Battery Thermal Management and Algorithmic 3D Temperature Prediction

点击翻译按钮获取中文摘要

Electric vehicles are becoming more widespread, with many manufacturers moving towards electric power for new vehicle models. Battery technology is a bottleneck in the development and uptake of electric vehicles as many members of the public worry about charging times, low driving range, and performance variations due to temperature. Therefore, a method for furthering development in this field is by optimising battery performance with regards to the battery temperature.Lithium-ion batteries have an optimal range of temperature at which they perform best (15 C - 35 C). Below this temperature, the battery performance suffers and battery capacity (ie. driving range) is lost. Above this temperature, unwanted chemical reactions occur inside the battery and risk of thermal runaway increases, leading to catastrophic failure of the battery. Battery temperature is therefore monitored by an on-board battery management system. However, the temperature readings used with these systems are taken at the surface of the battery casing and therefore do not give an accurate representation of the internal temperature at which the battery processes are taking place. This project aims to design and develop a predictive algorithm which allows internal battery temperatures to be predicted using external temperature measurements. The algorithm would then be integrated in a battery management system developed by the project industry partner, Horiba-MIRA. This algorithm would allow battery management systems to more accurately monitor battery temperatures, therefore, improving safety and performance of electric vehicles.In order to achieve this aim, the project has four key objectives:1. Obtain internal thermal measurements of a commercial lithium-ion cell2. Develop a predictive algorithm based on known theory around heat generation and diffusion3. Validate the predictive algorithm using data obtained by cycling a battery across various drive cycles4. Integrate the predictive algorithm in a Horiba-MIRA battery management system and incorporate this within an electric vehicle5. Validate the battery management system integration using standard and real world drive cyclesThe research methodology for this project is as follows:1. Design a quality assured method for instrumenting an internal temperature sensor inside a commercial lithium-ion battery2. Instrument a commercial lithium-ion battery using the developed method, ensuring performance and safety have not been compromised. Cycle the battery using various drive cycles to obtain internal and external temperature readings3. Develop a predictive algorithm which calculates internal temperature using the external measurement4. Integrate the predictive algorithm in a Horiba-MIRA battery management system. Validate the system by placing it in an electric vehicle and completing various standard and real-world drive cyclesThis project will investigate the placement of temperature sensors in prismatic style battery cells. This has been carried out in existing literature for cylindrical and pouch cells but little existing literature investigates the instrumentation of prismatic cells. These are widely used in electric vehicles and therefore instrumentation is important in developing this technology further. This project will also investigate various methods for modelling battery cells. This will include existing models, using various techniques to form hybrid models, and potentially the formation of a novel modelling method to meet the specific needs of this work.This work aligns with the EPSRC research area of Energy Storage and will be completed in collaboration with the industry partner Horiba-MIRA.