CN113269252A - Power electronic circuit fault diagnosis method for optimizing ELM based on improved sparrow search algorithm - Google Patents
Power electronic circuit fault diagnosis method for optimizing ELM based on improved sparrow search algorithm Download PDFInfo
- Publication number
- CN113269252A CN113269252A CN202110579753.2A CN202110579753A CN113269252A CN 113269252 A CN113269252 A CN 113269252A CN 202110579753 A CN202110579753 A CN 202110579753A CN 113269252 A CN113269252 A CN 113269252A
- Authority
- CN
- China
- Prior art keywords
- sparrow
- search algorithm
- electronic circuit
- power electronic
- fitness
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/28—Testing of electronic circuits, e.g. by signal tracer
- G01R31/2832—Specific tests of electronic circuits not provided for elsewhere
- G01R31/2836—Fault-finding or characterising
- G01R31/2846—Fault-finding or characterising using hard- or software simulation or using knowledge-based systems, e.g. expert systems, artificial intelligence or interactive algorithms
- G01R31/2848—Fault-finding or characterising using hard- or software simulation or using knowledge-based systems, e.g. expert systems, artificial intelligence or interactive algorithms using simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a power electronic circuit fault diagnosis method for optimizing ELM based on an improved sparrow search algorithm. Firstly, acquiring a load voltage signal of an actual power electronic circuit to be tested through a sensor, decomposing the voltage signal to obtain a characteristic vector of the signal, and dividing the characteristic vector of the signal into a test sample and a training sample. And building an ELM neural network, and taking the sum of the error rate of the training sample and the error rate of the test sample as a fitness function of the sparrow search algorithm. Aiming at the fact that the sparrow search algorithm is prone to falling into local optimum, the optimization method of the gravity center reverse learning sparrow search algorithm based on wavelet variation is provided, the global search capability is enhanced, the diversity of the population is enriched, the possibility of falling into the local optimum is reduced, and the optimization capability of the sparrow search algorithm is improved. And optimizing the weight and the threshold of the ELM neural network by using a sparrow search algorithm to obtain a power electronic circuit fault diagnosis model for optimizing the ELM neural network by using the sparrow search algorithm, and inputting the test sample into the trained power electronic circuit fault diagnosis model for optimizing the ELM neural network by using the sparrow search algorithm to finish the fault diagnosis of the power electronic circuit.
Description
Technical Field
The invention relates to the technical field of power electronic circuit fault diagnosis, in particular to a power electronic circuit fault diagnosis method for optimizing ELM based on an improved sparrow search algorithm.
Background
Power electronics technology has been developed in the 70 s of the 20 th century, through the high-speed development of fifty years, topological structure of circuit, novel material, continuous perfection of control mode and smart grid, integrated circuit, continuous development of high intelligent industrialization, high performance, power electronics of high complexity constantly emerge. Nowadays, power electronic devices play a very important role in many fields, but the structure, the number and the working environment are increasingly complex, the failure rate of power electronic devices is also increased, and thus, the production economy is greatly and unpredictably lost. Therefore, the fault diagnosis technology of the power electronic circuit is researched and applied to the actual power electronic device, and the fault diagnosis technology has great significance for industrial production, national defense safety and economic growth. With the intensive research of artificial intelligence and deep learning and the continuous application of numerous achievements, our country seizes the first opportunity, continuously improves the competitiveness of the country, continuously increases the research and development application investment, and is in a leading state in the aspect of computers nowadays. Therefore, artificial intelligence and deep learning are rapidly developed today, and a new opportunity is brought to fault diagnosis of the power electronic device.
Therefore, the invention provides a power electronic circuit fault diagnosis method for optimizing ELM based on an improved sparrow search algorithm. The sparrow search algorithm (sparrow search algorithm) is a group intelligence optimization algorithm proposed based on the behavior of sparrows to forage for food and evade predators. The sparrow searching algorithm mainly simulates the sparrow group foraging process and is divided into a finder, a follower and an alarm, and an individual with good food in sparrows is found as the finder and provides a foraging area and direction for the follower. Other individuals are used as followers, meanwhile, a certain proportion of individuals in the population are selected for detection and early warning, if danger is found, food is abandoned, sparrows at the edges of the population can rapidly move to a safe area to obtain a better position, and sparrows in the middle of the population can randomly move to be close to other sparrows. The novel group intelligent optimization algorithm has better global exploration and local development capabilities, and can enable sparrows in a group to move to a global optimal value and quickly converge near the optimal value. The convergence rate is obvious at the beginning, and the algorithm has the characteristics of good stability, strong global search capability and few parameters, so that aiming at the problems of low ELM accuracy and low stability, the initial weight and the threshold of the ELM can be optimized by utilizing the sparrow search algorithm aiming at the practical problems.
The sparrow search algorithm has the characteristics of good stability, strong global search capability and few parameters, but has the common problem with many group intelligent optimization algorithms, namely the problems of reduced population diversity, easy falling into local optimization and the like easily occur when the global optimal position is close to the intelligent optimization algorithm.
Disclosure of Invention
The invention aims to provide a power electronic circuit fault diagnosis method for optimizing ELM based on an improved sparrow search algorithm, and provides a gravity center reverse learning sparrow search algorithm optimization method based on wavelet variation, which aims at the problem that the sparrow search algorithm is easy to fall into local optimum, enhances the global search capability, enriches the diversity of population, reduces the possibility of falling into local optimum, and can further improve the accuracy of equipment fault diagnosis by establishing an SSA-ELM model through the improved sparrow search algorithm optimized ELM.
In order to achieve the above purpose, the solution of the invention is:
a power electronic circuit fault diagnosis method for optimizing ELM based on an improved sparrow search algorithm comprises the following steps:
step (1): the method comprises the steps that collected voltage signals of the power electronic circuit are decomposed through a variational mode, and the characteristic vectors of the signals are obtained;
step (2): building an ELM neural network, training the ELM neural network, and calculating the fitness value of each sparrow individual by selecting the sum of the error rate of a training sample and the error rate of a test sample as a fitness function of a sparrow search algorithm;
and (3): according to the problems that the sparrow search algorithm is easy to fall into local optimum and insufficient in population diversity in actual operation, the sparrow search algorithm is improved;
and (4): optimizing the weight and the threshold of the ELM neural network by using an improved sparrow search algorithm to obtain a power electronic circuit fault diagnosis model for optimizing the ELM neural network by using the sparrow search algorithm, inputting the feature vector in the step (1) into the power electronic circuit fault diagnosis model for optimizing the ELM neural network by using the sparrow search algorithm, and completing training;
and (5): and inputting the test sample into a trained power electronic circuit fault diagnosis model of which the ELM neural network is optimized by a sparrow search algorithm to finish the fault diagnosis of the power electronic circuit.
Specifically, the fitness function value of each sparrow in step (2) is calculated by using the following function:
fitness＝argmin(Tr_ErrorR-Te_ErrorR)
wherein, Tr _ ErrorR is the error rate of the training sample, and Te _ ErrorR is the error rate of the testing sample.
Specifically, the modified sparrow algorithm operation in step (3) includes the following steps:
step 3.1: setting the sparrow population scale as N and the number of discoverers as PNNumber of early-warners SNThe dimension D of the objective function, the upper bound Ub and the lower bound Lb of the initial value, and the maximum iteration number Tmax；
Step 3.2: evaluating the current individuals and the reverse individuals by using a gravity center reverse learning strategy, and selecting N individuals with the best fitness as an initialization population;
step 3.3: calculating the position x of each sparrow in D-dimensional spaceiAnd fitness F of each sparrowiSelecting the sparrow F with the best initial fitness in the populationbAnd position x of sparrowbAnd sparrow F with the worst initial fitnessdAnd sparrow position xd；
Step 3.4: calculating and sorting the fitness of sparrow populations, and selecting the P with the best fitness in each generation of populationsNSparrows as discoverers, the rest N-PNSparrow as follower, SNThe number of individuals are used for reconnaissance and early warning as alarm persons, sparrows at the edges of the population can rapidly move to a safe area during early warning so as to obtain a better position, and the positions of a finder, a follower and the alarm persons are updated according to a formula;
step 3.5: after iteration is finished, calculating the fitness value of each sparrow in the population and the average fitness value of the population, and performing wavelet variation operation on the individual positions with the fitness smaller than the average fitness of the population;
step 3.6: and judging whether the requirement of the maximum iteration number is met, if so, outputting the result as an initial weight and a threshold of the ELM neural network, and otherwise, returning to the step 3.3.
Specifically, the step 3.2 includes:
assuming that the size of a sparrow is N, the dimension of the search space is D, and the position of the ith sparrow in the search space is:
wherein i 1,2, N, j 1,2ijIndicating the position of the ith sparrow in the jth dimension;
the fitness value of the ith sparrow is expressed as:
where f represents the fitness value.
5. Specifically, in step 3.2, the gravity center reverse learning strategy is applied as follows:
set in a D-dimensional search space, [ X ]1,X2,...,Xn]Are n points with mass, since the center of gravity of the whole is defined as:
if the center of gravity of a discrete uniform whole is M, a certain point X in MiThe reverse point formula of (a) can be updated as:
the search space in which the reverse point is located has dynamic boundaries in which
Lbj＝min(Xij),Ubj＝max(Xij) K is [01 ]]Random numbers uniformly distributed in intervals; if the inverse solution becomes non-solvable beyond the boundary, the formula is as follows:
where rand represents a random number on (0, 1).
The population selection method by using the gravity center reverse learning strategy is as follows:
randomly generating D-dimensional vectors with values between (0,1)
Generating 2N vectors according to formulas (3) - (6); wherein i 1,2., N, j 1,2.. D; and then calculating the fitness value of each individual in the population, and selecting N individuals with the best fitness values as the positions for initializing the population.
Specifically, in step 3.4, the follower location update formula is:
wherein rand { -11} represents a randomly selected value of-1 or 1,
is the current best position of the sparrow.
In particular, it is characterized in that said step 3.6 is as follows for the wavelet variation formula:
wherein t represents the current iteration number, Tmax is the maximum iteration number, Ub is an upper boundary, Lb is a lower boundary, and in order to prevent the algorithm from exceeding the search boundary, the user sets
Is [ -1.5d]And g is 10000.
Drawings
Fig. 1 is a flowchart of a power electronic circuit fault diagnosis method for optimizing ELM based on an improved sparrow search algorithm.
Detailed Description
The technical scheme of the invention is further explained by combining the attached drawings.
The invention provides a power electronic circuit fault diagnosis method for optimizing ELM based on an improved sparrow search algorithm, which has the general idea that:
firstly, acquiring a load voltage signal of an actual power electronic circuit to be tested through a sensor, decomposing the voltage signal to obtain a characteristic vector of the signal, and dividing the characteristic vector of the signal into a test sample and a training sample. And building an ELM neural network, and taking the sum of the error rate of the training sample and the error rate of the test sample as a fitness function of the sparrow search algorithm. Aiming at the fact that the sparrow search algorithm is prone to falling into local optimum, the optimization method of the gravity center reverse learning sparrow search algorithm based on wavelet variation is provided, the global search capability is enhanced, the diversity of the population is enriched, the possibility of falling into the local optimum is reduced, and the optimization capability of the sparrow algorithm is improved. And optimizing the weight and the threshold of the ELM neural network by using a sparrow search algorithm to obtain a power electronic circuit fault diagnosis model for optimizing the ELM neural network by using the sparrow search algorithm, and inputting the test sample into the trained power electronic circuit fault diagnosis model for optimizing the ELM neural network by using the sparrow search algorithm to finish the fault diagnosis of the power electronic circuit.
As shown in fig. 1, the specific implementation of the power electronic circuit fault diagnosis method for optimizing ELM based on the improved sparrow search algorithm of the present invention includes the following steps:
step 1: the method comprises the steps of carrying out variable mode decomposition processing on collected voltage signals of the power electronic circuit, decomposing the signals through variable mode decomposition to obtain characteristic vectors of the signals, and dividing the characteristic vectors of the signals into test samples and training samples.
Step 2: building an ELM neural network, training the ELM neural network, and calculating the fitness value of each sparrow individual by selecting the sum of the error rate of a training sample and the error rate of a test sample as a fitness function of a sparrow search algorithm, wherein the fitness function is as follows:
fitness＝argmin(Tr_ErrorR-Te_ErrorR)
wherein, Tr _ ErrorR is the error rate of the training sample, and Te _ ErrorR is the error rate of the testing sample.
And step 3: setting the sparrow population scale as N and the number of discoverers as PNNumber of early-warners SNThe dimension D of the objective function, the upper bound Ub and the lower bound Lb of the initial value, and the maximum iteration number Tmax. Wherein the position of the ith sparrow in the search space is
Wherein i 1,2, N, j 1,2ijIndicating the position of the ith sparrow in the jth dimension; the fitness value of the ith sparrow is expressed as:
where f represents the fitness value.
And 4, step 4: and evaluating the current individuals and the reverse individuals by using a gravity center reverse learning strategy, and selecting N individuals with the best fitness as an initialization population. The application of the gravity center inverse learning strategy is as follows:
set in a D-dimensional search space, [ X ]1,X2,...,Xn]Are n points with mass, since the center of gravity of the whole is defined as:
if the center of gravity of a discrete uniform whole is M, a certain point X in MiThe formula of the reverse point is updated as follows:
the search space in which the reverse point is located has dynamic boundaries in which
Lbj＝min(Xij),Ubj＝max(Xij) K is [01 ]]Random numbers uniformly distributed in intervals; if the inverse solution becomes non-solvable beyond the boundary, the formula is as follows:
where rand represents a random number on (0, 1).
Selecting population by using gravity center reverse learning strategy, and randomly generating D-dimensional vector with numerical value between (0,1)
Generating 2N vectors according to formulas (1) - (4); wherein i 1,2., N, j 1,2.. D; and then calculating the fitness of each individual in the populationAnd selecting N individuals with the best fitness value as the positions of the initialized population.
And 5: calculating and sorting the fitness of sparrow populations, and selecting the P with the best fitness in each generation of populationsNSparrows as discoverers, the rest N-PNSparrow as follower, SNThe sparrows at the edges of the population can rapidly move to a safe area to obtain a better position during early warning by using a number of individuals to reconnoiter and early warn as sirens, and the positions of a finder, a follower and the sirens are updated according to a formula.
The location update formula of the discoverer is as follows:
where t denotes the current number of iterations, xijRepresenting the position information of the ith sparrow population in the jth dimension, alpha representing a random number from 0 to 1, TmaxRepresents the maximum number of iterations, Q represents a random number that follows a normal distribution, L is a matrix of 1 x D and all elements are 1, R2 e (0,1)]Represents an early warning value, and ST ∈ [0.5,1) represents a safety value. When R2<The ST represents that the early warning value is smaller than the safety value, and at the moment, no predator exists in the foraging environment, and the finder can perform extensive search operation; when R2>ST means that some sparrows in the population have found predators and give an early warning to other sparrows in the population, all of which need to fly to a safe area for foraging.
The follower's location update formula is:
wherein rand { -11} represents a randomly selected value of-1 or 1,
is the current best position of the sparrow. The formula shows that,
and adding the distance between the sparrow and each dimension of the optimal position to the current optimal sparrow position, and then dividing the sum to each dimension. Its value converges to the optimum position.
In a sparrow population, the number of dangerous sparrows, which accounts for 10% to 20% of the total number, are randomly located, and we call them sirens.
The siren location update formula is:
wherein the content of the first and second substances,
representing the current global optimum position, a random number obeying standard normal distribution is used as a step size control parameter, beta is a random number belonging to-1 to 1, fiRepresenting the fitness value of the current sparrow individual, fgRepresenting a global best fitness value, fwRepresenting the global worst fitness value. When f isi>fgThe time when sparrows are at the edge of the population, the sparrows are very easy to be attacked by predators, and the time when f isi＝fgIt is an indication that sparrows in the middle of the population are also at risk, and it is desirable to get close to other sparrows to reduce the risk of being prey.
Step 6: after iteration is completed, calculating the fitness value of each sparrow in the population and the average fitness value of the population, and performing wavelet variation operation on individual positions with fitness less than the average fitness value of the population to increase the diversity of the population, wherein the formula of the wavelet variation is as follows:
where T represents the current number of iterations, TmaxFor maximum iteration number, Ub is upper boundary, Lb is lower boundary, to prevent algorithm from exceeding search boundary, we set
Is [ -1.5d]And g is 10000.
And 7: and (3) optimizing the weight and the threshold of the ELM neural network by using the improved sparrow search algorithm to obtain a power electronic circuit fault diagnosis model for optimizing the ELM neural network by using the sparrow search algorithm, and inputting the feature vector in the step (1) into the power electronic circuit fault diagnosis model for optimizing the ELM neural network by using the sparrow search algorithm to finish training.
And 8: and inputting the test sample into a trained power electronic circuit fault diagnosis model of which the ELM neural network is optimized by a sparrow search algorithm to finish the fault diagnosis of the power electronic circuit.
The above embodiments are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modifications made on the basis of the technical scheme according to the technical idea of the present invention fall within the protection scope of the present invention.
Claims (8)
1. A power electronic circuit fault diagnosis method for optimizing ELM based on an improved sparrow search algorithm is characterized by comprising the following steps:
step (1): the method comprises the steps that collected voltage signals of the power electronic circuit are decomposed through a variational mode, and the characteristic vectors of the signals are obtained;
step (2): building an ELM neural network, training the ELM neural network, and calculating the fitness value of each sparrow individual by selecting the sum of the error rate of a training sample and the error rate of a test sample as a fitness function of a sparrow search algorithm;
and (3): according to the problems that the sparrow search algorithm is easy to fall into local optimum and insufficient in population diversity in actual operation, the sparrow search algorithm is improved;
and (4): optimizing the weight and the threshold of the ELM neural network by using an improved sparrow search algorithm to obtain a power electronic circuit fault diagnosis model for optimizing the ELM neural network by using the sparrow search algorithm, inputting the feature vector in the step (1) into the power electronic circuit fault diagnosis model for optimizing the ELM neural network by using the sparrow search algorithm, and completing training;
and (5): and inputting the test sample into a trained power electronic circuit fault diagnosis model of which the ELM neural network is optimized by a sparrow search algorithm to finish the fault diagnosis of the power electronic circuit.
2. A power electronic circuit fault diagnosis method based on improved sparrow search algorithm optimization ELM as claimed in claim 1, characterized in that the fitness function value of each sparrow in the step (2) is calculated by the following function:
fitness＝argmin(Tr_ErrorR-Te_ErrorR)
wherein, Tr _ ErrorR is the error rate of the training sample, and Te _ ErrorR is the error rate of the testing sample.
3. The method for diagnosing the fault of the power electronic circuit based on the improved sparrow search algorithm for optimizing the ELM as claimed in claim 1, wherein the operation of the improved sparrow algorithm in the step (3) comprises the following steps:
step 3.1: setting the sparrow population scale as N and the number of discoverers as PNNumber of early-warners SNThe dimension D of the objective function, the upper bound Ub and the lower bound Lb of the initial value, and the maximum iteration number Tmax；
Step 3.2: evaluating the current individuals and the reverse individuals by using a gravity center reverse learning strategy, and selecting N individuals with the best fitness as an initialization population;
step 3.3: calculating the position x of each sparrow in D-dimensional spaceiAnd fitness F of each sparrowiSelecting the sparrow F with the best initial fitness in the populationbAnd an optimal sparrow position xbAnd sparrow F with the worst initial fitnessdAnd worst sparrow position xd；
Step 3.4: calculating and sorting the fitness of sparrow populations, and selecting the P with the best fitness in each generation of populationsNSparrows as discoverers, the rest N-PNSparrow as follower, SNThe number of individuals are used for reconnaissance and early warning as alarm persons, sparrows at the edges of the population can rapidly move to a safe area during early warning so as to obtain a better position, and the positions of a finder, a follower and the alarm persons are updated according to an updating formula;
step 3.5: after iteration is finished, calculating the fitness value of each sparrow in the population and the average fitness value of the population, and performing wavelet variation operation on the individual positions with the fitness smaller than the average fitness of the population;
step 3.6: and judging whether the requirement of the maximum iteration number is met, if so, outputting the result as an initial weight and a threshold of the ELM neural network, and otherwise, returning to the step 3.3.
4. A power electronic circuit fault diagnosis method for optimizing ELM based on improved sparrow search algorithm as claimed in claim 3, characterized in that said step 3.2 includes:
assuming that the size of a sparrow is N, the dimension of the search space is D, and the position of the ith sparrow in the search space is:
wherein i 1,2, N, j 1,2ijIndicating the position of the ith sparrow in the jth dimension;
the fitness value of the ith sparrow is expressed as:
where f represents the fitness value.
5. A power electronic circuit fault diagnosis method based on improved sparrow search algorithm optimization ELM as claimed in claim 3, characterized in that in step 3.2, the gravity center reverse learning strategy is applied as follows:
set in a D-dimensional search space, [ X ]1,X2,...,Xn]Are n points with mass, since the center of gravity of the whole is defined as:
if the center of gravity of a discrete uniform whole is M, a certain point X in MiThe formula of the reverse point is updated as follows:
the search space in which the reverse point is located has dynamic boundaries in which
Lbj＝min(Xij),Ubj＝max(Xij) K is [01 ]]Random numbers uniformly distributed in intervals; if the inverse solution becomes non-solvable beyond the boundary, the formula is as follows:
where rand represents a random number on (0, 1).
6. The method for diagnosing the faults of the power electronic circuit based on the improved sparrow search algorithm for optimizing ELM is characterized in that a gravity center reverse learning strategy is used for selecting the population mode as follows:
randomly generating D-dimensional vectors with values between (0,1)
Generating 2N vectors according to formulas (3) - (6); wherein i 1,2., N, j 1,2.. D; and then calculating the fitness value of each individual in the population, and selecting N individuals with the best fitness values as the positions for initializing the population.
7. A power electronic circuit fault diagnosis method for optimizing ELM based on improved sparrow search algorithm as claimed in claim 3, wherein in step 3.4, the follower position update formula is:
wherein rand { -11} represents a randomly selected value of-1 or 1,
is the current best position of the sparrow.
8. A power electronic circuit fault diagnosis method based on improved sparrow search algorithm optimization ELM according to claim 3, characterized in that the step 3.6 is as follows for wavelet variation formula:
where T represents the current number of iterations, TmaxFor maximum iteration number, Ub is upper boundary, Lb is lower boundary, to prevent algorithm from exceeding search boundary, we set
Is [ -1.5d]And g is 10000.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110579753.2A CN113269252A (en) | 2021-05-26 | 2021-05-26 | Power electronic circuit fault diagnosis method for optimizing ELM based on improved sparrow search algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110579753.2A CN113269252A (en) | 2021-05-26 | 2021-05-26 | Power electronic circuit fault diagnosis method for optimizing ELM based on improved sparrow search algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113269252A true CN113269252A (en) | 2021-08-17 |
Family
ID=77233130
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110579753.2A Withdrawn CN113269252A (en) | 2021-05-26 | 2021-05-26 | Power electronic circuit fault diagnosis method for optimizing ELM based on improved sparrow search algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113269252A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113961362A (en) * | 2021-11-14 | 2022-01-21 | 苏州浪潮智能科技有限公司 | Process identification method, system, storage medium and equipment |
| CN114330438A (en) * | 2021-12-28 | 2022-04-12 | 福州大学 | Lower limb movement recognition method based on improved sparrow search algorithm optimization KELM |
| CN114397807A (en) * | 2021-12-02 | 2022-04-26 | 中国人民解放***箭军工程大学 | PID parameter optimization method based on improved sparrow algorithm |
| CN114743089A (en) * | 2021-12-22 | 2022-07-12 | 宁波送变电建设有限公司 | Image recognition GIS fault diagnosis device and method based on SSA-SVM |
| CN114969995A (en) * | 2021-11-22 | 2022-08-30 | 昆明理工大学 | Rolling bearing early fault intelligent diagnosis method based on improved sparrow search and acoustic emission |
| CN115150284A (en) * | 2022-06-13 | 2022-10-04 | 上海工程技术大学 | Communication network topology optimization method and system based on improved sparrow algorithm |
| CN115544890A (en) * | 2022-10-19 | 2022-12-30 | 重庆工商大学 | Short-term power load prediction method and system |
| CN117574057A (en) * | 2023-11-24 | 2024-02-20 | 昆明理工大学 | Intelligent fault diagnosis method for vertical water pump unit |
| CN117665510A (en) * | 2023-12-08 | 2024-03-08 | 西南交通大学 | High-voltage cabinet insulating material reliability assessment method based on electrification characteristics |
-
2021
- 2021-05-26 CN CN202110579753.2A patent/CN113269252A/en not_active Withdrawn
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113961362A (en) * | 2021-11-14 | 2022-01-21 | 苏州浪潮智能科技有限公司 | Process identification method, system, storage medium and equipment |
| CN113961362B (en) * | 2021-11-14 | 2024-01-16 | 苏州浪潮智能科技有限公司 | Process identification method, system, storage medium and equipment |
| CN114969995A (en) * | 2021-11-22 | 2022-08-30 | 昆明理工大学 | Rolling bearing early fault intelligent diagnosis method based on improved sparrow search and acoustic emission |
| CN114969995B (en) * | 2021-11-22 | 2024-02-27 | 昆明理工大学 | Rolling bearing early fault intelligent diagnosis method based on improved sparrow search and acoustic emission |
| CN114397807B (en) * | 2021-12-02 | 2024-03-29 | 中国人民解放***箭军工程大学 | PID parameter optimization method based on improved sparrow algorithm |
| CN114397807A (en) * | 2021-12-02 | 2022-04-26 | 中国人民解放***箭军工程大学 | PID parameter optimization method based on improved sparrow algorithm |
| CN114743089A (en) * | 2021-12-22 | 2022-07-12 | 宁波送变电建设有限公司 | Image recognition GIS fault diagnosis device and method based on SSA-SVM |
| CN114743089B (en) * | 2021-12-22 | 2024-06-04 | 宁波送变电建设有限公司 | Image recognition GIS fault diagnosis device and method based on SSA-SVM |
| CN114330438A (en) * | 2021-12-28 | 2022-04-12 | 福州大学 | Lower limb movement recognition method based on improved sparrow search algorithm optimization KELM |
| CN115150284A (en) * | 2022-06-13 | 2022-10-04 | 上海工程技术大学 | Communication network topology optimization method and system based on improved sparrow algorithm |
| CN115150284B (en) * | 2022-06-13 | 2023-04-25 | 上海工程技术大学 | Communication network topology optimization method and system based on improved sparrow algorithm |
| CN115544890A (en) * | 2022-10-19 | 2022-12-30 | 重庆工商大学 | Short-term power load prediction method and system |
| CN117574057B (en) * | 2023-11-24 | 2024-05-28 | 昆明理工大学 | Intelligent fault diagnosis method for vertical water pump unit |
| CN117574057A (en) * | 2023-11-24 | 2024-02-20 | 昆明理工大学 | Intelligent fault diagnosis method for vertical water pump unit |
| CN117665510A (en) * | 2023-12-08 | 2024-03-08 | 西南交通大学 | High-voltage cabinet insulating material reliability assessment method based on electrification characteristics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113269252A (en) | Power electronic circuit fault diagnosis method for optimizing ELM based on improved sparrow search algorithm | |
| CN105488528B (en) | Neural network image classification method based on improving expert inquiry method | |
| CN105608444B (en) | A kind of wild animal image-recognizing method for automatic Pilot | |
| CN105488563A (en) | Deep learning oriented sparse self-adaptive neural network, algorithm and implementation device | |
| CN111988329B (en) | Network intrusion detection method based on deep learning | |
| CN107223260B (en) | Method for dynamically updating classifier complexity | |
| CN110287985B (en) | Depth neural network image identification method based on variable topology structure with variation particle swarm optimization | |
| CN107403188A (en) | A kind of quality evaluation method and device | |
| Wu | A density adjustment based particle swarm optimization learning algorithm for neural network design | |
| CN113484785A (en) | Battery health state estimation and residual life prediction method based on sparrow search and least square support vector machine | |
| CN114117787A (en) | Short-term wind power prediction method based on SSA (simple sequence analysis) optimization BP (back propagation) neural network | |
| CN111062511A (en) | Aquaculture disease prediction method and system based on decision tree and neural network | |
| Liang et al. | Recent advances in particle swarm optimization via population structuring and individual behavior control | |
| CN111612246B (en) | Method, device, equipment and storage medium for predicting heavy metal content of farmland soil | |
| Jiang et al. | Hybrid PSO and GA for neural network evolutionary in monthly rainfall forecasting | |
| Skorpil et al. | Back-propagation and k-means algorithms comparison | |
| Yadav et al. | Use of different approaches to model catch per unit effort (CPUE) abundance of fish | |
| Wang et al. | Predator-prey biogeography-based optimization for bio-inspired visual attention | |
| Liu et al. | Swarm intelligence for classification of remote sensing data | |
| Hu et al. | Research on pest and disease recognition algorithms based on convolutional neural network | |
| Boniecki | The Kohonen neural network in classification problems solving in agricultural engineering | |
| Liu et al. | Research on prediction about fruit tree diseases and insect pests based on neural network | |
| Houssein et al. | Salp swarm algorithm: modification and application | |
| CN117784615B (en) | Fire control system fault prediction method based on IMPA-RF | |
| CN110837762A (en) | Convolutional neural network pedestrian recognition method based on GoogLeNet |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210817 |