Development of Remaining Useful Life (RUL) Prediction of Lithium-ion Battery Using Genetic Algorithm-Deep Learning Neural Network (GA-DNN) Hybrid Model
Muchamad Iman Karmawijaya (a*), Irsyad Nashirul Haq (b), Edi Leksono (c), Augie Widyotriatmo (d)

(a,b,c,d) Department of Engineering Physics, Faculty of Industrial Technology, Institut Teknologi Bandung, Indonesia
*iman95[at]gmail.com
(b) National Center for Sustainable Transportation Technology, Indonesia

Abstract

Determination of Remaining Useful Life (RUL) battery is essential in battery management system design. This study focused on the use of evolutionary algorithms to optimize Deep Learning Neural Network (DLNN) algorithm to predict the RUL battery. For this purpose, a Genetic Algorithm (GA) was developed to select the most significant features in the raw dataset. After that, a GA-DLNN hybrid model was developed to select optimal parameters for the DLNN model, including network algorithm, activation function for hidden neurons, number of hidden layers, and the number of neurons in each hidden layer. A database of real data set of lithium-ion battery cycle life from NASA was used. The dataset was divided into three parts, namely the training set, validation, and testing set for the construction, validation and testing phases of the proposed model, respectively. Various quality assessment criteria, namely the coefficient of determination (R2), Index of Agreement (IA), mean absolute error (MAE) and root mean squared error (RMSE), were used to evaluate the performance of the machine learning (ML) algorithms. The GA-DLNN hybrid model was shown to exhibit the ability to find the most optimal set of parameters for the prediction process. By using 11-input the GA-DLNN result showed R2 value of 0.959, MAE = 7.167, RMSE = 9.96 and IA = 0.989. The results showed that the performance of the hybrid model using only the most critical features gave the highest accuracy, compared with those obtained by the hybrid model using all input variables.

Keywords: Prediction, Remaining Useful Life, Deep Neural Network, Genetic Algorithm, Battery Management System&#8239-

Topic: Battery Technology and Management System