<p>Proton exchange membrane fuel cells
(PEMFCs) are promising devices for conversing chemical energy into electrical
energy due to their versatile properties, such as high power density, quick
start-up, lower operating temperature, and portability, etc. For PEMFC
technology to outperform the incumbent technologies, artificial intelligence
(AI) based multi-objective optimisation (AI-MOO) has been employed to facilitate
the design and applications of PEMFC since AI-MOO is flexible enough to consider
various factors simultaneously in the customized multiple objective functions
and under new or updated case situations. This review provides a comprehensive
literature survey on AI-MOO employed in PEMFC field. Firstly, AI-MOO were introduced in detail,
including the definition, categories and framework. Then the objectives,
intelligent algorithms and trade-off methods that are commonly used in PEMFC were
tabularised and evaluated. The
application of AI-MOO in PEMFC were summarised systematically based on the
application areas, including the PEMFC components, kinetics and thermodynamics,
control and monitoring systems, the overall performance, and the hybrid systems.
The related studies were tabularised and discussed, especially algorithms,
variables, objectives and optimisation results. Finally, <a>this review addressed the current challenges in the
research area and proposed research implications for future investigations.</a></p>
Multi-objective optimal control for proton exchange membrane fuel cell via large-scale deep reinforcement learning
