Electric Vehicle Battery Charging Estimation by ANN and Fuzzy Logic

Abstract

The state of charge (STOC) of lithium-ion batteries (LTIB) poses a significant challenge in the implementation and advancement of battery management systems, necessitating precise measurement of the battery capacity utilized in electric vehicle (EV) development, thereby emerging as a straightforward issue. Efficient regulation of battery energy, to mitigate dangers associated with overcharging and over-discharging, is feasible only with an accurate calculation of the STOCH, which supports several situations and constraints. In the STOCH esteem analysis, it is imperative to account for the influence of diverse components on the operational cycle of lithium-ion batteries, such as cell aging and cell imbalance, by employing various sophisticated similar circuit models of the batteries. Fitting assessment computations are employed to quantify and improve the precision of STOCH evaluations. The batteries and battery management systems (BMTS) are essential components of electric vehicles (EVs). The STOCH admiration, denoting the excess limit or capacity in the batteries, also constitutes the central border of the BMTS. This approach use a perpetual battery to control the STOCH value via an Arduino UNO microcontroller. Additionally, we will examine the battery display using a simulation of an associate degree Arduino regulator utilizing a MATLAB Simulink model. In MATLAB, an associate degree ANN model is to be developed for the detection of the battery's prudent STOCH.