PROJECT TITLE :

Estimation of the state of charge for a LFP battery using a hybrid

ABSTRACT:

LiFePO4 (LFB) batteries are increasingly being used in portable devices, and a method for precisely determining the battery's state of charge (SOC) is urgently needed. An RBF neural network, an orthogonal least squares (OLS) algorithm, and an adaptive genetic algorithm (AGA) are used in this paper to estimate the SOC in discharging state. The number of nodes in the RBF neural network's hidden layer is determined using the OLS method. When a neural network structure has been found that is ideal for RBF, the AGA is used to modify the parameters of RBF, including RBF centres and RBF width. An RBF neural network trained on an LFP battery's SOC is then utilised to predict the battery's capacity. The proposed estimation method is tested using LFP batteries under a variety of discharge situations to demonstrate its usefulness. The proposed method is compared to the Coulomb integration method and a back propagation (BP) neural network in terms of its effectiveness. According to the results, the proposed method outperforms the other methods on the market today.