Robust Estimation of State of Charge for Electric Vehicular Application

The primary focus of this paper is to observe the state of charge (SOC) of lithium-ion batteries that are utilized in electric vehicles. The SOC directly affects the performance and range of the vehicle, and accurate estimation of State of Charge can prevent over-discharging or overcharging, thereby improving the battery's useful life and predicting the remaining range. The project involves designing a data logger to measure current and voltage parameters and using the coulomb counting method to calculate SOC. However, due to some drawbacks of the coulomb counting method, this is the goal of the forward-feedback neural network (FNN).[1] The information-gathering process used by data-driven methodologies relies on predicting how the model will be able to leverage the battery's non-linear features.[4] Create a mathematical relationship for the SOC without the use of a physical current by using quantifiable characteristics like voltage and between the input and output data. The dataset used for training the FNN includes features like current, voltage, and capacity. The primary objective of the project is to enhance the effectiveness and dependability of battery systems in electric vehicles.