├── LRNN- MOTOR FAULT DETECTION.pdf
├── README.md
└── code.m


/LRNN- MOTOR FAULT DETECTION.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/ssv112/Motor-Fault-Diagnosis-Using-LRNN/HEAD/LRNN- MOTOR FAULT DETECTION.pdf


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # 3-Phase-Induction-Motor-Fault-Diagnosis-Using-Layer Recurrent Neural Networks
2 | Layer Recurrent Neural Network Based External Fault Diagnosis Model for Three Phase Induction Motor Using Current Signature Analysis
3 | 
4 | 
5 | Three Phase Induction Motors (IMs) are the most commonly used motors in industries so during regular use faults may occur in them. Fault Diagnosis and Conditioning monitoring (FDCA) has now become an essential part in the field of Electrical Engineering for the safe operations of motors as it can increase life and efficiency of an IM. In this paper, FDCA of induction motors is done on the basis of available dataset. Induction motors external faults are online controlled by LRNN i.e. Layered Recurrent Neural Network & then the results are compared with MLP neural network, which states that the fault identification by LRNN is better& less time consuming than MLP. For this proposed algorithm, RMS voltage and RMS current have been used as input variables for the diagnosis of normal and faulty conditions in IM.
6 | 


--------------------------------------------------------------------------------
/code.m:
--------------------------------------------------------------------------------
 1 | clc; clear all
 2 | tic
 3 | load('TrainData.mat')
 4 | y=training';
 5 | t=trainingS1';
 6 | 
 7 | %yy=testing';
 8 | pp = con2seq((testing)');
 9 | tt = con2seq((testingS1)');
10 | 
11 | p = con2seq(y);
12 | t = con2seq(t);
13 | lrn_net = layrecnet(1,3);
14 | lrn_net.trainFcn = 'traincgb';
15 | lrn_net.trainParam.show = 5;
16 | lrn_net.trainParam.epochs = 100;
17 | lrn_net = train(lrn_net,p,t);
18 | %after training the model
19 | y = lrn_net(p);
20 | %Train output plot
21 | figure(1)
22 | plot(cell2mat(y))
23 | hold on
24 | plot(cell2mat(t))
25 | hold off
26 | 
27 | %% Test the Model
28 | Y_test = sim(lrn_net,pp);
29 | YY_test=cell2mat(Y_test);
30 | YY_testT=cell2mat(tt);
31 | % plot(cell2mat(Y_test))
32 | 
33 | % Testing output plot
34 | figure(2)
35 | plot(cell2mat(Y_test))
36 | hold on
37 | plot(cell2mat(tt))
38 | hold off
39 | 
40 | figure(3)
41 | plotregression(YY_testT,YY_test,'Regresion Test')
42 | 
43 | YY_test=cell2mat(Y_test);
44 | YY_testT=cell2mat(tt);
45 | error_test=YY_testT-YY_test;
46 | %MSE_test=mse(error_test)
47 | %plotregression()
48 | Toc
49 | 


--------------------------------------------------------------------------------