electric energy metering device, is beneficial to improving the fault diagnosis accuracy and stability of the electric energy metering device, ensures the accurate, reliable and normal running of the metering device, and provides convenience for the maintenance of the electric energy metering device.
(2) The sine chaotic mapping initialization strategy and the wave state adaptive weight are introduced into the traditional gull algorithm, so that the method is more suitable for the optimization problem of the fault diagnosis model of the electric energy metering device. Sine chaotic mapping is introduced to initialize a gull population, and randomness and non-repeated ergodicity of chaotic sequences enable the gull population to be better distributed in the initialization process, so that the optimizing characteristic of a gull optimization algorithm is improved; wave state self-adaptive weight is introduced to update the individual positions of the seagulls, so that local rigidification in the individual updating process is avoided, the algorithm has higher convergence precision, the optimization of a fault diagnosis model of the electric energy metering device is facilitated, and the fault diagnosis accuracy is improved.
(3) The electric energy metering device fault diagnosis method and system based on the data mining technology are reasonable in design, and the constructed ISOA-SVM electric energy metering device fault diagnosis model is not limited to fault diagnosis of a specific type of simple electric energy metering device, but is suitable for fault diagnosis of an electric energy metering device with complex abnormal working conditions. The invention provides a new thought for realizing accurate and efficient fault diagnosis of the electric energy metering device, and is beneficial to maintaining fairness and justice of electric energy metering and safe and stable operation of a power grid.
Drawings
Fig. 1 is a schematic flow chart of a fault diagnosis method for an electric energy metering device based on a data mining technology according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of an electric energy metering device provided in an embodiment of the present invention.
Fig. 3 is a schematic flow chart of an optimized support vector machine model of the improved gull optimization algorithm according to the embodiment of the present invention.
Fig. 4 is a diagram of a result of diagnosing a fault of an ISOA-SVM model according to an embodiment of the present invention.
Fig. 5 is a schematic structural diagram of a fault diagnosis system of an electric energy metering device based on a data mining technology according to an embodiment of the present invention.
Detailed Description
In order to explain the technical solution of the present invention, the following description is made by using specific embodiments with reference to the accompanying drawings.
The embodiment of the invention provides a fault diagnosis method for an electric energy metering device based on a data mining technology, which comprises the following steps of:
(1) acquiring different fault state data of the electric energy metering device;
collecting power consumption data of a user within a period of time through a metering automation system; the power consumption data comprise running parameters of the electric energy metering device in an abnormal working state and a normal working state.
(2) Processing fault data of the electric energy metering device, extracting key operation parameters and classifying the operation state of the electric energy metering device;
1) carrying out data cleaning on the collected power utilization data to remove incomplete data with incomplete collected operation parameters;
2) extracting key operation parameters closely related to the fault of the electric energy metering device, and using the key operation parameters as input data of a support vector machine model; wherein the key operating parameters include: the phase voltage between two phases of the electric energy metering device A, B, the phase current of the a phase, the current and voltage of the metering unit, and the detection signal of the secondary side of the current transformer in the metering device.
In the embodiment of the invention, 8 key operation parameters closely related to the fault of the electric energy metering device are extracted, and are respectively UAB、IA、Ia、Ib、Ua、Ub、ua’、ub', the basic structure of the electric energy metering device is shown in fig. 2; wherein, UABRepresenting the phase voltage between A, B phases, IAPhase current of phase A, Ia、IbRepresents the current, U, of the metering units 1 and 2 in FIG. 2a、UbRepresenting the voltages, u, of the metering units 1 and 2a’、ub' separately represent the detection signals, CT, of the secondary side of the current transformer in the metering system1Representing a current transformer 1, CT2Indicating current transformers 2, PT1Indicating potential transformer 1, PT2 A voltage transformer 2 is shown.
3) Classifying the running states of the electric energy metering device according to the collected electricity utilization data, setting corresponding classification labels, namely each running state corresponds to a number, and taking the fault type label as the output of a support vector machine model; wherein the fault operation state of the electric energy metering device comprises the following steps: the short circuit of the primary side of the current transformer, the short circuit of the front end of the secondary side of the current transformer, the short circuit of the rear end of the secondary side of the current transformer, the short circuit of the secondary interphase of the current transformer and the open circuit of the voltage transformer.
In the embodiment of the invention, the running states of the electric energy metering device are divided into 10 types, including a normal working state and nine abnormal working states; wherein the abnormal operating state comprises: CT1Primary side short circuit, CT2Primary side short circuit, CT1Secondary side rear end short circuit, CT2Secondary side back end short circuit, CT1Secondary side front end short circuit, CT2Secondary side front end short circuit, CT secondary interphase short circuit, PT1Open circuit and PT2And (4) opening a circuit, taking the 10 operation states as classification labels, enabling each operation state to correspond to one digit, and enabling the support vector machine fault diagnosis model to output the digits, so that the corresponding operation state can be determined, and fault diagnosis of the electric energy metering device is realized.
The names of the running states of the electric energy metering device and the corresponding classification labels thereof in the embodiment of the invention are shown in table 1:
meter 1 electric energy metering device operation state classification result
(3) Constructing an electric energy metering device fault diagnosis model based on an improved gull optimization algorithm optimized support vector machine;
in the embodiment of the invention, in order to meet the accuracy requirement of fault diagnosis of the electric energy metering device, an improved gull optimization (ISOA) algorithm is adopted to optimize an internal parameter penalty factor C and a kernel function parameter g of a Support Vector Machine (SVM), the internal parameters (C, g) with the best fault diagnosis effect are output and substituted into an SVM model, and the ISOA-SVM model for fault diagnosis of the electric energy metering device is obtained.
The gull optimization algorithm (SOA) is a novel optimization algorithm, and solves the problem to be optimized by simulating gull migration and attack behaviors. In the embodiment of the invention, the gull individual position represents a candidate solution of the problem of the internal parameters (C, g) of the fault diagnosis model of the preferred electric energy metering device, and the individual position with the optimal fitness in the gull population represents the optimal solution of the internal parameters (C, g) of the support vector machine model.
Optimizing the SVM model by using an ISOA algorithm, and setting an objective function Obj of the algorithm as the fault diagnosis Accuracy (Accuracy) of the electric energy metering device:
in the formula, num.right is the number of data with correct model fault diagnosis; total is the total number of data participating in fault diagnosis.
Fig. 3 shows that the process of optimizing the support vector machine model by improving the gull optimization algorithm in the embodiment of the present invention is: start → initialization of gull population based on chaotic sinusoidal mapping → calculation and evaluation of individual fitness value to obtain globally optimal individual → start iteration → update of migratory position of gull individual → update of attack position of gull individual → introduction of wave state adaptive weight to update gull individual position → update of globally optimal position of gull population → satisfaction of algorithm end condition? No, add 1 to the number of iterations and start the next iteration; yes, end and output the optimal gull individual position (C, g) → optimal support vector machine model → end.
The specific steps of constructing the electric energy metering device fault diagnosis model based on the improved gull optimization algorithm optimized support vector machine are as follows:
1) initializing a gull population based on chaotic sine mapping;
determining the population scale and the maximum iteration times, and initializing the gull population by using chaotic sine mapping to obtain the position of the initialized gull. In the embodiment of the present invention, the population size is 50, and the maximum number of iterations is 500.
The method introduces a chaotic sine mapping initialization strategy, and the chaotic sequence has the characteristics of randomness and non-repeated ergodicity, so that the population of the gull realizes better spatial distribution in the initialization process, and the global optimization capability and the convergence characteristic of the gull optimization algorithm are improved.
The expression of the chaotic sine mapping is as follows:
in the formula, gn+1For SVM kernel function parameter to be optimized, delta is control factor with value range of [0, 4 ]]，xnIs a chaotic variable, n is 1, 2.
2) Calculating and evaluating individual fitness values to obtain globally optimal individuals;
and calculating the fitness of each gull position in the initial gull group according to the fitness function, sequencing the fitness, and finding out the gull individual with the optimal fitness. In the embodiment of the present invention, the fitness function Fit is an objective function Obj, as shown in formula (3):
in the formula, num.right is the number of data with correct model fault diagnosis; total is the total number of data participating in fault diagnosis.
3) Updating the migration position of the individual gull;
the migration behavior of the gull individuals represents the global search process in the algorithm, each internal parameter candidate solution of the fault diagnosis model of the electric energy metering device is different, and the gull individuals corresponding to each internal parameter candidate solution update the current individual positions according to the optimal individual positions.
The migration position updating formula of the seagull individual is as follows:
Anew＝A·Xs(iter) (4)
A＝fc-(fc·(iter/T)) (5)
Dbest＝B·(Xbest(iter)-Xp(iter)) (6)
B＝2×A2×rand (7)
Rdis＝|Anew+Dbest| (8)
in the formula, AnewRepresenting updated positions of individual gulls corresponding to internal parameters (C, g) of fault diagnosis model of electric energy metering device, Xp(iter) represents the current position of the individual gull, represents the current value of the internal parameters of the support vector machine model, A represents the movement behavior of the individual gull in the search space and is used for calculating the update of the variable, fcRepresenting the coefficient for controlling the use frequency of the variable A, iter representing the current iteration number, T representing the maximum iteration number, B representing a random number for global search and local development of the balance algorithm, rand being a random number in the range of (0, 1), DbestDirection, X, representing the position of the optimum gull individualbest(iter) represents the current optimal gull individual position, RdisThe distance between the individual gull and the optimal individual gull is represented.
4) Updating the attack position of the individual gull;
the attack behavior of the individual gull represents a local development process in the algorithm, and the attack position of the individual gull in the space is updated according to the formula:
Xp(iter)＝(Rdisu·v·w)+Xbest(iter) (11)
in the formula, u, v and w represent the motion behaviors of the individual gull in a three-dimensional space, r represents the radius of each spiral line when the attack position of the individual gull is updated, phi is a random number in the range of [0, 2 pi ], k and j represent constants defining the spiral shape, and e is the base number of a natural logarithm.
5) Introducing wave state adaptive weight to update the individual positions of the gulls;
according to the method, the wave state self-adaptive weight S is introduced into the position updating formula of the individual gull, the position of the individual gull is updated according to the optimal individual position, partial rigidity caused by trapping in the individual updating process is avoided, and the searching efficiency of the optimal parameters is improved.
The expression for updating the individual positions of the gulls is as follows:
Xp(iter)＝(Rdis×u×v×w)+S×Xbest(iter) (12)
wherein S is the wave state adaptive weight.
6) Updating the global optimal position of the gull group;
calculating the fitness of each gull position after position updating according to the formula (3), sequencing, finding out the gull position with the optimal current fitness, and taking the gull position as the global optimal position of the gull group.
7) Judging whether an end condition of the algorithm is met or not;
in the embodiment of the invention, whether the iteration times of the algorithm reach 500 times or not is checked, and whether the end condition of the algorithm is met or not is judged. If the maximum iteration number is 500, finishing the algorithm, and executing the step 8), if the maximum iteration number is not reached, starting the next iteration, and repeatedly executing the steps 3) to 6) to continuously search the parameters.
8) Outputting the best seagull individual position (C, g), optimizing SVM
And outputting the optimal individual positions (C, g) of the gull after the optimized improved gull algorithm is optimized, and obtaining the optimal kernel function parameter g and the penalty factor C, namely the optimal solution of the internal parameters of the fault diagnosis model of the electric energy metering device.
Substituting the C, g obtained through optimization into an SVM model to obtain an electric energy metering device fault diagnosis model based on an improved gull optimization algorithm optimized support vector machine (ISOA-SVM), wherein the input of the ISOA-SVM electric energy metering device fault diagnosis model is as follows: key operation parameters which are obtained from the electricity consumption data of users and closely related to the faults of the electric energy metering device; the output of the fault diagnosis model of the ISOA-SVM electric energy metering device is as follows: the fault operation state of the electric energy metering device corresponds to the classification label.
(4) Training fault diagnosis model of electric energy metering device
In the embodiment of the invention, 480 groups of data are randomly selected from 10 working state data of a plurality of groups of electric energy metering devices subjected to data processing to serve as a sample set, the sample set is divided, the first 400 groups of data serve as a training sample set, and the last 80 groups of data serve as a testing sample set.
Inputting 8 key operation parameters closely related to the fault of the electric energy metering device in the training sample set data into an ISOA-SVM fault diagnosis model of the electric energy metering device, and training the fault diagnosis model of the electric energy metering device; the key operating parameters include: u shapeAB、IA、Ia、Ib、Ua、Ub、ua’、ub’。
(5) Fault diagnosis by using fault diagnosis model of electric energy metering device
And performing fault diagnosis by using the ISOA-SVM fault diagnosis model of the electric energy metering device, inputting data of the electric energy metering device needing fault diagnosis, outputting a fault type label, obtaining a fault diagnosis result, and determining whether a fault occurs and the fault type.
In the embodiment of the invention, 8 key operation parameters closely related to the fault of the electric energy metering device in the test sample set data are input into the ISOA-SVM electric energy metering deviceA device fault diagnosis model, the key operating parameters comprising: u shapeAB、IA、Ia、Ib、Ua、Ub、ua’、ub'; and outputting the fault type label to obtain a fault diagnosis result. FIG. 4 is a diagram of a result of a fault diagnosis of an ISOA-SVM model provided by an embodiment of the present invention; the abscissa is an electricity consumption data label corresponding to 80 groups of test sample data, and the ordinate is a fault type label representing corresponding classification labels of 10 operating states of the electric energy metering device.
Fig. 4 shows a comparison result of actual fault types and fault diagnosis types of the electric energy metering device in the embodiment of the present invention, in 80 groups of data, only 5 groups of data have been subjected to misdiagnosis, and the fault diagnosis accuracy of the ISOA-SVM model obtained by calculation is 93.75%, which proves that the ISOA-SVM model has higher precision in fault diagnosis of the electric energy metering device, and further illustrates that the fault diagnosis method of the electric energy metering device based on the data mining technology of the present invention can accurately distinguish complex fault types, and is more advantageous in solving the fault diagnosis problem of the electric energy metering device.
An embodiment of the present invention further provides a fault diagnosis system for an electric energy metering device based on a data mining technology, as shown in fig. 5, the system includes: the system comprises a data acquisition module S100, a data processing module S200, a fault diagnosis model parameter optimization module S300, a fault diagnosis model training module S400 and a fault diagnosis module S500.
The data acquisition module S100 is used for acquiring different fault data of the electric energy metering device, the different fault data of the electric energy metering device are power consumption data of a user acquired by the metering automation system, part of the data are used as an electric energy metering device training sample set for training an electric energy metering device fault diagnosis model, and the other part of the data are used as an electric energy metering device test sample set for fault diagnosis.
The data processing module S200 is used for processing different fault data of the electric energy metering device, extracting key operation parameters and classifying metering fault types; wherein the key operating parameters include: the phase voltage between two phases of the electric energy metering device A, B, the phase current of the A phase, the current and the voltage of the metering unit, and the detection signal of the secondary side of the current transformer in the metering device, and the metering fault types include: the short circuit of the primary side of the current transformer, the short circuit of the front end of the secondary side of the current transformer, the short circuit of the rear end of the secondary side of the current transformer, the short circuit of the secondary interphase of the current transformer and the open circuit of the voltage transformer.
The fault diagnosis model parameter optimization module S300 is used for optimizing a kernel function parameter g and a penalty factor C of the support vector machine by improving a gull algorithm to obtain an electric energy metering device fault diagnosis model based on the improved gull optimization algorithm optimized support vector machine; the improved gull algorithm takes a kernel function parameter g and a penalty factor C of a support vector machine as solutions to be optimized, continuously iterates and optimizes after gull population initialization based on chaotic sine mapping, outputs an optimal gull individual position (C, g) and substitutes the optimal gull individual position into the support vector machine model, and parameter optimization of a fault diagnosis model is realized;
the fault diagnosis model training module S400 is used for inputting the electric energy metering device training sample set data into the electric energy metering device fault diagnosis model and training the electric energy metering device fault diagnosis model.
The fault diagnosis module S500 is used for inputting the data of the test sample set of the electric energy metering device into the fault diagnosis model, outputting the fault type label, obtaining the fault diagnosis result, and determining whether the electric energy metering device has a fault and the fault type.
It is emphasized that the examples described herein are illustrative and not restrictive, and thus the present invention includes, but is not limited to, those examples described herein and that other embodiments derived from the teachings of the present invention by those of ordinary skill in the art are also within the scope of the present invention.
Claims (6)
1. A fault diagnosis method for an electric energy metering device based on a data mining technology is characterized by comprising the following steps:
the method comprises the following steps: acquiring different fault state data of the electric energy metering device; collecting power consumption data of a user within a period of time through a metering automation system; the electricity consumption data comprises an abnormal working state and operating parameters under a normal working state of the electric energy metering device;
step two: processing fault data of the electric energy metering device, extracting key operation parameters and classifying the metering fault types;
step three: constructing an electric energy metering device fault diagnosis model based on an improved gull optimization algorithm optimized support vector machine;
step four: training a fault diagnosis model of the electric energy metering device;
step five: and carrying out fault diagnosis by using the fault diagnosis model of the electric energy metering device.
2. An electric energy metering device fault diagnosis system based on data mining technology is characterized by comprising:
the data acquisition module is used for acquiring different fault data of the electric energy metering device, wherein the different fault data of the electric energy metering device are power consumption data of a user acquired by the metering automation system, part of the data are used as an electric energy metering device training sample set and are used for training an electric energy metering device fault diagnosis model, and the other part of the data are used as an electric energy metering device test sample set and are used for fault diagnosis;
the data processing module is used for processing different fault data of the electric energy metering device, extracting key operation parameters and classifying metering fault types;
the fault diagnosis model parameter optimization module is used for optimizing a kernel function parameter g and a penalty factor C of the support vector machine through an improved gull algorithm to obtain an electric energy metering device fault diagnosis model based on the improved gull optimization algorithm optimized support vector machine; the improved gull algorithm takes a kernel function parameter g and a penalty factor C of a support vector machine as solutions to be optimized, continuously iterates and optimizes after gull population initialization based on chaotic sine mapping, outputs an optimal gull individual position (C, g) and substitutes the optimal gull individual position into the support vector machine model, and realizes parameter optimization of a fault diagnosis model;
the fault diagnosis model training module is used for inputting the electric energy metering device training sample set data into the electric energy metering device fault diagnosis model and training the electric energy metering device fault diagnosis model;
and the fault diagnosis module is used for inputting the test sample set data of the electric energy metering device into the fault diagnosis model, outputting the fault type label, obtaining the fault diagnosis result and determining whether the electric energy metering device has a fault and the fault type.
3. The method of claim 1, wherein the improving gull optimization algorithm comprises:
introducing a chaotic sine mapping initialization strategy, and initializing a gull population by adopting chaotic sine mapping to obtain the position of an initialized gull;
the expression of the chaotic sine mapping is as follows:
in the formula, gn+1For SVM kernel function parameter to be optimized, delta is control factor with value range of [0, 4 ]]，xnIs a chaotic variable, n is 1, 2.
Introducing a wave state self-adaptive weight S, introducing the wave state self-adaptive weight S into a position updating formula of the individual gull, and updating the position of the individual gull according to the optimal individual position;
the expression for updating the individual positions of the gulls is as follows:
Xp(iter)＝(Rdis×u×v×w)+S×Xbest(iter)
in the formula, Xp(iter) represents the current position of the individual gull, RdisRepresenting the distance between the individual gull and the optimal gull, u, v and w representing the motion behavior of the individual gull in three-dimensional space, S representing wave-state adaptive weight, and Xbest(iter) represents the position of the current best gull individual.
4. The method for fault diagnosis of the electric energy metering device according to claim 1, wherein the processing fault data of the electric energy metering device, extracting key operation parameters and classifying the operation state of the electric energy metering device comprises:
carrying out data cleaning on the collected power utilization data to remove incomplete data with incomplete collected operation parameters;
extracting key operation parameters closely related to the fault of the electric energy metering device, and using the key operation parameters as input data of a support vector machine model; wherein the key operating parameters include: phase voltage between two phases of the electric energy metering device A, B, phase current of the A phase, current and voltage of the metering unit, and detection signals of the secondary side of a current transformer in the metering device;
classifying the running states of the electric energy metering device according to the collected electricity utilization data, setting corresponding classification labels, namely each running state corresponds to a number, and taking the fault type label as the output of a support vector machine model; wherein the fault operation state of the electric energy metering device comprises the following steps: the short circuit of the primary side of the current transformer, the short circuit of the front end of the secondary side of the current transformer, the short circuit of the rear end of the secondary side of the current transformer, the short circuit of the secondary interphase of the current transformer and the open circuit of the voltage transformer.
5. The method of claim 1, wherein the internal parameters of a Support Vector Machine (SVM) are optimized using an improved gull optimization (ISOA) algorithm, the internal parameters including a penalty factor C and a kernel parameter g;
when the internal parameters of the fault diagnosis model of the electric energy metering device are optimized, gull individuals represent candidate solutions of the internal parameters of the SVM model, gull groups represent a solution set generated by the internal parameters of the SVM model, and the positions of the individuals with the optimal fitness in the gull groups represent the optimal solution of the internal parameters (C, g) of the support vector machine model.
6. A fault diagnosis method for an electric energy metering device based on a data mining technology comprises the following specific steps:
(1) acquiring different fault state data of electric energy metering device
Collecting power consumption data of a user within a period of time through a metering automation system; the power consumption data comprises running parameters of the electric energy metering device in an abnormal working state and a normal working state;
(2) processing fault data of the electric energy metering device, extracting key operation parameters and classifying the operation state of the electric energy metering device
1) Carrying out data cleaning on the collected power utilization data to remove incomplete data with incomplete collected operation parameters;
2) extracting key operation parameters closely related to the fault of the electric energy metering device, and using the key operation parameters as input data of a support vector machine model; wherein the key operating parameters include: phase voltage between two phases of the electric energy metering device A, B, phase current of the A phase, current and voltage of the metering unit, and detection signals of the secondary side of a current transformer in the metering device;
3) classifying the running states of the electric energy metering device according to the collected electricity utilization data, setting corresponding classification labels, namely each running state corresponds to a number, and taking the fault type label as the output of a support vector machine model; wherein the fault operation state of the electric energy metering device comprises the following steps: a primary side short circuit of the current transformer, a secondary side front end short circuit of the current transformer, a secondary side rear end short circuit of the current transformer, a secondary interphase short circuit of the current transformer and an open circuit of the voltage transformer;
(3) electric energy metering device fault diagnosis model based on improved gull optimization algorithm optimized support vector machine
Optimizing the SVM model by using an ISOA algorithm, and setting an objective function Obj of the algorithm as the fault diagnosis Accuracy (Accuracy) of the electric energy metering device:
in the formula, num.right is the number of data with correct model fault diagnosis; num.total is the number of all data participating in fault diagnosis, and the specific steps of optimizing internal parameters (C, g) of a fault diagnosis model of the electric energy metering device by the improved seagull optimization algorithm are as follows:
1) initializing a gull population based on chaotic sine mapping;
determining the population scale and the maximum iteration times, and initializing the gull population by using chaotic sine mapping to obtain the position of the initialized gull;
the expression of chaotic sine mapping is:
in the formula, gn+1For SVM kernel function parameter to be optimized, delta is control factor with value range of [0, 4 ]]，xnIs a chaotic variable, n is 1, 2.
2) Calculating and evaluating individual fitness values to obtain globally optimal individuals;
and calculating and sequencing the fitness of each gull position in the initial gull group, and finding out the gull position with the optimal fitness. The fitness function Fit of the invention is an objective function Obj as follows:
3) updating the migration position of the individual gull;
the migration behavior of the gull individual represents the global search process in the algorithm, the internal parameters of the fault diagnosis model of the electric energy metering device are different, and the gull individual corresponding to each internal parameter candidate solution updates the current individual position according to the optimal individual position;
the migration position updating formula of the seagull individual is as follows:
Anew＝A·Xs(iter)
A＝fc-(fc·(iter/T))
Dbest＝B·(Xbest(iter)-Xp(iter))
B＝2×A2×rand
Rdis＝|Anew+Dbest|
in the formula, AnewRepresents the updated position of individual gull corresponding to the internal parameters (C, g) of the fault diagnosis model of the electric energy metering device, Xp(iter) represents the current position of individual gull, A represents the motion behavior of individual gull in search space, and is used to calculate the update of variables, fcRepresenting the coefficient for controlling the use frequency of the variable A, iter representing the current iteration number, T representing the maximum iteration number, B representing a random number for global search and local development of the balance algorithm, rand being a random number in the range of (0, 1), DbestDirection, X, representing the location of the optimal gull individualbest(iter) represents the current optimal gull individual position, RdisRepresenting the distance between the individual gull and the optimal individual gull;
4) updating the attack position of the individual gull;
the attack behavior of the individual gull represents a local development process in the algorithm, and the attack position of the individual gull in the space is updated according to the formula:
Xp(iter)＝(Rdisu·v·w)+Xbest(iter)
in the formula, u, v and w represent the motion behaviors of individual gulls in a three-dimensional space, r represents the radius of each spiral line when the attack position of the individual gull is updated,
is [0, 2 π ]]Random numbers in the range, k and j representing constants defining the shape of the helix, e being the base of the natural logarithm;
5) introducing wave state self-adaptive weight to update the individual positions of the gulls;
introducing a wave state self-adaptive weight S into a position updating formula of the individual gull, and updating the position of the individual gull according to the optimal individual position;
the expression for updating the individual positions of the gulls is as follows:
Xp(iter)＝(Rdis×u×v×w)+S×Xbest(iter)
in the formula, S is wave state self-adaptive weight;
6) updating the global optimal position of the gull group;
calculating and sequencing the fitness of each gull position after position updating, finding out the gull position with the optimal current fitness, and taking the gull position as the global optimal position of the gull group;
7) judging whether an end condition of the algorithm is met;
and checking whether the algorithm reaches the maximum iteration times or not, and judging whether the end condition of the algorithm is met or not. If the maximum iteration times are reached, finishing the algorithm, executing the step 8), if the maximum iteration times are not reached, starting the next iteration, and repeatedly executing the steps 3) to 6) to continuously search the parameters;
8) outputting the best seagull individual position (C, g), optimizing the support vector machine
Outputting the optimal individual positions (C, g) of the gull after the optimized improved gull algorithm, and obtaining an optimal kernel function parameter g and a penalty factor C, namely an optimal solution of internal parameters of the fault diagnosis model of the electric energy metering device;
substituting the C, g obtained through optimization into an SVM model to obtain an electric energy metering device fault diagnosis model based on an improved gull optimization algorithm optimization support vector machine (ISOA-SVM);
(4) training fault diagnosis model of electric energy metering device
Taking different fault state data of the electric energy metering device after data processing as a training sample set, inputting the training sample set data into an ISOA-SVM electric energy metering device fault diagnosis model, and training the electric energy metering device fault diagnosis model;
(5) fault diagnosis by using fault diagnosis model of electric energy metering device
And performing fault diagnosis by using the ISOA-SVM fault diagnosis model of the electric energy metering device, inputting data of the electric energy metering device needing fault diagnosis, outputting a fault type label, obtaining a fault diagnosis result, and determining whether a fault occurs and the fault type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210328343.5A CN114638170A (en) | 2022-03-31 | 2022-03-31 | Electric energy metering device fault diagnosis method and system based on data mining technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210328343.5A CN114638170A (en) | 2022-03-31 | 2022-03-31 | Electric energy metering device fault diagnosis method and system based on data mining technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114638170A true CN114638170A (en) | 2022-06-17 |
Family
ID=81951677
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210328343.5A Pending CN114638170A (en) | 2022-03-31 | 2022-03-31 | Electric energy metering device fault diagnosis method and system based on data mining technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114638170A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115660024A (en) * | 2022-09-07 | 2023-01-31 | 中国人民解放军陆军装甲兵学院 | Vehicle-mounted network fault diagnosis method based on intelligent optimization algorithm and machine learning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017128455A1 (en) * | 2016-01-25 | 2017-08-03 | 合肥工业大学 | Analogue circuit fault diagnosis method based on generalized multiple kernel learning-support vector machine |
| CN111734961A (en) * | 2020-06-24 | 2020-10-02 | 东北石油大学 | Natural gas pipeline leakage detection method |
| CN113392513A (en) * | 2021-05-28 | 2021-09-14 | 国网河北省电力有限公司衡水供电分公司 | Multi-objective optimization method, device and terminal for combined cooling, heating and power system |
| CN114091334A (en) * | 2021-11-18 | 2022-02-25 | 国网山东省电力公司电力科学研究院 | Partial discharge fault diagnosis method based on improved bat algorithm and support vector machine |
-
2022
- 2022-03-31 CN CN202210328343.5A patent/CN114638170A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017128455A1 (en) * | 2016-01-25 | 2017-08-03 | 合肥工业大学 | Analogue circuit fault diagnosis method based on generalized multiple kernel learning-support vector machine |
| CN111734961A (en) * | 2020-06-24 | 2020-10-02 | 东北石油大学 | Natural gas pipeline leakage detection method |
| CN113392513A (en) * | 2021-05-28 | 2021-09-14 | 国网河北省电力有限公司衡水供电分公司 | Multi-objective optimization method, device and terminal for combined cooling, heating and power system |
| CN114091334A (en) * | 2021-11-18 | 2022-02-25 | 国网山东省电力公司电力科学研究院 | Partial discharge fault diagnosis method based on improved bat algorithm and support vector machine |
Non-Patent Citations (3)
| Title |
|---|
| 朱少斌等: ""基于BSO-BPNN模型的电能计量装置异常诊断方法研究"", 《中国测试》, 9 July 2021 (2021-07-09), pages 1 - 7 * |
| 毛清华等: ""融合改进Logistics混沌和正弦余弦算子的自适应t分布海鸥算法"", 《小型微型计算机系统》, 22 October 2021 (2021-10-22), pages 1 - 9 * |
| 续婷: ""基于群智能算法与机器学习的预测与分类研究"", 《中国博士学位论文全文数据库基础科学辑》, 15 January 2022 (2022-01-15), pages 15 - 102 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115660024A (en) * | 2022-09-07 | 2023-01-31 | 中国人民解放军陆军装甲兵学院 | Vehicle-mounted network fault diagnosis method based on intelligent optimization algorithm and machine learning |
| CN115660024B (en) * | 2022-09-07 | 2024-01-19 | 中国人民解放军陆军装甲兵学院 | Vehicle-mounted network fault diagnosis method based on intelligent optimization algorithm and machine learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109931678B (en) | Air conditioner fault diagnosis method based on deep learning LSTM | |
| CN109359665B (en) | Household appliance load identification method and device based on support vector machine | |
| CN105279397A (en) | Method for identifying key proteins in protein-protein interaction network | |
| CN110580502A (en) | Factor hidden Markov load decomposition method based on Gaussian mixture | |
| CN111628494B (en) | Low-voltage distribution network topology identification method and system based on logistic regression method | |
| CN113452018B (en) | Method for identifying standby shortage risk scene of power system | |
| CN111639586A (en) | Non-invasive load identification model construction method, load identification method and system | |
| CN113125960A (en) | Vehicle-mounted lithium ion battery charge state prediction method based on random forest model | |
| CN114638170A (en) | Electric energy metering device fault diagnosis method and system based on data mining technology | |
| CN115759393A (en) | Cumulative load baseline prediction method based on ensemble learning | |
| CN117078114B (en) | Water quality evaluation method and system for water-bearing lakes under influence of diversion engineering | |
| CN116842459B (en) | Electric energy metering fault diagnosis method and diagnosis terminal based on small sample learning | |
| CN111539492B (en) | Abnormal electricity utilization judgment system and method based on reinforcement learning | |
| CN116774086B (en) | Lithium battery health state estimation method based on multi-sensor data fusion | |
| CN113358157A (en) | RST-PNN-GA-based power equipment temperature rise detection and early warning method | |
| CN111337833B (en) | Lithium battery capacity integrated prediction method based on dynamic time-varying weight | |
| CN112595918A (en) | Low-voltage meter reading fault detection method and device | |
| CN116662893A (en) | Water quality prediction method for optimizing SVM (support vector machine) based on improved goblet sea squirt algorithm | |
| CN116561569A (en) | Industrial power load identification method based on EO feature selection and AdaBoost algorithm | |
| Ma et al. | State-of-Health assessment for aero-engine based on density-distance clustering and fuzzy Bayesian risk | |
| CN111090679A (en) | Time sequence data representation learning method based on time sequence influence and graph embedding | |
| CN115712871A (en) | Power electronic system fault diagnosis method combining resampling and integrated learning | |
| CN115238951A (en) | Power load prediction method and device | |
| CN113889274B (en) | Method and device for constructing risk prediction model of autism spectrum disorder | |
| CN116127447A (en) | Virtual power plant false data injection attack detection method, device, terminal and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |