CN112307906B - Energy storage battery fault classification feature screening dimension reduction method under neighbor propagation clustering - Google Patents
Energy storage battery fault classification feature screening dimension reduction method under neighbor propagation clustering Download PDFInfo
- Publication number
- CN112307906B CN112307906B CN202011094254.6A CN202011094254A CN112307906B CN 112307906 B CN112307906 B CN 112307906B CN 202011094254 A CN202011094254 A CN 202011094254A CN 112307906 B CN112307906 B CN 112307906B
- Authority
- CN
- China
- Prior art keywords
- feature
- characteristic
- energy storage
- features
- cluster
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method for screening and dimension reduction of fault classification characteristics of an energy storage battery under neighbor propagation clustering. The implementation process of the method is as follows: firstly, acquiring terminal voltage signal data samples of N energy storage batteries with different faults in the process of completing one-time charge and discharge, and excavating characteristics of terminal voltage signals to form a characteristic set; then, defining a similarity matrix of the features by utilizing cosine similarity, clustering each feature based on a neighbor propagation method to form a plurality of feature clusters; calculating the mean value and the standard deviation of the mean value of each feature in the same cluster under different types of samples, and defining the feature with the largest standard deviation of the mean value as the typical feature for representing the feature cluster; and taking typical features of each feature cluster as the filtered dimensionality-reduced features for classification tasks. According to the invention, the salient features for classification tasks can be automatically screened out, the problem of dimensional explosion under the condition of small samples is relieved, the adverse effects of redundant and invalid features on the performance of the classifier are reduced, and the classification accuracy is improved.
Description
Technical field:
the invention relates to the technical field of energy storage batteries, in particular to a method for screening and dimension reduction of fault classification characteristics of an energy storage battery under neighbor propagation clustering.
The background technology is as follows:
under the development of machine learning and artificial intelligence technology, the automation level of classification tasks of fault diagnosis of power equipment and energy storage systems is greatly improved. The feature engineering and the learner are the key and core for solving the traditional classification task, and the significance of the mined features and the generalization capability of the learner are important conditions for restricting the accuracy, the robustness and the popularization capability of the classification task. At present, a great deal of research exists around the design and improvement of classification models, and in the face of different application requirements, feature extraction is not a uniform and definite effective method. Often defined in terms of an index with a physical meaning, which is prone to redundancy and the presence of invalid features. Meanwhile, under the condition of limited data samples, the existence of invalid and redundant features can cause too high feature space dimension, so that sample spacing is too large, a dimension explosion phenomenon is formed, and the generalization capability of the classification model is reduced. Therefore, in feature engineering, it is necessary to select or transform the extracted feature variables to reduce the feature space dimension or enhance the significance of the features.
Currently, existing feature selection manners for fault diagnosis of energy storage systems can be broadly classified into a filtering type, a packing type and an embedded type. The filtering method is characterized in that the characteristic selection is carried out according to the characteristics of the data set, and then the learner is trained, namely the characteristic selection process is independent of the learner, and has the advantages of small calculated amount and no dependence on the particularity of a learning machine model, and has the disadvantages that the directional characteristic selection is difficult to carry out in a classification task-oriented mode based on the characteristics of the data; the wrapped method directly takes the performance of the used learning period as the evaluation criterion of the feature subset after feature selection, and has the advantages that the learner performance is taken as an index to provide a direction for the feature selection process, but the learner is required to be trained for many times in the feature selection process, so that the calculation cost is high; the embedded method is characterized in that a characteristic selection process is related to a learner, the characteristic selection process is fused with a learner training process, and the characteristic selection is automatically carried out in the learner training process. Therefore, a new screening and dimension reduction method for fault diagnosis of the energy storage system is needed.
The invention comprises the following steps:
in order to improve the applicability of the fault diagnosis feature selection method of the energy storage system and reduce the feature selection calculated amount, the invention adopts a filtering type feature selection process, uses the clustering thought as a reference, realizes the distinction of feature sets based on neighbor propagation clustering, analyzes the expressive power of each feature in the similar feature clusters on samples of different categories, and selects the typical feature with the maximized distinction capability. The feature space dimension reduction is realized while the fault diagnosis feature selection of the energy storage system is completed, and a foundation is laid for the design of a high-performance learning machine model.
The invention adopts the technical scheme that:
a method for screening and dimension reduction of fault classification features of energy storage batteries under neighbor propagation clustering comprises the following steps:
step 1.1, acquiring N terminal voltage signal data samples of R failed energy storage batteries in the process of completing one-time charge and discharge by utilizing an experimental measurement mode;
step 1.2, defining and normalizing the feature vector of the voltage characteristics of the energy storage end represented by m feature variables according to a common feature extraction mode, namely the feature vector of the nth sample
Form feature set a= { (X) (n) ,L (n) )|n＝1,2,…,N；L (n) ∈{L 1 ,L 2 ,…,L R -expressed in matrix form as follows:
the row represents a sample, the first m represents a characteristic, and the m+1 represents a category of the sample;
step 2, evaluating the similarity degree among the features by utilizing cosine similarity, clustering the features by utilizing a neighbor propagation clustering method, and forming a plurality of feature clusters; the specific process is as follows:
step 2.1, vectors are composed with values of all samples under characteristic variables, i.e
i=1, 2, …, N, and calculating the cosine similarity sim between every two features, where the cosine similarity between the i-th feature and the j-th feature is:
and forming a symmetrical square matrix taking the number of the features as the number of rows (columns) according to the cosine similarity among the features, namely a similarity matrix S, wherein the negative value of the cosine similarity among the element features in the square matrix is as follows:
step 2.2, let t=1, initialize the attraction degree matrix Re t And a home degree matrix Av t And the attraction degree matrix Re t And a home degree matrix Av t The damping coefficient zeta and the maximum iteration number T are defined as row and column square matrixes which are equal to the similarity matrix S;
step (a)2.3 calculating and updating the attraction degree matrix Re of the t+1 generation t+1 Elements of (a) and (b);
step 2.4, calculating and updating the attribution degree matrix Av of the t+1 generation t+1 Elements of (a) and (b);
step 2.5, judging whether the maximum iteration times are reached, namely T is less than or equal to T; if yes, t=t+1 returns to step 2.3, if not, step 2.6 is entered;
step 2.6 let e=re t+1 +Av t+1 Diagonalizing the matrix E, counting elements larger than zero on the diagonal, and counting the characteristics of row (column) numbers where the elements larger than zero are located, namely, a neighbor propagation clustering center, selecting columns where the clustering centers are located in the similarity matrix S, selecting the category attribution of other characteristics according to the rows, and forming a plurality of characteristic clusters; assuming that the diagonal elements ei and ej in the diagonalization matrix E are greater than zero, then feature x ei And x ej For the clustering center, the ei and ej columns in the similarity matrix S are selected, and according to the sizes of-sim (ek, ei) and-sim (ek, ej), if-sim (ek, ei) is more than or equal to-sim (ek, ej), the ek-th feature x ek Similar to feature x ei Belonging to the feature x ei Is a category of (2); if-sim (ek, ei)<Sim (ek, ej), ek th feature x ek Similar to feature x ej Belonging to the feature x ej Is a category of (2); so that two dry feature clusters can be formed for all the features, and the centers of the feature clusters are respectively formed by the features x ek And feature x ej Representative of;
step 3, screening the characteristics which are most favorable for fault diagnosis of the energy storage battery in the same characteristic cluster, and defining the characteristics as typical characteristics of the characteristic cluster; the method specifically comprises the following steps:
step 3.1, calculating the average value of the voltage characteristic values of the sample ends of different fault energy storage batteries under each similar characteristic in the same characteristic cluster according to the dividing result of the characteristic cluster obtained in the step 2; let H (k) = { x in the kth feature cluster ki ,x kj ,..}, feature x ki The average value of the characteristic values of different types of samples is mu ki ＝[μ ki,1 ,μ ki,2 ,...,μ ki,R ]Wherein the L < th > is r Class sample at feature x ki The lower mean value can be expressed as
Calculating standard deviation of voltage characteristic value mean values of sample ends of different fault energy storage batteries in the same characteristic cluster in the step 3.1, and defining the characteristic with the largest standard deviation (the most obvious difference of different fault energy storage batteries under the characteristic) in the characteristic cluster as the typical characteristic for representing the characteristic cluster; let the kth feature cluster H (k) contain two feature variables, i.e., H (k) = { x ki ,x kj Characteristic x ki And x kj Sample terminal voltage characteristic value mean { mu ] of different fault energy storage batteries ki ,μ kj Standard deviations of } are respectively
And->
Then selecting the feature with the largest standard deviation as representative feature x of the feature cluster H (k) ki (σ ki ≥σ kj )+x kj (σ ki ＜σ kj )；
Step 4, forming a screened dimension reduction feature set according to typical features of each feature cluster; according to the typical characteristics of each characteristic cluster obtained in the step 3.2, removing other characteristics of atypical characteristics in the step 1.2 to form a new characteristic set for describing difference of fault terminal voltage of R energy storage batteries, realizing characteristic screening and dimension reduction processes, and being used for designing a model of a subsequent learning machine to finish fault diagnosis tasks of the energy storage batteries.
Compared with the closest prior art, the method has the advantages that in the technical scheme, the similarity degree of different features is described by cosine similarity, then, a proximity propagation clustering method is adopted, feature variables with high similarity are aggregated to form a feature cluster, the average value of each feature in the same cluster in different types of samples is analyzed, the standard deviation of the average value is counted, and the feature with the largest standard deviation of the average value is selected to represent the typical feature of the cluster. Compared with the existing multi-clustering method used for sample clustering and other feature dimension reduction methods, the method provided by the invention has the advantages that the clustering process is used for aggregating feature variables to form a plurality of feature clusters describing feature differences, and typical features of each feature cluster are established in the most favorable classification mode, so that the feature screening process which is favorable for completing classification tasks is realized while feature dimensions are greatly reduced, the problem of dimension explosion possibly caused by excessive feature dimensions is reduced, adverse effects of high feature dimensions on subsequent classification model design are alleviated, and generalization capability and robustness of classification tasks are improved.
Description of the drawings:
FIG. 1 is a schematic flow chart of the method of the invention.
Fig. 2 is a schematic flow chart of forming a plurality of feature clusters by using a neighbor propagation clustering method in step 2.
The specific embodiment is as follows:
examples:
a method for screening and dimension reduction of fault classification features of energy storage batteries under neighbor propagation clustering comprises the following steps:
step 1.1, acquiring N terminal voltage signal data samples of R failed energy storage batteries in the process of completing one-time charge and discharge by utilizing an experimental measurement mode;
step 1.2, defining and normalizing the feature vector of the voltage characteristics of the energy storage end represented by m feature variables according to a common feature extraction mode, namely the feature vector of the nth sample
Form feature set a= { (X) (n) ,L (n) )|n＝1,2,…,N；L (n) ∈{L 1 ,L 2 ,…,L R -expressed in matrix form as follows:
the row represents a sample, the first m represents a characteristic, and the m+1 represents a category of the sample;
step 2, evaluating the similarity degree among the features by utilizing cosine similarity, clustering the features by utilizing a neighbor propagation clustering method, and forming a plurality of feature clusters; the specific process is as follows:
step 2.1, vectors are composed with values of all samples under characteristic variables, i.e
And calculating cosine similarity sim between every two features, wherein the cosine similarity between the ith feature and the jth feature is as follows:
and forming a symmetrical square matrix taking the number of the features as the number of rows (columns) according to the cosine similarity among the features, namely a similarity matrix S, wherein the negative value of the cosine similarity among the element features in the square matrix is as follows:
step 2.2, let t=1, initialize the attraction degree matrix Re t And a home degree matrix Av t And the attraction degree matrix Re t And a home degree matrix Av t The damping coefficient zeta and the maximum iteration number T are defined as row and column square matrixes which are equal to the similarity matrix S;
step 2.3, calculating and updating the attraction degree matrix Re of the t+1 generation t+1 Elements of (a) and (b);
step 2.4, calculating and updating the attribution degree matrix Av of the t+1 generation t+1 Elements of (a) and (b);
step 2.5, judging whether the maximum iteration times are reached, namely T is less than or equal to T; if yes, t=t+1 returns to step 2.3, if not, step 2.6 is entered;
step 2.6 let e=re t+1 +Av t+1 Diagonalizing the matrix E, counting elements larger than zero on the diagonal, and counting the characteristics of row (column) numbers where the elements larger than zero are located, namely, a neighbor propagation clustering center, selecting columns where the clustering centers are located in the similarity matrix S, selecting the category attribution of other characteristics according to the rows, and forming a plurality of characteristic clusters; assuming that the diagonal elements ei and ej in the diagonalization matrix E are greater than zero, then feature x ei And x ej For the clustering center, the ei and ej columns in the similarity matrix S are selected, and according to the sizes of-sim (ek, ei) and-sim (ek, ej), if-sim (ek, ei) is more than or equal to-sim (ek, ej), the ek-th feature x ek Similar to feature x ei Belonging to the feature x ei Is a category of (2); if-sim (ek, ei)<Sim (ek, ej), ek th feature x ek Similar to feature x ej Belonging to the feature x ej Is a category of (2); so that two dry feature clusters can be formed for all the features, and the centers of the feature clusters are respectively formed by the features x ek And feature x ej Representative of;
step 3, screening the characteristics which are most favorable for fault diagnosis of the energy storage battery in the same characteristic cluster, and defining the characteristics as typical characteristics of the characteristic cluster; the method specifically comprises the following steps:
step 3.1, calculating the average value of the voltage characteristic values of the sample ends of different fault energy storage batteries under each similar characteristic in the same characteristic cluster according to the dividing result of the characteristic cluster obtained in the step 2; let H (k) = { x in the kth feature cluster ki ,x kj ,..}, feature x ki The average value of the characteristic values of the samples of different categories is as follows:
μ ki ＝[μ ki,1 ,μ ki,2 ,...,μ ki,R ]wherein the L < th > is r Class sample at feature x ki The lower mean value can be expressed as
Step 3.2, calculating standard deviation of voltage characteristic value mean values of different fault energy storage battery sample ends under each characteristic in the same characteristic cluster in the step 3.1, and maximizing the standard deviation in the characteristic cluster (different fault energy storage circuitsPool most significant under this feature) is defined as a typical feature characterizing this feature cluster; let the kth feature cluster H (k) contain two feature variables, i.e., H (k) = { x ki ,x kj Characteristic x ki And x kj Sample terminal voltage characteristic value mean { mu ] of different fault energy storage batteries ki ,μ kj Standard deviations of } are respectively
And->
Then selecting the feature with the largest standard deviation as representative feature x of the feature cluster H (k) ki (σ ki ≥σ kj )+x kj (σ ki ＜σ kj )；
Step 4, forming a screened dimension reduction feature set according to typical features of each feature cluster; according to the typical characteristics of each characteristic cluster obtained in the step 3.2, removing other characteristics of atypical characteristics in the step 1.2 to form a new characteristic set for describing difference of fault terminal voltage of R energy storage batteries, realizing characteristic screening and dimension reduction processes, and being used for designing a model of a subsequent learning machine to finish fault diagnosis tasks of the energy storage batteries.
Finally, it should be noted that the described embodiments are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
Claims (1)
1. The energy storage battery fault classification feature screening dimension reduction method under the neighbor propagation clustering is characterized by comprising the following steps of:
step 1, acquiring terminal voltage signal data samples of N energy storage batteries with different faults in the process of completing one-time charge and discharge, and excavating characteristics of terminal voltage signals to form a characteristic set; the specific process is as follows:
step 1.1, acquiring N terminal voltage signal data samples of R failed energy storage batteries in the process of completing one-time charge and discharge by utilizing an experimental measurement mode;
step 1.2, defining and normalizing the feature vector of the voltage characteristics of the energy storage end represented by m feature variables according to a common feature extraction mode, namely the feature vector of the nth sample
Form feature set a= { (X) (n) ,L (n) )|n＝1,2,…,N；L (n) ∈{L 1 ,L 2 ,…,L R -expressed in matrix form as follows:
the row represents a sample, the first m represents a characteristic, and the m+1 represents a category of the sample;
step 2, evaluating the similarity degree among the features by utilizing cosine similarity, clustering the features by utilizing a neighbor propagation clustering method, and forming a plurality of feature clusters; the specific process is as follows:
step 2.1, vectors are composed with values of all samples under characteristic variables, i.e
And calculating cosine similarity sim between every two features, wherein the cosine similarity between the ith feature and the jth feature is as follows:
and forming a symmetrical square matrix taking the number of the features as the number of rows and columns according to the cosine similarity among the features, namely a similarity matrix S, wherein the negative value of the cosine similarity among the element features in the square matrix is as follows:
step 2.2, let t=1, initialize the attraction degree matrix Re t And a home degree matrix Av t And the attraction degree matrix Re t And a home degree matrix Av t The damping coefficient zeta and the maximum iteration number T are defined as row and column square matrixes which are equal to the similarity matrix S;
step 2.3, calculating and updating the attraction degree matrix Re of the t+1 generation t+1 Elements of (a) and (b);
step 2.4, calculating and updating the attribution degree matrix Av of the t+1 generation t+1 Elements of (a) and (b);
step 2.5, judging whether the maximum iteration times are reached, namely T is less than or equal to T; if yes, t=t+1 returns to step 2.3, if not, step 2.6 is entered;
step 2.6 let e=re t+1 +Av t+1 Diagonalizing the matrix E, counting elements larger than zero on the diagonal, wherein the characteristics of the row numbers and the column numbers of the elements larger than zero are the neighbor propagation clustering centers, selecting the columns of the clustering centers in the similarity matrix S, selecting the category attribution of other characteristics according to the rows, and forming a plurality of characteristic clusters; assuming that the diagonal elements ei and ej in the diagonalization matrix E are greater than zero, then feature x ei And x ej For the clustering center, the ei and ej columns in the similarity matrix S are selected, and according to the sizes of-sim (ek, ei) and-sim (ek, ej), if-sim (ek, ei) is more than or equal to-sim (ek, ej), the ek-th feature x ek Similar to feature x ei Belonging to the feature x ei Is a category of (2); if-sim (ek, ei)<Sim (ek, ej), ek th feature x ek Similar to feature x ej Belonging to the feature x ej Is a category of (2); so that two dry feature clusters can be formed for all the features, and the centers of the feature clusters are respectively formed by the features x ek And feature x ej Representative of;
step 3, screening the characteristics which are most favorable for fault diagnosis of the energy storage battery in the same characteristic cluster, and defining the characteristics as typical characteristics of the characteristic cluster; the method specifically comprises the following steps:
step 3.1, calculating the average value of the voltage characteristic values of the sample ends of different fault energy storage batteries under each similar characteristic in the same characteristic cluster according to the dividing result of the characteristic cluster obtained in the step 2; let H (k) = { x in the kth feature cluster ki ,x kj ,..}, feature x ki The average value of the characteristic values of different types of samples is mu ki ＝[μ ki,1 ,μ ki,2 ,...,μ ki,R ]Wherein the L < th > is r Class sample at feature x ki The lower mean value can be expressed as
Calculating standard deviation of voltage characteristic value mean values of different fault energy storage battery sample ends under each characteristic in the same characteristic cluster in the step 3.1, and defining the characteristic with the largest standard deviation in the characteristic cluster as a typical characteristic for representing the characteristic cluster, wherein the difference of different fault energy storage batteries is most obvious under the characteristic; let the kth feature cluster H (k) contain two feature variables, i.e., H (k) = { x ki ,x kj Characteristic x ki And x kj Sample terminal voltage characteristic value mean { mu ] of different fault energy storage batteries ki ,μ kj Standard deviations of } are respectively
And->
Then selecting the feature with the largest standard deviation as representative feature x of the feature cluster H (k) ki (σ ki ≥σ kj )+x kj (σ ki ＜σ kj )；
Step 4, forming a screened dimension reduction feature set according to typical features of each feature cluster;
according to the typical characteristics of each characteristic cluster obtained in the step 3.2, removing other characteristics of atypical characteristics in the step 1.2 to form a new characteristic set for describing difference of fault terminal voltage of R energy storage batteries, realizing characteristic screening and dimension reduction processes, and being used for designing a model of a subsequent learning machine to finish fault diagnosis tasks of the energy storage batteries.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011094254.6A CN112307906B (en) | 2020-10-14 | 2020-10-14 | Energy storage battery fault classification feature screening dimension reduction method under neighbor propagation clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011094254.6A CN112307906B (en) | 2020-10-14 | 2020-10-14 | Energy storage battery fault classification feature screening dimension reduction method under neighbor propagation clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112307906A CN112307906A (en) | 2021-02-02 |
| CN112307906B true CN112307906B (en) | 2023-07-04 |
Family
ID=74488060
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011094254.6A Active CN112307906B (en) | 2020-10-14 | 2020-10-14 | Energy storage battery fault classification feature screening dimension reduction method under neighbor propagation clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112307906B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113552496B (en) * | 2021-06-29 | 2024-04-02 | 哈尔滨理工大学 | Battery series module internal short circuit fault diagnosis method based on voltage cosine similarity |
| CN114088388B (en) * | 2021-12-10 | 2024-09-06 | 深圳市出新知识产权管理有限公司 | Fault diagnosis method and fault diagnosis device for gear box |
| CN114497770B (en) * | 2022-01-26 | 2022-09-09 | 上海玫克生储能科技有限公司 | Method, system and terminal for analyzing state of battery box in battery cluster |
| CN114818881B (en) * | 2022-04-07 | 2024-04-26 | 青岛大学 | Fault detection and positioning method for voltage sensor of vehicle-mounted power battery pack |
| CN117457094B (en) * | 2023-12-20 | 2024-03-29 | 安徽农业大学 | Oxyhydrogen fuel cell energy efficiency performance evaluation method and system based on AP algorithm |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010064414A1 (en) * | 2008-12-02 | 2010-06-10 | ソニー株式会社 | Gene clustering program, gene clustering method, and gene cluster analyzing device |
| JP2016197406A (en) * | 2015-04-06 | 2016-11-24 | 国立研究開発法人産業技術総合研究所 | Information processor, information processing system, information processing method, program, and recording medium |
| CN109245100A (en) * | 2018-11-07 | 2019-01-18 | 国网浙江省电力有限公司经济技术研究院 | Consider the Dynamic Load Modeling method of alternating current-direct current distribution network load composition time variation |
| CN109976308A (en) * | 2019-03-29 | 2019-07-05 | 南昌航空大学 | A kind of extracting method of the fault signature based on Laplce's score value and AP cluster |
| CN110084520A (en) * | 2019-04-30 | 2019-08-02 | 国网上海市电力公司 | Charging station site selecting method and device based on public bus network Yu gridding AP algorithm |
| WO2020066257A1 (en) * | 2018-09-26 | 2020-04-02 | 国立研究開発法人理化学研究所 | Classification device, classification method, program, and information recording medium |
| CN111506730A (en) * | 2020-04-17 | 2020-08-07 | 腾讯科技（深圳）有限公司 | Data clustering method and related device |
-
2020
- 2020-10-14 CN CN202011094254.6A patent/CN112307906B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010064414A1 (en) * | 2008-12-02 | 2010-06-10 | ソニー株式会社 | Gene clustering program, gene clustering method, and gene cluster analyzing device |
| JP2016197406A (en) * | 2015-04-06 | 2016-11-24 | 国立研究開発法人産業技術総合研究所 | Information processor, information processing system, information processing method, program, and recording medium |
| WO2020066257A1 (en) * | 2018-09-26 | 2020-04-02 | 国立研究開発法人理化学研究所 | Classification device, classification method, program, and information recording medium |
| CN109245100A (en) * | 2018-11-07 | 2019-01-18 | 国网浙江省电力有限公司经济技术研究院 | Consider the Dynamic Load Modeling method of alternating current-direct current distribution network load composition time variation |
| CN109976308A (en) * | 2019-03-29 | 2019-07-05 | 南昌航空大学 | A kind of extracting method of the fault signature based on Laplce's score value and AP cluster |
| CN110084520A (en) * | 2019-04-30 | 2019-08-02 | 国网上海市电力公司 | Charging station site selecting method and device based on public bus network Yu gridding AP algorithm |
| CN111506730A (en) * | 2020-04-17 | 2020-08-07 | 腾讯科技（深圳）有限公司 | Data clustering method and related device |
Non-Patent Citations (2)
| Title |
|---|
| Battery Grouping with Time Series Clustering Based on Affinity Propagation;Zhiwei He等;《Energies 》;第9卷(第7期);第561页 * |
| 基于近邻传播算法的电池配组技术研究;唐丽君;《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》(第2期);第C042-1570页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112307906A (en) | 2021-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112307906B (en) | Energy storage battery fault classification feature screening dimension reduction method under neighbor propagation clustering | |
| CN112379269B (en) | Battery abnormality detection model training and detection method and device thereof | |
| CN110752410B (en) | Method for rapidly sorting and recombining retired lithium battery | |
| CN109635928A (en) | A kind of voltage sag reason recognition methods based on deep learning Model Fusion | |
| CN109190672A (en) | Operation of Electric Systems operating condition unsupervised clustering and device | |
| CN111008726B (en) | Class picture conversion method in power load prediction | |
| CN110942091A (en) | Semi-supervised few-sample image classification method for searching reliable abnormal data center | |
| CN111259917B (en) | Image feature extraction method based on local neighbor component analysis | |
| CN117201122B (en) | Unsupervised attribute network anomaly detection method and system based on view level graph comparison learning | |
| CN110726898A (en) | Power distribution network fault type identification method | |
| CN113033678A (en) | Lithium battery pack fault diagnosis method based on adaptive countermeasure network | |
| CN115277888B (en) | Method and system for analyzing message type of mobile application encryption protocol | |
| CN112116950B (en) | Protein folding identification method based on depth measurement learning | |
| CN114609994A (en) | Fault diagnosis method and device based on multi-granularity regularization rebalance incremental learning | |
| CN114936599A (en) | Base station energy consumption abnormity monitoring method and system based on wavelet decomposition and migration discrimination | |
| CN117556369A (en) | Power theft detection method and system for dynamically generated residual error graph convolution neural network | |
| CN116805051A (en) | Double convolution dynamic domain adaptive equipment fault diagnosis method based on attention mechanism | |
| CN113780402A (en) | User electricity stealing detection method based on improved generation countermeasure network | |
| CN113406503A (en) | Lithium battery SOH online estimation method based on deep neural network | |
| CN115659254A (en) | Power quality disturbance analysis method for power distribution network with bimodal feature fusion | |
| CN114841266A (en) | Voltage sag identification method based on triple prototype network under small sample | |
| CN115115113A (en) | Equipment fault prediction method and system based on graph attention network relation embedding | |
| CN111783688B (en) | Remote sensing image scene classification method based on convolutional neural network | |
| CN114596445B (en) | Method for improving fault diagnosis precision of oil extraction machine | |
| CN112733067A (en) | Data set selection method for robot target detection algorithm |
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 |