Comparison of SOC Estimation Performance with Different Training Functions Using Neural Network

Jian, Wei; Jiang, Xuehuan; Zhang, Jinliang; Xiang, Zhengtao; Jian, Yubing

doi:10.1109/uksim.2012.69

Cited by 8 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Normalization data is aimed more efficiency and robust when computational. This is formula of normalization data before training data [5]: (2) Where Xmin and Xmax are X minimum and maximal input vectors in the BPNN. The testing process use same data scale with Xmin and Xmax when training data 5.…”

Section: Back Propagation Neural Networkmentioning

confidence: 99%

“…Training will be stopped when error of system is very low or amount of epoch has been defined from users and convergence condition. The formula MSE and R 2 describe in equations 4 and 5 [5]. ,…”

Section: -2 Figure 2 Training Pattern Of Neural Networkmentioning

confidence: 99%

“…The bias can trainable when associated with a trainable weight to be modified as other weight . Hence the bias is realized in terms of an input with some constant (+1 or +B) input, and the exact bias bi at the i th neuron is then given [5]: ,…”

Section: -2 Figure 2 Training Pattern Of Neural Networkmentioning

confidence: 99%

“…Activation function in the hidden layer is hyperbolic tangent sigmoid 5. Hyperbolic tangent sigmoid can be expressed with [5]: ,…”

Section: Proposed Bpnn Modelmentioning

confidence: 99%

“…Here is a equation of linear function [5]: (8) Because of the special structure of the BPNN, we can use multiple variables as independent variables,putting them into the network, the output voltage or current as a dependent variable. In Fig.…”

Section: Proposed Bpnn Modelmentioning

confidence: 99%

See 4 more Smart Citations

Backpropagation neural network models for LiFePO4 battery

Primadusi

Cahyadi

Prasetyo

et al. 2016

AIP Conference Proceedings

View full text Add to dashboard Cite

Abstract. Neural Networks have been used in system control, medicine, pattern recognition and business. The backpropagation neural network (BPNN) appear to be most popular and have been widely used in many applications. BPNN is a supervised learning technique for training multilayer feedforward neural networks. The gradient or steepest descent method is used to train a BPNN by adjusting the weights . The purpose of update numerical weights is minimize error of network between target and output. In this paper, focus with BPNN modeling with data battery for training and testing. We used discharge and Urban Dynamometer Driving Schedule (UDDS) as training data and testing data, respectively Architecture of BPNN consist of input layer, hidden layer and output layer. The otherhand, using BPNN has problem to define amount of hidden neurons. In this study, we used current or voltage as input in input layer, one hidden layer with 8 neurons and one output layer. We used Levenberg-Marquardt algorithm to get fast iteration when computation. The experiment used 2200 mAh of LiFePO4 battery. Result of this research show that Mean Squared Error (MSE) value when current as input and voltage as target is 0.021135 with regresion is 0.626. Then MSE value when voltage as input and current as target 0.029925 with regresion is 0.5213. In this study relationship between voltage and current battery is nonlinear.

show abstract

Section: Back Propagation Neural Networkmentioning

confidence: 99%

Section: -2 Figure 2 Training Pattern Of Neural Networkmentioning

confidence: 99%

Section: -2 Figure 2 Training Pattern Of Neural Networkmentioning

confidence: 99%

“…Activation function in the hidden layer is hyperbolic tangent sigmoid 5. Hyperbolic tangent sigmoid can be expressed with [5]: ,…”

Section: Proposed Bpnn Modelmentioning

confidence: 99%

Section: Proposed Bpnn Modelmentioning

confidence: 99%