fig1

Figure 1. Flowchart depicting traditional ML methods for estimating SOC. Input variables (voltage, current, and temperature) undergo data cleaning and feature transformation before being processed by machine learning models: KNN, Decision Trees, SVM, ELM and GPR. The final step involves results processing, accuracy evaluation, and error analysis. SOC: State of charge; KNN: k-nearest neighbor; SVM: support vector machine; ELM: extreme learning machine; GPR: gaussian process regression.