CN112288147B - Method for predicting insulation state of generator stator by BP-Adaboost strong predictor - Google Patents
Method for predicting insulation state of generator stator by BP-Adaboost strong predictor Download PDFInfo
- Publication number
- CN112288147B CN112288147B CN202011120840.3A CN202011120840A CN112288147B CN 112288147 B CN112288147 B CN 112288147B CN 202011120840 A CN202011120840 A CN 202011120840A CN 112288147 B CN112288147 B CN 112288147B
- Authority
- CN
- China
- Prior art keywords
- initial
- neural network
- sample data
- particle
- predicting
- 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.)
- Active
Links
- 238000009413 insulation Methods 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 33
- 239000002245 particle Substances 0.000 claims abstract description 67
- 238000013528 artificial neural network Methods 0.000 claims abstract description 61
- 230000015556 catabolic process Effects 0.000 claims abstract description 16
- 238000002922 simulated annealing Methods 0.000 claims abstract description 14
- 230000032683 aging Effects 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 239000013598 vector Substances 0.000 claims abstract description 5
- 230000001066 destructive effect Effects 0.000 claims description 8
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000000137 annealing Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000010521 absorption reaction Methods 0.000 claims description 3
- 230000006978 adaptation Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 claims description 3
- 230000010287 polarization Effects 0.000 claims description 3
- 238000012847 principal component analysis method Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims 1
- 238000012423 maintenance Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Water Supply & Treatment (AREA)
- Entrepreneurship & Innovation (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Tests Of Circuit Breakers, Generators, And Electric Motors (AREA)
Abstract
The invention discloses a method for predicting the insulation state of a generator stator by a BP-Adaboost strong predictor, which comprises the following steps: 1) Collecting initial sample data, preprocessing the initial sample data, extracting the characteristics of the preprocessed initial sample data, and taking the extracted characteristic vector as sample data; 2) Optimizing a particle swarm algorithm by using a simulated annealing algorithm to obtain a simulated annealing particle swarm algorithm, and calculating initial weights and thresholds of a plurality of BP neural networks based on the simulated annealing particle swarm algorithm; 3) Constructing a plurality of BP neural networks according to the initial weights and the threshold values of the BP neural networks obtained in the step 2), and training the BP neural networks; 4) Determining the weight of each BP neural network according to the training error of each BP neural network; 5) The method comprises the steps of forming a strong predictor, and then predicting breakdown voltage capable of representing the insulation aging state of the stator of the generator by using the strong predictor.
Description
Technical Field
The invention belongs to the field of generator stator insulation state prediction and the field of artificial intelligence, and relates to a method for predicting the generator stator insulation state by a BP-Adaboost strong predictor.
Background
The operational reliability of the large-scale generator is related to the operational stability of the power grid, and is one of the key equipment of the power system. One of the threats of safe operation is mainly from the reliability problem of an insulation system, and in the operation process of a motor, a stator winding is subjected to the combined action of multiple factors such as electricity, heat, machinery, environment and the like, so that the insulation performance is gradually deteriorated. In the case of severe ageing of the insulation, insulation faults of the generator may be caused.
In order to avoid stator insulation faults in the running process of the generator, a method of regular shutdown maintenance is generally adopted, but the method is not flexible to implement, and the waste of the load of the generator is caused. If the state of the stator insulation of the generator can be predicted in advance, and corresponding maintenance measures are adopted according to the predicted state, the operation efficiency and economic benefit of the generator can be greatly improved.
Most of the existing generator stator insulation state prediction methods adopt a formula method. The method makes great contribution to the state prediction of the stator bar, but has the limitation that the actual insulation state of the generator cannot be accurately estimated. In summary, the method for accurately predicting the insulation state of the stator of the generator has great significance for the stable operation of the power grid, and has good economic benefit and wide application prospect.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a method for predicting the insulation state of a generator stator by using a BP-Adaboost strong predictor, which can accurately predict the insulation state of the generator stator.
In order to achieve the above purpose, the method for predicting the insulation state of the stator of the generator by the BP-Adaboost strong predictor comprises the following steps:
1) Collecting initial sample data, wherein the initial sample is a nondestructive characteristic parameter and a destructive parameter capable of representing the insulation aging state of a generator stator, preprocessing the initial sample data, extracting the characteristics of the preprocessed initial sample data, and taking the extracted characteristic vector as sample data;
2) Optimizing a particle swarm algorithm by using a simulated annealing algorithm to obtain a simulated annealing particle swarm algorithm, and calculating initial weights and thresholds of a plurality of BP neural networks based on the simulated annealing particle swarm algorithm;
3) Constructing a plurality of BP neural networks according to the initial weight and the threshold value of each BP neural network obtained in the step 2), and training each BP neural network by utilizing the sample data obtained in the step 1);
4) Determining the weight of each BP neural network according to the training error of each BP neural network;
5) And combining the BP neural networks by using an Adaboost iterative algorithm according to the weight of each BP neural network to form a strong predictor, and predicting breakdown voltage capable of representing the insulation aging state of the generator stator by using the strong predictor.
The non-destructive characteristic parameters comprise absorption ratio, leakage current, polarization index, dielectric loss increment and maximum discharge amount increment rate; the destructive parameter is the breakdown voltage.
The specific operation of preprocessing the initial sample data is as follows: and determining an abnormal value of the initial sample data by using the MATLAB box diagram to remove the abnormal value, and then performing interpolation of the initial sample data by using an approximate interpolation method or a Newton interpolation method.
And extracting the characteristics of the preprocessed sample data by adopting a kernel principal component analysis method.
The specific operation of the step 2) is as follows:
21 Setting a learning factor, the evolution times and the population scale, and initializing the position and the speed of particles, wherein the position of each particle represents a group of initial weights and thresholds of the BP neural network;
22 Calculating fitness of the initial particles;
23 Selecting individual optimal particles p according to the fitness of the initial particles best Population optimal particle g best ；
24 Setting an initial annealing temperature T, wherein T=fitnesszbest/ln (5), and fitnesszbest is an initial population optimal particle g best Is adapted to the degree of adaptation of (a);
25 Judging whether the evolution times of the current particles meet the preset evolution times conditions, if so, outputting the positions of the optimal particles of the group, namely the optimal initial weight and threshold value of the BP neural network, otherwise, turning to the step 26);
26 Updating the position and velocity of each particle;
27 Calculating the fitness of each new particle;
28 Calculating the difference Vf between the fitness of the new particle and the fitness of the previous particle, and accepting the speed and the position of the new particle when Vf is smaller than 0 or exp (-Vf/T) > rand, otherwise, retaining the speed and the position of the old particle;
29 Updating individual best particles and group best particles through fitness;
210 A desuperheating operation, and then goes to step 24).
The specific operation of the step 5) is as follows: and predicting breakdown voltage by utilizing each trained BP neural network, then carrying out weighted calculation on the prediction result of each BP neural network according to the weight of each BP neural network, and taking the weighted calculation result as the breakdown voltage capable of representing the insulation aging state of the generator stator.
The invention has the following beneficial effects:
the method for predicting the insulation state of the stator of the generator by using the BP-Adaboost strong predictor optimizes a particle swarm algorithm by using a simulated annealing algorithm to obtain a simulated annealing particle swarm algorithm, calculates initial weights and thresholds of a plurality of BP neural networks based on the simulated annealing particle swarm algorithm, constructs and trains the BP neural networks based on the initial weights and thresholds, combines the BP neural networks by using the Adaboost iterative algorithm to form the strong predictor, and can predict breakdown voltage capable of representing the insulation aging state of the stator of the generator based on the strong predictor to realize accurate prediction of the insulation state of the stator of the generator.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The invention is described in further detail below with reference to the attached drawing figures:
referring to fig. 1, the method for predicting the insulation state of the stator of the generator by using the BP-Adaboost strong predictor comprises the following steps:
acquiring initial sample data, wherein the initial sample data are nondestructive characteristic parameters and destructive parameters capable of representing the insulation aging state of a generator stator, and the nondestructive characteristic parameters comprise an absorption ratio, leakage current, polarization index, dielectric loss increment and maximum discharge amount increment rate; the destructive parameter is breakdown voltage, and a better method for evaluating the insulation state of the stator bar of the generator is to infer the residual breakdown voltage of the stator insulation, so that the nondestructive parameter is used as the input of a strong predictor, the breakdown voltage is used as the target output, and all initial sample data are obtained through a multi-factor aging test platform.
Because the sample data obtained by the test platform may have the problems of low instrument precision, manual misoperation and the like, the initial sample data needs to be preprocessed, wherein the preprocessing comprises missing value processing and abnormal value processing, and the MATLAB box line diagram is adopted to identify abnormal values of the initial sample data so as to reject the abnormal data; because the sample data has a sequential rule, the sample data is divided into training data and verification data, and data interpolation is carried out by adopting a nearest interpolation method or a Newton interpolation method, wherein the Newton difference algorithm is an optimal scheme.
And extracting features of the preprocessed sample data by adopting a kernel principal component analysis method, and particularly mapping the preprocessed sample data to a multidimensional feature space, so that the sample data has better separability. And performing principal component analysis on the mapping data in the multidimensional feature space to extract feature vectors, and taking the extracted feature vectors as sample data of the BP-Adaboost strong predictor.
After clean data is obtained, the sample data is divided into training data and prediction data, and the training data and the prediction data are normalized, so that errors caused by overlarge magnitude differences of characteristic parameters capable of representing the insulation aging state of the generator stator are reduced.
Setting the number of input nodes, the number of hidden layer nodes and the number of output nodes of the BP neural network to form a basic structure of the neural network.
The specific process of Adaboost iteration is as follows:
setting the number t of BP neural networks to be combined and optimized and initializing sample weights, wherein t=5, namely, combining the results of 5 BP neural networks after optimization as the output of a strong predictor, and giving the same weight D to each sample in the beginning stage 1 (i) The method comprises the following steps:
where n is the number of samples.
Judging whether the current circulation times are equal to the number of the neural networks, outputting the predicted result of the weight and breakdown voltage of each optimized BP neural network in the final integrated model when the current circulation times are equal to the number of the neural networks, otherwise, continuing the circulation evolution.
And calling a sub-function for simulating the initial weight and the threshold value of the annealing particle swarm algorithm to optimize the neural network, wherein the output result of the sub-function is the initial weight and the threshold value of the optimized neural network. The prediction precision of the BP neural network can be improved by setting a proper initial weight and a proper threshold, wherein the modeling process of the subfunction is as follows:
1) Learning factors, number of evolutions, and population size. In the invention, two learning factors are 1.49445; the evolution times were 50; population size was 20. The position and velocity of the particles are initialized. The location of each particle represents a set of weights and thresholds for the neural network, i.e., a potential solution.
2) Calculating the fitness of the initial particles;
specifically, the position of the particle is given to the BP neural network by using the subfunction as a weight and a threshold, the BP neural network is trained, and then the average value of the absolute prediction error is output as the fitness of the simulated annealing particle swarm algorithm.
3) Individual optimal particles p are selected according to the fitness of the initial particles best Population optimal particle g best ；
4) Setting an initial annealing temperature T, wherein T=fitnesszbest/ln (5), and fitnesszbest is an initial population optimal particle g best Is adapted to the degree of adaptation of (a);
5) Judging whether the evolution times of the current particles meet the preset evolution times conditions, if so, outputting the positions of the optimal particles of the group, namely the optimal initial weight and threshold value of the BP neural network, and if not, turning to the step 6);
6) Updating the position and speed of each particle;
7) Calculating the fitness of each new particle;
8) Calculating a difference Vf between the fitness of the new particle and the fitness of the previous particle, and accepting the speed and the position of the new particle when Vf is smaller than 0 or exp (-Vf/T) > rand, otherwise, retaining the speed and the position of the old particle;
9) Updating individual optimal particles and group optimal particles through fitness;
10 A temperature-removing operation is performed, and then the process goes to step 4).
Obtaining an optimal initial weight and a threshold value of the BP neural network based on a simulated annealing particle swarm algorithm, giving the optimal initial weight and the optimal threshold value to the BP neural network, and training the BP neural network;
according to training results of the neural networks, calculating total errors of the optimized BP neural networks, setting a current network to be trained as a j-th optimized BP neural network, j=1, 2, and the sample weight is D j (i) Calculate training error E for n samples j (i) I=1, 2,..n, T is a threshold;
the total error of the jth optimized BP neural network is:
meanwhile, the sample weight of the j+1th neural network is updated according to the following principle: when the sample training error is greater than a threshold value T, increasing the weight of the sample, otherwise, keeping the weight of the sample unchanged, wherein the sample weight of the j+1th neural network is as follows:
weighting the sample by D j+1 (i) Carrying out normalization treatment;
calculating the weight a of the jth optimized BP neural network in the final integrated model j ：
And carrying out weight normalization on the weight of each optimized BP neural network in a final integrated model, and carrying out integrated output on breakdown voltage results predicted by 5 optimized BP neural networks to obtain the breakdown voltage capable of representing the insulation aging state of the generator stator.
Claims (4)
1. The method for predicting the insulation state of the stator of the generator by using the BP-Adaboost strong predictor is characterized by comprising the following steps of:
1) Collecting initial sample data, wherein the initial sample is a nondestructive characteristic parameter and a destructive parameter capable of representing the insulation aging state of a generator stator, preprocessing the initial sample data, extracting the characteristics of the preprocessed initial sample data, and taking the extracted characteristic vector as sample data;
2) Optimizing a particle swarm algorithm by using a simulated annealing algorithm to obtain a simulated annealing particle swarm algorithm, and calculating initial weights and thresholds of a plurality of BP neural networks based on the simulated annealing particle swarm algorithm;
3) Constructing a plurality of BP neural networks according to the initial weight and the threshold value of each BP neural network obtained in the step 2), and training each BP neural network by utilizing the sample data obtained in the step 1);
4) Determining the weight of each BP neural network according to the training error of each BP neural network;
5) Combining the BP neural networks by using an Adaboost iterative algorithm according to the weight of each BP neural network to form a strong predictor, and predicting breakdown voltage capable of representing the insulation aging state of the generator stator by using the strong predictor;
the specific operation of the step 2) is as follows:
21 Setting a learning factor, the evolution times and the population scale, and initializing the position and the speed of particles, wherein the position of each particle represents a group of initial weights and thresholds of the BP neural network;
22 Calculating fitness of the initial particles;
23 Selecting individual optimal particles p according to the fitness of the initial particles best Population optimal particle g best ；
24 Setting an initial annealing temperature T, wherein t=fitnesszbest/ln (5), fitnesszbest is the best particle g of the initial population best Is adapted to the degree of adaptation of (a);
25 Judging whether the evolution times of the current particles meet the preset evolution times conditions, if so, outputting the positions of the optimal particles of the group, namely the optimal initial weight and threshold value of the BP neural network, otherwise, turning to the step 26);
26 Updating the position and velocity of each particle;
27 Calculating the fitness of each new particle;
28 Calculating the difference Vf between the fitness of the new particle and the fitness of the previous particle, and accepting the speed and the position of the new particle when Vf is smaller than 0 or exp (-Vf/T) > rand, otherwise, retaining the speed and the position of the old particle;
29 Updating individual best particles and group best particles through fitness;
210 A temperature-reducing operation is performed, and then the process goes to the step 24);
the specific operation of the step 5) is as follows: and predicting breakdown voltage by utilizing each trained BP neural network, then carrying out weighted calculation on the prediction result of each BP neural network according to the weight of each BP neural network, and taking the weighted calculation result as the breakdown voltage capable of representing the insulation aging state of the generator stator.
2. The method for predicting the insulation state of a generator stator by using the BP-Adaboost strong predictor according to claim 1, wherein the non-destructive characteristic parameters comprise an absorption ratio, a leakage current, a polarization index, a dielectric loss increment and a maximum discharge amount increment rate; the destructive parameter is the breakdown voltage.
3. The method for predicting the insulation state of a stator of a generator by using the BP-Adaboost strong predictor according to claim 1, wherein the specific operation of preprocessing initial sample data is as follows: and determining an abnormal value of the initial sample data by using the MATLAB box diagram to remove the abnormal value, and then performing interpolation of the initial sample data by using an approximate interpolation method or a Newton interpolation method.
4. The method for predicting the insulation state of the stator of the generator by using the BP-Adaboost strong predictor according to claim 1, wherein a kernel principal component analysis method is adopted to perform feature extraction on the preprocessed sample data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011120840.3A CN112288147B (en) | 2020-10-19 | 2020-10-19 | Method for predicting insulation state of generator stator by BP-Adaboost strong predictor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011120840.3A CN112288147B (en) | 2020-10-19 | 2020-10-19 | Method for predicting insulation state of generator stator by BP-Adaboost strong predictor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112288147A CN112288147A (en) | 2021-01-29 |
| CN112288147B true CN112288147B (en) | 2023-06-30 |
Family
ID=74496501
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011120840.3A Active CN112288147B (en) | 2020-10-19 | 2020-10-19 | Method for predicting insulation state of generator stator by BP-Adaboost strong predictor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112288147B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113569507A (en) * | 2021-09-27 | 2021-10-29 | 中国人民解放军海军工程大学 | Machine learning-based stator bar insulation aging state composite prediction method |
| CN115471501B (en) * | 2022-10-31 | 2023-10-13 | 长江勘测规划设计研究有限责任公司 | Method and system for identifying air gap distribution state of generator on line by utilizing machine vision |
| CN115830411B (en) * | 2022-11-18 | 2023-09-01 | 智慧眼科技股份有限公司 | Biological feature model training method, biological feature extraction method and related equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102135593A (en) * | 2010-12-28 | 2011-07-27 | 太原理工大学 | On-line diagnosis and evaluation method of insulation state of large electric machine |
| CN105137349A (en) * | 2015-07-22 | 2015-12-09 | 广东电网有限责任公司电力科学研究院 | Large-scale generator stator winding major insulation aging state test method based on frequency domain spectroscopy |
| CN107644231A (en) * | 2017-09-19 | 2018-01-30 | 广东工业大学 | A kind of generator amature method for diagnosing faults and device |
| CN108520154A (en) * | 2018-04-16 | 2018-09-11 | 重庆邮电大学 | Generator stator end winding construction optimization method based on particle cluster algorithm and support vector machines |
| CN111563622A (en) * | 2020-04-30 | 2020-08-21 | 西安交通大学 | Stator bar insulation aging degree prediction method based on gray level co-occurrence matrix and deep learning |
| CN111709182A (en) * | 2020-05-25 | 2020-09-25 | 温州大学 | Electromagnet fault prediction method based on SA-PSO (SA-particle swarm optimization) optimized BP (Back propagation) neural network |
| WO2020191800A1 (en) * | 2019-03-27 | 2020-10-01 | 东北大学 | Method for predicting remaining service life of lithium-ion battery employing wde-optimized lstm network |
-
2020
- 2020-10-19 CN CN202011120840.3A patent/CN112288147B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102135593A (en) * | 2010-12-28 | 2011-07-27 | 太原理工大学 | On-line diagnosis and evaluation method of insulation state of large electric machine |
| CN105137349A (en) * | 2015-07-22 | 2015-12-09 | 广东电网有限责任公司电力科学研究院 | Large-scale generator stator winding major insulation aging state test method based on frequency domain spectroscopy |
| CN107644231A (en) * | 2017-09-19 | 2018-01-30 | 广东工业大学 | A kind of generator amature method for diagnosing faults and device |
| CN108520154A (en) * | 2018-04-16 | 2018-09-11 | 重庆邮电大学 | Generator stator end winding construction optimization method based on particle cluster algorithm and support vector machines |
| WO2020191800A1 (en) * | 2019-03-27 | 2020-10-01 | 东北大学 | Method for predicting remaining service life of lithium-ion battery employing wde-optimized lstm network |
| CN111563622A (en) * | 2020-04-30 | 2020-08-21 | 西安交通大学 | Stator bar insulation aging degree prediction method based on gray level co-occurrence matrix and deep learning |
| CN111709182A (en) * | 2020-05-25 | 2020-09-25 | 温州大学 | Electromagnet fault prediction method based on SA-PSO (SA-particle swarm optimization) optimized BP (Back propagation) neural network |
Non-Patent Citations (2)
| Title |
|---|
| Experimental research on stator insulation system of 5 MW offshore Wind turbine generator under simulated marine environments;Liu Xuezhong;《2013 Annual Report Conference on Electrical Insulation and Dielectric Phenomena》;摘要 * |
| 基于GA-SAPSO优化BP神经网络的光伏系统发电量预测;黄超;葛愿;葛宜然;张刚;;宿州学院学报(03);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112288147A (en) | 2021-01-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112288147B (en) | Method for predicting insulation state of generator stator by BP-Adaboost strong predictor | |
| CN110348615B (en) | Cable line fault probability prediction method based on ant colony optimization support vector machine | |
| CN110083951B (en) | Solid insulation life prediction method based on relevant operation data of transformer | |
| CN111523778A (en) | Power grid operation safety assessment method based on particle swarm algorithm and gradient lifting tree | |
| CN114460445B (en) | Transformer aging unavailability evaluation method considering aging threshold and service life | |
| Liu et al. | Combined forecasting method of dissolved gases concentration and its application in condition-based maintenance | |
| CN110929835B (en) | Novel silicon carbide-based aviation power converter fault diagnosis method and system | |
| Huang et al. | Research on transformer fault diagnosis method based on GWO optimized hybrid kernel extreme learning machine | |
| CN116842337A (en) | Transformer fault diagnosis method based on LightGBM (gallium nitride based) optimal characteristics and COA-CNN (chip on board) model | |
| CN114595883A (en) | Oil-immersed transformer residual life personalized dynamic prediction method based on meta-learning | |
| CN114280490A (en) | Lithium ion battery state of charge estimation method and system | |
| CN113554229A (en) | Three-phase voltage unbalance abnormality detection method and device | |
| CN117454939A (en) | Wind power prediction method for optimizing BP neural network based on NSABO-Gold algorithm | |
| CN117077052A (en) | Dry-type transformer abnormality detection method based on working condition identification | |
| CN115840119A (en) | Power cable line degradation diagnostic system and method using database samples | |
| CN115292820A (en) | Method for predicting residual service life of urban rail train bearing | |
| CN115239971A (en) | GIS partial discharge type recognition model training method, recognition method and system | |
| CN113884936A (en) | Lithium ion battery health state prediction method based on ISSA coupling DELM | |
| CN114117937A (en) | Method and system for identifying key nodes of cascading failures of power system | |
| CN111967593A (en) | Method and system for processing abnormal data based on modeling | |
| CN117390418B (en) | Transient stability evaluation method, system and equipment for wind power grid-connected system | |
| Qiu et al. | Fault prediction of elevator door lock based on MPGA-BP algorithm | |
| Ma et al. | Risk Assessment of Tower Transmission Based on Insulator Online Monitoring | |
| CN115641152B (en) | Method, device, equipment and storage medium for identifying faults of comprehensive energy system | |
| CN117634652B (en) | Dam deformation interpretable prediction method based on machine learning |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |