Parameter Identification of PEMFC Air Compressor Based on Trust Region Reflection Least Square Method

The modeling and parameters of the air compressor in a Proton Exchange Membrane Fuel Cell (PEMFC) system are crucial for optimizing the system's overall performance. This paper introduces a semi-empirical model of the PEMFC air compressor, which accurately describes the input-output relationship of the compressor. Next, the trust region reflective least squares algorithm is employed to identify the parameters of the air compressor. This approach enhances the accuracy and stability of parameter identification by formulating trust region sub-problems and integrating reflection techniques. Finally, the identified parameters are substituted into the established model for calculation and compared with actual operational data. The results demonstrate that the identified parameters accurately reflect the actual operating characteristics of the air compressor, validating the effectiveness and accuracy of both the established model and the identification algorithm.