Fault Diagnosis Method Based on KPCA and Selective Neural Network Ensemble

2014 ◽  
Vol 915-916 ◽  
pp. 1272-1276 ◽  
Author(s):  
Yan Yan Pang ◽  
Hai Ping Zhu ◽  
Fan Mao Liu
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Binary Particle Swarm Optimization ◽  
Large Network ◽  
Neural Network Ensemble ◽  
Tennessee Eastman Process ◽  
Training Time ◽  
Diagnosis Method ◽  
Neural Network Algorithm ◽  
Bagging Method

Aiming at the problems of less study sample, large network scale and long training time existing in current fault diagnosis field, we develop a new method based on KPCA and selective neural network ensemble. First, reducing the data size by using KPCA to extract the sample features. Then achieving a selective neural network ensemble method based on improved binary particle swarm optimization algorithm (IBPSOSEN), and combining the two methods for fault diagnosis. In selective neural network algorithm, bagging method is used to take a number of different training sets of fault samples to solve the problem of less fault samples. Finally, simulation experiments and comparisons over Tennessee Eastman Process (TE) demonstrate the effectiveness and feasibility of the proposed method.

A two-stage classifier switchable aluminum electrolysis fault diagnosis method

2021 ◽  
pp. 014233122110596
Author(s):  
Tianhao Gao ◽  
Ke Zhang ◽  
Huaitao Shi ◽  
Jinbao Zhao ◽  
Jiejia Li
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Simple Structure ◽  
Aluminum Electrolysis ◽  
Electrolytic Cell ◽  
Training Method ◽  
Random Forest Algorithm ◽  
Electrolytic Cells ◽  
Diagnosis Method ◽  
Neural Network Algorithm

Traditional aluminum electrolysis fault diagnosis methods have problems such as low accuracy, small forecast advance, and high CPU usage, which make their popularity low in enterprises. Aiming at the above problems, a fault diagnosis method with switchable two-level classifiers is designed. The input data are first judged by the first-level algorithm. If it is determined that there is no fault, the result will be output directly. If it is determined that there is a fault in the electrolytic cell, the data will be transferred to the second-level network for specific fault diagnosis. The first level is based on the Random Forest algorithm with simple structure and good two-class classification effect and is optimized by the improved cuckoo algorithm. The second level is based on an improved DBN-DNN (Deep Belief Neural Network–Deep Neural Network) algorithm, and the training method is given. Experimental results show that this method can switch between different algorithms according to different situations, save computing resources, realize that a computer can monitor multiple electrolytic cells, and reduce investment costs. In addition, the accuracy and forecast advance have been significantly improved, which has promoted the popularization of fault diagnosis systems in aluminum electrolysis enterprises.

The Fault Diagnosis Method of Diesel Based on Wavelet Pocket and BP Neural Network

2014 ◽  
Vol 635-637 ◽  
pp. 910-913 ◽  
Author(s):  
Hong Hui Sun ◽  
Jun Xu ◽  
Qing Hua Zhang ◽  
Hong Xia Wang
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Adaptive Learning ◽  
Bp Neural Network ◽  
Wavelet Packet ◽  
Vibration Signals ◽  
Signal Energy ◽  
Time Frequency ◽  
Diagnosis Method ◽  
Neural Network Algorithm

Because of the well time-frequency spectrum disposal capability of wavelet packet, the wavelet packet algorithm is used to analyze the time - frequency characteristics of diesel vibration signals. The signal energy distributing characteristics based on wavelet packet transform. are extracted and taken as diagnostic characteristic vector, then improved BP neural network algorithm that connects additional momentum with self-adaptive learning rate was used to classify and recognize faults of diesel valves. The experimental results show the fault diagnosis method of diesel based on wavelet pocket and BP neural network is effective and feasible.

Study of the Fault Diagnosis Method Based on Wavelet Time and Frequency Analysis

2012 ◽  
Vol 472-475 ◽  
pp. 2166-2170
Author(s):  
Qun Qi ◽  
Xue Zhang Zhao
Keyword(s):  
Neural Network ◽  
Wavelet Transform ◽  
Fault Diagnosis ◽  
Asynchronous Motor ◽  
Wavelet Neural Network ◽  
Experiment Data ◽  
Complex Fault ◽  
Diagnosis Method ◽  
Neural Network Algorithm ◽  
Network Fault

In order to better solve asynchronous motor complex fault characteristics, improve the reliability of the diagnosis and accuracy, combined with wavelet transform technique, construct a wavelet neural network, wavelet transform technology feature extraction asynchronous motor as a signal wavelet neural network's input vector, and the wavelet neural network algorithm was used to optimize, realize the motor identify types of fault, through the simulation experiment data diagnosis results show that this method is effective and feasible. Based on the wavelet analysis and neural network fault diagnosis method of research.

Research of CNC Fault Diagnosis Based on RBF Neural Network

2012 ◽  
Vol 150 ◽  
pp. 211-216 ◽  
Author(s):  
Jian Sheng Zhang ◽  
Yan Hong Zhang
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Fault Diagnosis ◽  
Rbf Neural Network ◽  
Numerical Control ◽  
Time Data ◽  
Artificial Intelligence Technology ◽  
Diagnosis Method ◽  
Neural Network Algorithm ◽  
Control Machine

The artificial intelligence technology has been widely used in fault diagnosis, but under the unusual circumstances, if the real time data is lacking, it is difficult to build new mathematical model. Now the artificial intelligence technology has been used in the fault diagnosis field. The CNC hinders diagnosis using neural network is introduced in this paper, and the fault diagnosis method of numerical control machine is analyzed, and hence the CNC fault diagnoses are realized by RBF neural network algorithm and program.

scholarly journals Sensor and Actuator Fault Diagnosis for Robot Joint Based on Deep CNN

Entropy ◽  
2021 ◽  
Vol 23 (6) ◽  
pp. 751
Author(s):  
Jinghui Pan ◽  
Lili Qu ◽  
Kaixiang Peng
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Convolutional Neural Network ◽  
Deep Convolutional Neural Network ◽  
Support Vector ◽  
Long Term Memory ◽  
Sensors And Actuators ◽  
Training Time ◽  
Gain Error ◽  
Diagnosis Method

This paper proposes a data-driven method-based fault diagnosis method using the deep convolutional neural network (DCNN). The DCNN is used to deal with sensor and actuator faults of robot joints, such as gain error, offset error, and malfunction for both sensors and actuators, and different fault types are diagnosed using the trained neural network. In order to achieve the above goal, the fused data of sensors and actuators are used, where both types of fault are described in one formulation. Then, the deep convolutional neural network is applied to learn characteristic features from the merged data to try to find discriminative information for each kind of fault. After that, the fully connected layer does prediction work based on learned features. In order to verify the effectiveness of the proposed deep convolutional neural network model, different fault diagnosis methods including support vector machine (SVM), artificial neural network (ANN), conventional neural network (CNN) using the LeNet-5 method, and long-term memory network (LTMN) are investigated and compared with DCNN method. The results show that the DCNN fault diagnosis method can realize high fault recognition accuracy while needing less model training time.

Research of Fault Diagnosis of Belt Conveyor Based on Fuzzy Neural Network

2014 ◽  
Vol 8 (1) ◽  
pp. 916-921
Author(s):  
Yuan Yuan ◽  
Wenjun Meng ◽  
Xiaoxia Sun
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Bp Neural Network ◽  
Fault Location ◽  
Fuzzy Neural Network ◽  
Fuzzy Theory ◽  
Belt Conveyor ◽  
Fuzzy Neural ◽  
Neural Network Algorithm ◽  
Network Fault

To address deficiencies in the process of fault diagnosis of belt conveyor, this study uses a BP neural network algorithm combined with fuzzy theory to provide an intelligent fault diagnosis method for belt conveyor and to establish a BP neural network fault diagnosis model with a predictive function. Matlab is used to simulate the fuzzy BP neural network fault diagnosis of the belt conveyor. Results show that the fuzzy neural network can filter out unnecessary information; save time and space; and improve the fault diagnosis recognition, classification, and fault location capabilities of belt conveyor. The proposed model has high practical value for engineering.

Fault diagnosis method of peak-load-regulation steam turbine based on improved PCA-HKNN artificial neural network

2021 ◽  
pp. 1748006X2110105
Author(s):  
Yifan Wu ◽  
Wei Li ◽  
Deren Sheng ◽  
Jianhong Chen ◽  
Zitao Yu
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Steam Turbine ◽  
Power Plants ◽  
Peak Load ◽  
Clean Energy ◽  
Steam Turbines ◽  
Peak Shaving ◽  
Diagnosis Method ◽  
Load Regulation

Clean energy is now developing rapidly, especially in the United States, China, the Britain and the European Union. To ensure the stability of power production and consumption, and to give higher priority to clean energy, it is essential for large power plants to implement peak shaving operation, which means that even the 1000 MW steam turbines in large plants will undertake peak shaving tasks for a long period of time. However, with the peak load regulation, the steam turbines operating in low capacity may be much more likely to cause faults. In this paper, aiming at peak load shaving, a fault diagnosis method of steam turbine vibration has been presented. The major models, namely hierarchy-KNN model on the basis of improved principal component analysis (Improved PCA-HKNN) has been discussed in detail. Additionally, a new fault diagnosis method has been proposed. By applying the PCA improved by information entropy, the vibration and thermal original data are decomposed and classified into a finite number of characteristic parameters and factor matrices. For the peak shaving power plants, the peak load shaving state involving their methods of operation and results of vibration would be elaborated further. Combined with the data and the operation state, the HKNN model is established to carry out the fault diagnosis. Finally, the efficiency and reliability of the improved PCA-HKNN model is discussed. It’s indicated that compared with the traditional method, especially handling the large data, this model enhances the convergence speed and the anti-interference ability of the neural network, reduces the training time and diagnosis time by more than 50%, improving the reliability of the diagnosis from 76% to 97%.

Sign in / Sign up

Export Citation Format

Share Document