Cascaded H-Bridge Multilevel Inverter System Fault Diagnosis Using a PCA and Multiclass Relevance Vector Machine Approach

Wang, Tianzhen; Xu, Hao; Han, Jingang; Elbouchikhi, Elhoussin; Benbouzid, Mohamed

doi:10.1109/tpel.2015.2393373

Cited by 229 publications

(105 citation statements)

References 34 publications

Supporting

Mentioning

105

Contrasting

Order By: Relevance

“…It can assign zero-mean Gaussian prior distribution to the weight vector. The maximizing marginal likelihood function method is applied to estimate hyper parameter, and the automatic regulation mechanism is adopted to avoid the problem of the difficulty in determining the regularization coefficient [10,11]. Let…”

Section: Relevance Vector Machine (Rvm)mentioning

confidence: 99%

Transformer fault diagnosis based on chemical reaction optimization algorithm and relevance vector machine

et al. 2017

View full text Add to dashboard Cite

Abstract.Power transformer is one of the most important equipment in power system. In order to predict the potential fault of power transformer and identify the fault types correctly, we proposed a transformer fault intelligent diagnosis model based on chemical reaction optimization (CRO) algorithm and relevance vector machine(RVM). RVM is a powerful machine learning method, which can solve nonlinear, high-dimensional classification problems with a limited number of samples. CRO algorithm has well global optimization and simple calculation, so it is suitable to solve parameter optimization problems. In this paper, firstly, a multi-layer RVM classification model was built by binary tree recognition strategy. Secondly, CRO algorithm was adopted to optimize the kernel function parameters which could enhance the performance of RVM classifiers. Compared with IEC three-ratio method and the RVM model, the CRO-RVM model not only overcomes the coding defect problem of IEC three-ratio method, but also has higher classification accuracy than the RVM model. Finally, the new method was applied to analyze a transformer fault case, Its predicted result accord well with the real situation. The research provides a practical method for transformer fault intelligent diagnosis and prediction.

show abstract

Section: Relevance Vector Machine (Rvm)mentioning

confidence: 99%

Transformer fault diagnosis based on chemical reaction optimization algorithm and relevance vector machine

et al. 2017

View full text Add to dashboard Cite

show abstract

“…The most popular methods are Phase Shifted Pulse Width Modulation (PSPWM) and level-shifted PWM [21,22]. PSPWM method is the most suitable PWM and is recommended for cascaded H-Bridge converters [23] and is extensively used for its ease of implementation and even power distribution amongst the cells.…”

Section: Multilevel Chb Convertermentioning

confidence: 99%

“…This method is capable of detecting the fault and its location in a few hundreds of microseconds, which is several times faster compared to the previous methods [12,19,21] and is comparable with [20]. Moreover, the proposed method uses only simple math, relational and state machine blocks and therefore its implementation on a digital target like FPGA would be easy.…”

Section: Introductionmentioning

confidence: 97%

“…Also this approach may experience difficulties in finding faulty cell"s location when a higher number of cells is utilized. In [21], Fast Fourier Transform is used on the output voltages to preprocess these signals. Then PCA (Principal Component Analysis) is used to extract the fault signatures and finally multiclass Relevance Vector Machine (mRVM) is used to classify fault samples.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fast Detection of Open-Switch Fault in Cascaded H-Bridge Multilevel Converter

et al. 2017

View full text Add to dashboard Cite

Abstract-Cascaded H-Bridge converter has recently been utilized in different high-power applications due to its modular and simple structure. In order to have a balanced operation after a fault occurrence in this converter, it is necessary to detect the switch fault and its location. In this paper, a fast power switch fault detection method is presented to identify the fault and its location. Only one voltage measurement per phase is required by this method, and the fault detection is faster compared to the existing methods. Moreover, it is suitable for implementation on an FPGA device, due to the use of simple math, relational and state machine blocks. The proposed method is verified by computer simulations and FPGA-based experimental tests.

show abstract

“…Multi-Relevance Vector Machine (mRVM) based classification is carried out with the Principal Component Analysis (PCA) for the feature extraction. This method proved to be highly sparse and fast in detection and response time [14]. A FPGA based implementation of the short circuit fault of the MLI is discussed which needs only one voltage sensor per phase [15].…”

Section: Introductionmentioning

confidence: 99%

Robust Fault Detection of Multilevel Inverter using Optimized Radial Basis Function based Artificial Neural Network in Renewable Energy Power Generation Application

Manjunath¹,

Kusagur²

2018

IJCA

View full text Add to dashboard Cite

The Optimized Artificial Neural Network based fault detection of the multilevel inverters are of research interest since the HVDC system and Electrical Drives system has started using the multilevel inverters as the power processing units. While the power supply to the multilevel inverters is constant the results of the THD obtained would be stable. But the multilevel inverters that are used with Renewable Energy Resources as the power supply, the variation in the supply would largely affect the THD values thus a robust model for the fault detection of multilevel inverters which not only considers the THD but the positive sequence voltage, negative sequence voltage, zero sequence voltage and angle of the inverter output is considered. This paper is an attempt to develop a robust fault detection method using the optimized Artificial Neural Network (ANN) using the Radial Basis function. The objective function of this optimization algorithm is the minimization of the Mean Square Error while the ANN is getting trained. The optimization is carried out by the use of the weight and the bias value search in the search space, which would enhance the training of the ANN. Particle Swarm Optimization (PSO) and the Cuckoo Search Algorithm (CSA) are considered for the optimization algorithms. Matlab based implementation is carried out and the results are measured and tabulated. It is observed that the CSA algorithm is performing better while training the ANN for the fault detection.

show abstract

Cascaded H-Bridge Multilevel Inverter System Fault Diagnosis Using a PCA and Multiclass Relevance Vector Machine Approach

Cited by 229 publications

References 34 publications

Transformer fault diagnosis based on chemical reaction optimization algorithm and relevance vector machine

Transformer fault diagnosis based on chemical reaction optimization algorithm and relevance vector machine

Fast Detection of Open-Switch Fault in Cascaded H-Bridge Multilevel Converter

Robust Fault Detection of Multilevel Inverter using Optimized Radial Basis Function based Artificial Neural Network in Renewable Energy Power Generation Application

Contact Info

Product

Resources

About