Lithium batteries are a type of battery that uses lithium metal or lithium alloys as the positive and negative electrode materials and a non-aqueous electrolyte solution. Gilbert N. Lewis first proposed and researched lithium metal batteries in 1912. In the 1970s, M. S. Whittingham proposed and began researching lithium-ion batteries. Due to the highly reactive chemical properties of lithium metal, its processing, storage, and use place high demands on the environment. With the advancement of science and technology, lithium batteries have become mainstream.

I. Basic Module
(I) Overview of Battery Management Technology
This module explains the operating principles and performance indicators of batteries, clarifies the core functions of battery management systems (BMS), and introduces key BMS software development concepts, such as SOC (State of Charge), SOH (State of Health), and remaining life prediction.

(II) Foundations of Artificial Intelligence and Machine Learning
This module outlines the development of artificial intelligence, explains key concepts in machine learning, and lists case studies of machine learning applications in battery management, building a foundation for understanding the application of this technology. II. Battery State Estimation Applications
(I) State of Charge (SOC) Estimation
This course covers methods such as support vector machines (SVMs) and neural networks (BP/CNN/LSTM), including battery testing, data collection, model building and verification, as well as transfer learning for adapting to different operating conditions and comparing multiple inputs.

This course includes practical examples, such as demonstrations of SOC estimation using support vector machines, neural networks, and transfer learning.
(II) State of Health (SOH) Estimation
This course covers various methods, including estimation based on BP/CNN/LSTM, evaluation under operating conditions such as full charge/discharge, multi-stage constant current/fractional constant current, and dynamic discharge, as well as strategies for integrating cloud-based big data and model error compensation.

This course also includes practical examples, such as practical SOH estimation under operating conditions such as full charge/discharge, multi-stage constant current, and estimation based on cloud-based data. III. Battery Life and Safety Management
(I) Remaining Life (RUL) and Related Predictions
This course covers lithium-ion battery life, thermal runaway characteristics prediction, RUL prediction based on traditional machine learning (SVR, etc.) and joint models, and degradation trajectory prediction, involving data processing, feature extraction, and method validation.
This course provides practical examples, such as practical life prediction based on support vector regression and deep transfer learning.
(II) Thermal Runaway Warning and Fault Diagnosis
This course explains thermal runaway warning methods, including battery fault detection and diagnosis using unsupervised learning methods such as KMeans and DBSCAN, as well as local outlier factors (LOFs) and deep learning. This course covers the principles, process, and results analysis of these methods.
Through practical examples, such as practical abnormal cell detection based on KMeans, DBSCAN, LOFs, and deep learning, students will master application methods.