Use case improved system identification for battery models

用例改进了电池模型的系统识别

• 批准号: 2283221
• 金额: --
• 依托单位: University of Bath
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2019
• 资助国家: 英国
• 起止时间: 2019 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2283221
• 关键词: Use; case; improved; system; identification

In the area of battery modelling, due to their simplicity and capability to run in real-time, equivalent RC circuit models are often employed. Previous work conducted by AVL indicates that the third-order version of this model (3RC) coupled with a lumped-capacitance thermal model can capture the system-level behaviour of Li-ion batteries under static as well as dynamic conditions. The overall model is referred to as the 'MoBat model' and has been selected to represent an element of a predefined library The main aim of the research is the development of an effective system identification process targeting the parametrisation of the MoBat model, and its expansion to be applied for other battery models representative for Li-ion batteries. The ideal outcome is represented by a demonstrated methodology capable to produce optimally parametrised models for given requirements, recognized by a set of KPIs targeting the validity and quality of the parametrisation, while reducing testbed as well as computational effort. While the main route will imply that parameterization data is already available, a separate route to parametrisation starting without any data will also be investigated. This part of the project will aim towards developing a highly effective testing routine in the format of model-based or active design of experiments capable of providing highly effective datasets to be used for the parametrisation of the MoBat model relative to the intended use case.The main benefit of the proposed research for automotive companies as well as academic researchers and research councils is represented by facilitated access for creating Li-ion battery models. This is set to be achieved through a predefined seamless process which requires a limited amount of expert knowledge without compromising results. The main advantages of the parametrisation process to be proposed have been summarised as it follows:Increased quality of the parameter optimization, promoting higher fidelity models to be obtained without additional efforts.Reduction of computational cost, resources required for model development and testbed efforts.Improved communication between the user and the system employed for model development.The research presented is set to be directly integrated as part of the ModelFactory framework made readily available by AVL. Consequently, the outcome of the project will lead to a reduction in the overall time and cost associated with the research activities targeting Li-ion batteries and ultimately contribute to more efficient and environmental-friendly vehicles. At the same time, the outcome of the project presented will represent a new milestone in the area of model development to be used as the groundwork for future research.The parties funding the project are the EPSRC and AVL List GmbH, as it is set for completion by an AAPS CDT student.

在电池建模领域，由于其简单性和实时运行的能力，经常使用等效的RC电路模型。 AVL进行的先前工作表明，该模型的三阶版本（3RC）与集体电容热模型相结合可以捕获在静态和动态条件下锂离子电池的系统级行为。总体模型被称为“ MOBAT模型”，并已选择代表预定义库的要素。研究的主要目的是开发针对MOBAT模型参数化的有效系统识别过程，并将其扩展用于其他电池模型代表使用Li-ion电池。理想的结果由能够为给定要求产生最佳参数化模型的演示方法来表示，该模型由一组针对参数化的有效性和质量的KPI认可，同时减少测试平台和计算工作。虽然主要路由将暗示已经可用的参数化数据，但也将研究参数构造的单独途径，而无需研究任何数据。 This part of the project will aim towards developing a highly effective testing routine in the format of model-based or active design of experiments capable of providing highly effective datasets to be used for the parametrisation of the MoBat model relative to the intended use case.The main benefit of the proposed research for automotive companies as well as academic researchers and research councils is represented by facilitated access for creating Li-ion battery models.这将通过预定义的无缝过程来实现，该过程需要有限的专家知识，而不会损害结果。以下是汇总的参数化过程的主要优势：参数优化的质量提高，促进较高的保真度模型，而无需额外的努力。还原计算成本，模型开发所需的资源和测试工作所需的资源。对模型开发的用户和系统之间的沟通进行了验证。所介绍的研究以整合了一部分，该模型已成为一个模拟的一部分。因此，该项目的结果将导致与针对锂离子电池的研究活动相关的整体时间和成本降低，并最终有助于更高效，更环保的车辆。同时，提出的项目的结果将代表模型开发领域的一个新里程碑，将用作未来研究的基础。当事方为该项目提供资金是EPSRC和AVL列表GmbH，因为AAPS CDT学生将其设定为完成。