CN113419172A - New energy automobile three-phase asynchronous motor fault identification based on GCN and VMD-ED - Google Patents
New energy automobile three-phase asynchronous motor fault identification based on GCN and VMD-ED Download PDFInfo
- Publication number
- CN113419172A CN113419172A CN202110688351.6A CN202110688351A CN113419172A CN 113419172 A CN113419172 A CN 113419172A CN 202110688351 A CN202110688351 A CN 202110688351A CN 113419172 A CN113419172 A CN 113419172A
- Authority
- CN
- China
- Prior art keywords
- gcn
- asynchronous motor
- phase asynchronous
- new energy
- vmd
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 claims abstract description 23
- 238000003745 diagnosis Methods 0.000 claims abstract description 8
- 239000013598 vector Substances 0.000 claims description 13
- 238000013528 artificial neural network Methods 0.000 claims description 10
- 238000000354 decomposition reaction Methods 0.000 claims description 7
- 238000011161 development Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims 1
- 238000004088 simulation Methods 0.000 abstract 1
- 239000011159 matrix material Substances 0.000 description 8
- 239000003208 petroleum Substances 0.000 description 6
- 239000000306 component Substances 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 239000002551 biofuel Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000003502 gasoline Substances 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000000474 nursing effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Images
Classifications
-
- 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
- G01R31/34—Testing dynamo-electric machines
- G01R31/343—Testing dynamo-electric machines in operation
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Control Of Electric Motors In General (AREA)
Abstract
The method for identifying the fault part of the three-phase asynchronous motor of the new energy automobile based on the GCN and the VMD-ED is provided. Firstly, VMD, ED and GCN algorithms are simply introduced, then GCN is combined with VMD-ED, and a method based on GCN and VMD-ED is provided. The three-phase asynchronous motor of the new energy automobile is used as an application object to carry out simulation experiments, and experimental results show that the algorithm is high in diagnosis speed and minimum in loss rate.
Description
Technical Field
The invention relates to a method for identifying faults of a three-phase asynchronous motor of a new energy automobile, and particularly provides a fault diagnosis algorithm based on GCN and VMD-ED (general packet network-based fault diagnosis) aiming at the characteristics that fault signals of a large power converter are unstable and effective information is easily covered by noise.
Background art:
new energy automobile three-phase asynchronous machine
Under the pressure of energy and environmental protection, new energy automobiles will undoubtedly become the development direction of future automobiles. If the new energy automobile is rapidly developed, calculated by 1.4 hundred million of the automobile reserves in China in 2020, 3229 million tons of oil can be saved, 3110 million of the oil can be replaced, and 6339 million of the oil can be saved and replaced, which is equivalent to the reduction of the oil requirement of the automobile by 22.7%. The saving and the replacement of petroleum before 2020 are mainly realized by developing advanced diesel vehicles, hybrid vehicles and the like. By 2030, the development of new energy automobiles can save 7306 million tons of petroleum, 9100 million tons of alternative petroleum, and 16406 million tons of petroleum saving and alternative petroleum, which is equivalent to reducing the petroleum demand of automobiles by 41%. At that time, the biological fuel and the fuel cell play important roles in automobile oil replacement. By combining the energy resource status of China and the development trend of international automobile technology, it is predicted that after 2025 years, China's common gasoline vehicles will only account for about 50% of the passenger cars, while advanced diesel vehicles, gas vehicles, biofuel vehicles and other new energy vehicles will develop rapidly. The motor serves as a core component in the three-phase power supply of the new energy automobile. The three-phase asynchronous motor has the advantages of reliable operation, high rotating speed, low cost, convenient maintenance, higher efficiency and the like, and plays an irreplaceable role in various aspects of industrial production, agricultural production, transportation, national defense, commerce, household appliances, medical electrical appliances and the like. As an indispensable part of a power system, the normal operation of the three-phase asynchronous motor has very important practical significance. Therefore, the method has certain value in diagnosing the faults of the three-phase asynchronous motor. Through various checking and testing methods, whether a fault exists in the equipment is found, and the position of the fault or the type of the fault is further determined to be called fault diagnosis, and the fault diagnosis can also be regarded as a fault identification problem.
Disclosure of Invention
The invention aims to solve the problems of multiple faults, low diagnosis speed and the like of a three-phase asynchronous motor, and provides a three-phase asynchronous motor fault part identification method based on VMD decomposition and graph convolution neural network. The method comprises the steps of firstly collecting normal operation current signals and several fault current signals of the three-phase asynchronous motor, nursing the signals, then decomposing and preprocessing the signals by using a variable empirical mode, constructing a graph structure, then realizing fault diagnosis of the three-phase asynchronous motor by using a graph convolution neural network, and effectively separating normal three-phase asynchronous motor operation data from fault data.
In order to achieve the purpose, the invention adopts the following technical scheme:
a three-phase asynchronous motor fault part identification method based on GCN and VMD-ED. The method comprises a data preprocessing module, variable empirical mode decomposition and a graph convolution neural network. The data preprocessing is to normalize the acquired original current signal data and eliminate the influence of magnitude order. The variable empirical mode decomposition is used for extracting data features, and the main function of the graph convolution neural network is to identify different final faults.
The invention mainly utilizes the following technologies:
1. empirical mode decomposition
The Variable Mode Decomposition (VMD) decomposes an input signal f (t) into a plurality of sub-signals u (t) with specific bandwidthsk(t) these sub-signals are concentrated at their respective center frequencies WkNearby, i.e. eigenmode components. The construction of the constraint variational model comprises the following three steps: (1) performing corresponding Hilbert transform for each mode quantity; (2) tuning the index of each modal quantity to an estimated center frequency such that the modal spectrum is shifted onto the baseband; (3) and performing two-norm operation on the gradient of the demodulation signal, thereby estimating the bandwidth of each modal quantity frequency band. The resulting constrained variational model is as follows:
in the formula { uk}={u1,u2,...,uK},{ωk}={ω1,ω2,...,ωKRepresents all the modal component sets and their respective center frequency sets, respectively, "+" represents the convolution operation; δ (t) is a unit pulse function;representing a partial derivative operation; j is an imaginary unit; f (t) is the objective function.
In order to solve the constraint variation model, a quadratic penalty term alpha and a Lagrange multiplier lambda are introduced for calculation to obtain an augmented Lagrange function:
where < > represents the inner product calculation.
The VMD algorithm flow is as follows:
iteratively updating u by computing the saddle points of the equations by an alternating direction multiplier algorithmkAnd ωkAnd finally obtaining the optimal solution of the constraint variable model as shown in the following formula.
Updating lambda
If the precision epsilon meets the requirement, the iteration is ended
2. European distance
Euclidean Distance (ED) is a measure of distance, a commonly used distance definition, which is the true distance between two points in an m-dimensional space, and the Euclidean distance between two vectors is calculated as follows:
3. graph convolution neural network
The GCN is a neural network layer, and assuming that a group of data has N nodes, each node has its own characteristics, and the characteristics of the nodes form an N × D matrix X, and then the relationship between the nodes also forms an N × N matrix a, also called an adjacency matrix. X and A are the inputs to our model. The propagation method between layers of the GCN is as follows:
Drawings
FIG. 1 is a flow chart of a wind driven generator fault site identification method based on GCN and WPT-ED.
The present invention will be described in detail below with reference to the accompanying drawings and using various techniques.
The implementation process of the whole wind driven generator fault part identification method is described in fig. 1, and the process is as follows.
The first step is data acquisition. The experimental system of the three-phase asynchronous motor is supplied with power by a symmetrical three-phase power supply, a voltage regulator and a frequency converter are respectively used for regulating the voltage and the frequency of the motor, and an oscilloscope capable of generating data is connected through a current clamp in the experimental process to sample the current of a two-phase stator. And accessing the computer after the acquisition is finished.
And a second step of feature extraction. In order to avoid a frequency mixing phenomenon in the feature extraction process, the VMD algorithm is used for decomposing current data acquired during fault to obtain a plurality of IMF components.
The third step is to build the graph structure. The similarity between feature vectors may represent the relationship between them. Since the euclidean distance can take into account the connection between various characteristics, we select the euclidean distance to calculate the similarity of each unknown sample set. Different distances are obtained by calculating the similarity among the characteristic vectors, the smaller the distance is, the greater the similarity is, and conversely, the greater the distance is, the smaller the similarity is. Through multiple experiments, when the threshold value of the Euclidean distance is 0.2, the relation between most feature vectors can exist as far as possible, namely: when the distance between any two feature vectors is less than 0.2, the two feature vectors may be defined as having a certain relationship. And constructing a topological graph structure by using the relation between the feature vectors. And taking each feature vector as a node, and constructing a topological structure for the edges according to the relation among the feature vectors. And (3) calculating the center distance between fault types in the training sample by using an Euclidean distance formula, screening the fault types which are easy to distinguish, and finally training and testing. And fourthly, recognizing the fault part of the three-phase asynchronous motor. And inputting the graph structure constructed in the third step into a graph convolution neural network. The GCN inputs two matrices, N × N adjacency matrix and N × D feature matrix. And N is the number of nodes. Wherein the steering matrix N x N is formed by the relationships between the nodes.
Claims (4)
1. A three-phase asynchronous motor fault part identification method based on GCN and WPT-ED is characterized in that a graph convolution neural network (GCN), variable empirical mode decomposition (VMD) and Mahalanobis distance (ED) are fused, and a new energy automobile three-phase asynchronous motor is used as an application object to identify a fault part;
the VMD is used for extracting different features by performing signal decomposition on each level of signals to construct feature vectors;
the ED is used for calculating the distance between different feature vectors, and the similarity between different feature vectors is obtained by setting a threshold value, so that the method is an effective method for calculating the similarity of a sample set;
the GCN is used for processing data such as high-dimensional and nonlinear big data, the collected and preprocessed data are input into the GCN, and fault classification is realized by using the GCN, so that a fault part identification method is realized.
2. The method for identifying the fault location of the three-phase asynchronous motor of the new energy vehicle based on the GCN and the VMD-ED as claimed in claim 1, wherein the fault location of the three-phase asynchronous motor of the new energy vehicle can be identified rapidly and effectively.
3. The GCN and WPT-VMD based new energy automobile three-phase asynchronous motor fault location identification method according to claim 2 can quickly and effectively identify the fault location of the new energy automobile three-phase asynchronous motor. The method comprises three steps: wavelet packet transformation realizes signal feature extraction, similarity between Mahalanobis distance calculation feature vectors, and graph convolution neural network realizes fault position identification of the new energy automobile three-phase asynchronous motor.
4. The GCN and VMD-ED based new energy vehicle three-phase asynchronous motor fault location identification method according to claim 1, characterized in that the graph convolution neural network is a novel intelligent algorithm, and the algorithm has fast diagnosis speed, high accuracy and good development prospect.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110688351.6A CN113419172A (en) | 2021-06-21 | 2021-06-21 | New energy automobile three-phase asynchronous motor fault identification based on GCN and VMD-ED |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110688351.6A CN113419172A (en) | 2021-06-21 | 2021-06-21 | New energy automobile three-phase asynchronous motor fault identification based on GCN and VMD-ED |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113419172A true CN113419172A (en) | 2021-09-21 |
Family
ID=77789642
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110688351.6A Pending CN113419172A (en) | 2021-06-21 | 2021-06-21 | New energy automobile three-phase asynchronous motor fault identification based on GCN and VMD-ED |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113419172A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114966403A (en) * | 2022-08-01 | 2022-08-30 | 山东博源精密机械有限公司 | New energy automobile motor locked-rotor fault detection method and system |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111812507A (en) * | 2020-05-27 | 2020-10-23 | 浙江工业大学 | Motor fault diagnosis method based on graph convolution |
CN112270273A (en) * | 2020-10-30 | 2021-01-26 | 湘潭大学 | Wind driven generator fault part identification method based on GCN and WPT-MD |
-
2021
- 2021-06-21 CN CN202110688351.6A patent/CN113419172A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111812507A (en) * | 2020-05-27 | 2020-10-23 | 浙江工业大学 | Motor fault diagnosis method based on graph convolution |
CN112270273A (en) * | 2020-10-30 | 2021-01-26 | 湘潭大学 | Wind driven generator fault part identification method based on GCN and WPT-MD |
Non-Patent Citations (1)
Title |
---|
时献江等: "VMD方法在轴承故障定子电流信号诊断中的应用", 《哈尔滨理工大学学报》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114966403A (en) * | 2022-08-01 | 2022-08-30 | 山东博源精密机械有限公司 | New energy automobile motor locked-rotor fault detection method and system |
CN114966403B (en) * | 2022-08-01 | 2022-10-25 | 山东博源精密机械有限公司 | New energy automobile motor locked-rotor fault detection method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109635928B (en) | Voltage sag reason identification method based on deep learning model fusion | |
CN103136587A (en) | Power distribution network operating state classification recognition method based on support vector machine | |
CN110687450B (en) | Lithium battery residual life prediction method based on phase space reconstruction and particle filtering | |
CN111628494B (en) | Low-voltage distribution network topology identification method and system based on logistic regression method | |
CN112069727B (en) | Intelligent transient stability evaluation system and method with high reliability for power system | |
CN111103477B (en) | Electrical load monitoring method and system based on 3D convolutional neural network | |
Hu et al. | Hierarchical fault diagnosis for power systems based on equivalent-input-disturbance approach | |
CN112633098A (en) | Fault diagnosis method and system for rotary machine and storage medium | |
CN107944705A (en) | A kind of all terminal reliability computational methods that communication corporations are divided based on modularity | |
CN110569888A (en) | transformer fault diagnosis method and device based on directed acyclic graph support vector machine | |
CN115618249A (en) | Low-voltage power distribution station area phase identification method based on LargeVis dimension reduction and DBSCAN clustering | |
Hu et al. | Design of fault diagnosis algorithm for electric fan based on LSSVM and Kd-Tree | |
CN113419172A (en) | New energy automobile three-phase asynchronous motor fault identification based on GCN and VMD-ED | |
CN113627685B (en) | Wind driven generator power prediction method considering wind power internet load limit | |
Gao et al. | Design and application of a fault diagnosis and monitoring system for electric vehicle charging equipment based on improved deep belief network | |
CN117171544B (en) | Motor vibration fault diagnosis method based on multichannel fusion convolutional neural network | |
CN117609818A (en) | Power grid association relation discovery method based on clustering and information entropy | |
Chen et al. | Meta rule-based energy management strategy for battery/supercapacitor hybrid electric vehicles | |
Hassani et al. | Design of a cost-effective deep convolutional neural network–based scheme for diagnosing faults in smart grids | |
CN112215490A (en) | Power load cluster analysis method based on correlation coefficient improved K-means | |
Wang et al. | An Automatic Identification Framework for Complex Power Quality Disturbances Based on Ensemble CNN | |
Nithiyananthan et al. | Enhanced R package-based cluster analysis fault identification models for three phase power system network | |
CN111126449A (en) | Battery fault classification diagnosis method based on cluster analysis | |
Wang et al. | Cluster division in wind farm through ensemble modelling | |
CN111988192A (en) | Train communication network fault diagnosis 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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210921 |
|
WD01 | Invention patent application deemed withdrawn after publication |