Fuel Cell Model Development

随着汽车行业继续脱碳，燃料电池汽车为传统内燃机汽车提供了一种很有前途的替代方案。确定燃料电池的特性实际上是释放其运行和开发中的性能和效率收益的基础。根据每个用例中所需的参数数量和保真度级别，使用制造商提供的数据和规范来开发理论模型通常非常耗时。因此，这项工作的主要目标是参数化和开发汽车燃料电池模型，该模型将改进和满足当前来自工业用例的模型要求，例如硬件在环系统和用于车载监控的虚拟传感器。该项目的初始阶段将包括通过深入的文献搜索和审查缩小范围，以及与行业合作伙伴沟通以定义模型本身。从那时起，燃料电池模型开发过程将对研究项目的很大一部分做出贡献，模型需要满足用例，同时还需要改进行业合作伙伴已有的可用模型。随后，将仔细设计实验程序，以优化数据采集过程。这将导致对模型的验证，减少模型值和测量值之间的误差。预计该项目所需的数据将由AVL提供，并在晚些时候使用能源部方法进行实验测试。参数化程序旨在通过优化参数化过程来减少UUT上所需的硬件。与该项目相关的主要挑战是整个过程中处理的数据量。燃料电池的建模很复杂，因此包括一组巨大的参数，这导致了全面描述电池所需的大量数据。这反过来又增加了实验过程中所需的测量。鉴于这项工作的预期输出是开发和定义燃料电池模型、参数化过程和更高效、更容易获得的实验过程，因此产生了明显的好处和影响。模型开发过程将允许简化工业中使用的模型生成和选择，并打开在该主题内探索的新的潜在途径，如细胞老化，目前这是一个相对较少了解的领域。这将使人们更好地了解燃料电池开发过程中的性能、效率和寿命方面的增长。简化实验设计程序和参数流程可以减少与测试相关的财务和时间成本，但也允许更广泛地访问模型，因为这些过程将得到优化。

As the automotive industry continues to de-carbonise, Fuel Cell vehicles provide a promising alternative to conventional ICE vehicles. Characterising a fuel cell virtually is fundamental in unlocking performance and efficiency gains in its operation and development. Using data and specifications from the manufacturer to develop a theoretical model is often time consuming given the number of parameters and the level of fidelity desired in each use case. Therefore, the main aim of this work is to parameterise and develop automotive fuel cell models which will improve and fulfil the current model requirements derived from industry use cases such as Hardware in the Loop systems and virtual sensors for onboard monitoring. The initial stages of the project will include narrowing the scope through both an in-depth literature search and review and communication with the industry partner to define the model itself. From here the fuel cell model development process will contribute to a significant proportion of the research project, with the model needing to satisfy the use case but also improve upon what is already available to the industry partner. Following this, careful design of the experimental procedure will take place in order to optimise the data acquisition process. This will then lead to the verification of the model, reducing error between modelled and measured values. It is envisaged that the required data for the project will be supplied by AVL as well as experimental testing at a later date using DoE methods. The parameterisation procedure will aim to reduce the hardware required on the UUT by optimising the parametrisation process. The main challenge associated with this project is the amount of data handled throughout. Fuel cells are complex to model and therefore include a huge set of parameters leading to a large amount of data needed to fully characterise the cell. This in turn increases the measurements required during the experimental procedures. Given that the intended output of this work is to develop and define a fuel cell model, parameterisation process and experimental process which is more efficient and accessible there are clear benefits and impacts as a result. The model development process will allow for a simplification of the model generation and selection used in industry, as well as unlocking new potential avenues to explore within the subject such as cell aging which at this moment is a relatively poorly understood area. This will allow better understanding of performance, efficiency, and longevity gains in the development of fuel cells. Streamlining the experimental design procedure and parametrisation process reduces both financial and time costs associated with testing but also allows models to be more widely accessible, given these processes will have been optimised.