On improving training time of neural networks in mixed signal circuit fault diagnosis applications

Mohammadi, Karim; Mahdavi, Seyyed Javad Seyyed

doi:10.1016/j.microrel.2007.12.009

Cited by 10 publications

(5 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…discriminated finally which kind of with the most votes just think x what kind. When the phase at the same time, x is classified as class standard small that [5] .…”

Section: Fault Classification Based On Svmmentioning

confidence: 99%

Study on the Transducer Power Electronic Circuits Fault Diagnosis Based on SVM

Li¹,

Yan-qiu²,

Li³

et al. 2018

Proceedings of the 2018 7th International Conference on Energy and Environmental Protection (ICEEP 2018)

View full text Add to dashboard Cite

The main circuit of inverter power transform fault mode analysis in the paper, using multiple support vector machine (SVM), implements the main circuit of inverter power transform fault diagnosis. In equipment fault diagnosis, online or offline to analyze the circuit, its running state, the learning ability of neural network, the fault parameters and the relationship between the cause of the problem through neural network learning after stored in its structure and power, you can through the analysis of fault parameters, it is concluded that the cause of the problem, so as to realize the automatic diagnosis of faults. Four characteristic parameters of reflecting fault state are distilled with analysis fault types and reasons of transducer power electronic circuits. Identification of fault mode can be completed utilizing multiply support vector machine(MSVM).

show abstract

“…discriminated finally which kind of with the most votes just think x what kind. When the phase at the same time, x is classified as class standard small that [5] .…”

Section: Fault Classification Based On Svmmentioning

confidence: 99%

Study on the Transducer Power Electronic Circuits Fault Diagnosis Based on SVM

Li¹,

Yan-qiu²,

Li³

et al. 2018

Proceedings of the 2018 7th International Conference on Energy and Environmental Protection (ICEEP 2018)

View full text Add to dashboard Cite

show abstract

“…As the improved algorithm performance better, we use Levenberg-marquardt back propagation (LM-BP), as is showed in the following equations. 4 shows the model of a neural network [3]. We can conclude that ANN is trained to perform pattern recognition by adjusting the connection weights to find out the mapping relationship of input feature vectors and their corresponding target vectors.…”

Section: B Design Of Neural Network Structurementioning

confidence: 99%

“…The latter bypasses the step of obtaining a mathematical mode and deals directly with the data. This is more appealing when the process being monitored is unknown to be linear or when this is too complicated to be extracted from the data [3]. Therefore the purpose of using ANN is the realization of nonlinear functions which can estimate a suitable output from any inputs after training with a sample dataset.…”

Section: Introductionmentioning

confidence: 99%

Artificial Neural Network Classifier Design Using Genetic Algorithm and Wavelet Transform in Fault Diagnosis

Wang

2010

2010 2nd International Conference on Information Engineering and Computer Science

View full text Add to dashboard Cite

A diagnosis method basing on neural network classifier, genetic algorithm (GA) and wavelet transform is proposed for a pulse width modulation voltage source inverter. It is used to detect and identify the transistor open-circuit fault. BP neural network (BPNN) is capable of recognition. However, it has shortcomings obviously. These are just advantages of GA, which has ability of global search. Thus GA is integrated into BPNN for obtaining complementary advantages. Besides, Wavelet transform is employed as a fast and effective means analyzing the transient waveforms, as an alternative to the traditional Fourier transform. The Hybrid algorithm can offer higher detection efficiency and reliability.

show abstract

“…Neural network-based method is one of the most commonly used pattern recognition-based condition monitoring methods, and is composed of two parts: feature extraction and pattern recognition [9][10][11][12]39]. Many researchers have studied on neural network-based condition recognition and fault diagnosis and have achieved good results [13][14][15][16]. However, this method depends too much on prior knowledge about signal processing techniques and diagnosis expertise hence the recognition results rely much on diagnosticians.…”

Section: Introductionmentioning

confidence: 99%