Hydraulic System Fault Diagnosis Based on EMD and Improved PSO-Elman ANN

2013 ◽  
Vol 284-287 ◽  
pp. 2936-2940
Author(s):  
Zeng Shou Dong ◽  
Xiao Yu Zhang ◽  
Jian Chao Zeng
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Convergence Rate ◽  
Hydraulic System ◽  
Learning Algorithm ◽  
Threshold Value ◽  
Pso Algorithm ◽  
Elman Neural Network ◽  
Neural Network Learning ◽  
Network Convergence

The element parameters of engineering machinery hydraulic system are detected, the fault eigenvector is extracted, and the information is applied to neural network fault diagnosis. Experience mode decomposition (EMD) is used to extract fault characteristic vectors in this paper, combined with the pressure, temperature and flow rate of dominant signal as neural network's inputs. In addition, the paper improves the Elman neural network learning algorithm by PSO algorithm. It can effectively increase network convergence rate and computing power. The particle swarm is used to optimize Elman neural network weights and the threshold value and then applied in the fault diagnosis system by training the network. The results show that the method increases the neural network convergence rate and reduces diagnoses error.

HYDRAULIC SYSTEM FAULT DIAGNOSIS BASED ON EMD AND IMPROVED PSO-ELMAN ANN

2013 ◽  
Vol 37 (4) ◽  
pp. 1189-1197 ◽  
Author(s):  
Zengshou Dong ◽  
Xiaoyu Zhang ◽  
Jianchao Zeng
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Convergence Rate ◽  
Hydraulic System ◽  
Learning Algorithm ◽  
Threshold Value ◽  
Pso Algorithm ◽  
Elman Neural Network ◽  
Neural Network Learning ◽  
Network Convergence

The element parameters of engineering machinery hydraulic system are detected, the fault eigenvector is extracted, and the information is applied to neural network fault diagnosis. Experience mode decomposition (EMD) is used to extract fault characteristic vectors in this paper, combined with the pressure, temperature and flow rate of dominant signal as neural network’s inputs. In addition, the paper improves the Elman neural network learning algorithm by the PSO algorithm. It can effectively increase network convergence rate and computing power. The particle swarm is used to optimize Elman neural network weights and the threshold value and then applied in the fault diagnosis system by training the network. The results show that the method increases the neural network convergence rate and reduces diagnoses error.

WNN Model Based on Particle Swarm Optimization for Fault Diagnosis in Analog Circuit

2013 ◽  
Vol 427-429 ◽  
pp. 1048-1051
Author(s):  
Xu Sheng Gan ◽  
Hao Lin Cui ◽  
Ya Rong Wu
Keyword(s):  
Neural Network ◽  
Particle Swarm Optimization ◽  
Fault Diagnosis ◽  
Convergence Rate ◽  
Analog Circuit ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Wavelet Neural Network ◽  
Swarm Optimization ◽  
Fast Convergence Rate

In order to diagnose the fault in analog circuit correctly, a Wavelet Neural Network (WNN) method is proposed that uses the Particle Swarm Optimization (PSO) algorithm to optimize the network parameters. For the improvement of convergence rate in WNN based on PSO algorithm, a compressing method in research space is introduced into the traditional PSO algorithm to improve the convergence in WNN training. The simulation shows that the proposed method has a good diagnosis with fast convergence rate for the fault in analog circuit.

An adaptive Elman neural network with C-PSO learning algorithm based pitch angle controller for DFIG based WECS

2016 ◽  
Vol 23 (5) ◽  
pp. 716-730 ◽  
Author(s):  
R Sitharthan ◽  
M Geethanjali
Keyword(s):  
Neural Network ◽  
Output Power ◽  
Pitch Angle ◽  
Learning Algorithm ◽  
Back Propagation ◽  
Pso Algorithm ◽  
Power Fluctuation ◽  
Elman Neural Network ◽  
Learning Capability ◽  
Wind Energy Conversion

Frequent variation in the wind flow affects the Wind Turbine (WT) to generate fluctuating output power and this can negatively impact the entire power network. This paper aims at modelling an Enhanced-Elman Neural Network (EENN) based pitch angle controller to mitigate the output power fluctuation in a grid connected Wind Energy Conversion System. The outstanding aspect of the proposed controller is that, they can smoothen the output power fluctuation, when the wind speed is above or below rated speed of the WT. The proposed EENN pitch controller is trained online using Gradient Descent (GD) algorithm and the network learning is carried out using Customized-Particle swarm optimization (C-PSO) algorithm. The C-PSO is adopted, in order to increase the learning capability of the training process by adjusting the networks learning rate. Further, the node connecting weights of the EENN is updated by means of GD algorithm using back-propagation methodology. The performance of the proposed controller is analysed using the simulation studies carried out in MATLAB /Simulink environment.

A Modified PSO-BP Algorithm in Hydraulic System Fault Diagnosis Application

2012 ◽  
Vol 182-183 ◽  
pp. 1145-1148 ◽  
Author(s):  
Zeng Shou Dong ◽  
Xiao Yu Zhang ◽  
Jian Chao Zeng
Keyword(s):  
Fault Diagnosis ◽  
Convergence Rate ◽  
Hydraulic System ◽  
Particle Swarm ◽  
Bp Algorithm ◽  
Particle Swarm Algorithm ◽  
Bp Network ◽  
Learning Factor ◽  
Swarm Algorithm ◽  
Modified Pso

BP neural network for failure pattern recognition has been used in hydraulic system fault diagnosis.However, its convergence rate is relatively small and always trapped at the local minima. So a new modified PSO-BP hydraulic system fault diagnosis method was proposed,which combined the respective advantages of particle swarm algorithm and BP algorithm. Firstly, the inertia weight and learning factor of the standard particle swarm algorithm was improved, then BP neural network’s weights and thresholds were optimized by modified PSO algorithm. BP network performance was ameliorated. The simulation results showed that this method improved the convergence rate of the BP network, and it could reduce the diagnostic errors.

Adaptive Learning Algorithm for Cerebellar Model Articulation Controller: Neural Network Based Hybrid-Type Controller—Part I

2000 ◽  
Author(s):  
Magdy Mohamed Abdelhameed ◽  
Sabri Cetinkunt
Keyword(s):  
Neural Network ◽  
Control System ◽  
Adaptive Learning ◽  
Learning Algorithm ◽  
Cerebellar Model Articulation Controller ◽  
Neural Network Learning ◽  
Network Learning ◽  
Hybrid Type ◽  
Cmac Neural Network ◽  
Learning Technique

Abstract Cerebellar model articulation controller (CMAC) is a useful neural network learning technique. It was developed two decades ago but yet lacks an adequate learning algorithm, especially when it is used in a hybrid- type controller. This work is intended to introduce a simulation study for examining the performance of a hybrid-type control system based on the conventional learning algorithm of CMAC neural network. This study showed that the control system is unstable. Then a new adaptive learning algorithm of a CMAC based hybrid- type controller is proposed. The main features of the proposed learning algorithm, as well as the effects of the newly introduced parameters of this algorithm have been studied extensively via simulation case studies. The simulation results showed that the proposed learning algorithm is a robust in stabilizing the control system. Also, this proposed learning algorithm preserved all the known advantages of the CMAC neural network. Part II of this work is dedicated to validate the effectiveness of the proposed CMAC learning algorithm experimentally.

Sign in / Sign up

Export Citation Format

Share Document