Parameter identification of electrical and thermal model of a lithium polymer battery using particle swarm optimization
In recent years, the demand for lightweight, high-endurance batteries has driven a significant shift in battery cell chemistry, particularly toward Lithium-ion (Li-ion) types. These chemistries offer advantages such as higher volumetric and gravimetric energy densities, making them suitable for appl...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Thesis-Master by Coursework |
Language: | English |
Published: |
Nanyang Technological University
2024
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/180713 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | In recent years, the demand for lightweight, high-endurance batteries has driven a significant shift in battery cell chemistry, particularly toward Lithium-ion (Li-ion) types. These chemistries offer advantages such as higher volumetric and gravimetric energy densities, making them suitable for applications requiring lower weight and higher energy storage. Lithium Polymer (LiP) batteries is a type of Li-ion technology. LiP batteries differ from other Li-ion technologies primarily due to their use of a solid or gel-like polymer electrolyte instead of the liquid electrolyte found in typical Li-ion batteries. This structural change allows LiP batteries to be lighter and more flexible, making them suitable for compact, portable devices. However, these benefits come with the challenge of modelling the complex behaviours of Li-ion batteries. The accurate representation of their electrical and thermal dynamics is crucial for performance prediction and management, but traditional identification methods often fail to capture their behaviour with precision, leading to inaccurate models.To overcome the limitations of conventional methods, this project focuses on using the Particle Swarm Optimization (PSO) algorithm for parameter estimation in LiP battery models. PSO, an evolutionary algorithm inspired by the social behaviour of bird flocking, excels in optimization problems and offers a robust solution to parameter identification. By iteratively adjusting and refining the parameters using real-world experimental data, PSO helps to significantly enhance the accuracy of electrical and thermal model. A pulse discharge test and thermal tests are conducted to obtain experimental data, which is then used as input for the PSO algorithm. The test data is compared against the simulated curves generated by the model, ensuring that the identified parameters align with the actual battery behaviour. The results indicate a high correlation between the experimental data and the responses predicted by the model, validating the effectiveness of PSO in improving parameter estimation. This approach not only addresses the shortcomings of traditional methods but also provides a pathway for the development of a more reliable digital twin for LiP batteries. By utilizing PSO, this project offers a promising solution for enhancing the precision of battery models, with potential applications in the optimization of battery performance in fields such as electric vehicles, drones, and portable electronics. This work contributes to the broader goal of advancing battery technology and addressing the growing demand for efficient energy storage solutions. |
---|