Quantum Restricted Boltzmann Machines-Based Feature Selection for Electric Vehicle Battery Health Monitoring

Battery health monitoring plays a critical role in ensuring the longevity and efficiency of battery-operated systems such as electric vehicles (EVs). Feature selection is a crucial step in designing battery health monitoring algorithms, which often face challenges due to the high dimensionality of the data and the complex, nonlinear relationships among features. This paper introduces a Quantum Restricted Boltzmann Machines (QRBM)-based feature selection method for EV battery health monitoring by leveraging RBM and quantum computing (i.e., D-Wave quantum computer in this paper). The feasibility of the proposed method has been validated by experiments. The experimental results validate that the proposed QRBM-based feature selection method outperforms traditional feature selection methods such as Particle Swarm Optimization and Pearson Correlation in terms of feature uniqueness and reducing number of features.