Deep Neural Network based SOH Monitoring of Battery module

Today, lithium battery is used in various fields as cell phone, electric vehicle, unmanned submarine and ESS as the main power. Therefore, for stable use, it is important for the device and the system quickly to detect the defect occurring in the battery and diagnosis the fault accurately. Battery fault can be diagnosed by measuring the state of health (SOH) of the battery, and SOH is changed by various operating conditions. In this paper, a system was implemented to diagnosis the fault of a battery cell through the Deep Neural Network state classifier. In this method, DNN state classifier utilized the discharge voltage data that was obtained by operating the lithium battery cell at high temperature. From the experiment results, we know that the proposed battery SOH monitoring method diagnosing the state of battery very well.