Realistic accelerated stress tests for PEM fuel cells: Operation condition dependent load profile optimization
Paul Thiele, Luís Gouveia, O. Ulrich, Yue Yang, Yingxu Liu, Maximilian Wick, Stefan Pischinger
Abstract
An algorithm to realistically shorten the test duration of accelerated stress tests based on driving cycles was developed. This algorithm considers all relevant operation conditions of the fuel cell and its reactants and eliminates non-degradation-inducing periods from the original load profile. It consists of four sub-algorithms representing the most critical states in fuel cell electric vehicles (FCEVs): Voltage cycling, idling, humidity cycling and high load operation. To prove the algorithm's functionality, the load profile derived from measurement data of a commercial FCEV - operated under Worldwide Harmonized Light Vehicles Test Cycle (WLTC) and New European Driving Cycle (NEDC) conditions - is optimized. The cycle durations are reduced by 46 % for the WLTC and 60 % for the NEDC. To ensure that the algorithm detects all relevant cycle occurrences, three independent degradation models are used to compare the electrochemical surface area loss, the activation and Ohmic losses and the degradation rates of the original cycle and its optimized variant. All degradation parameters show only small deviations for a test scenario representing one year of operation in the FCEV. The algorithm is open-access and its modular structure makes it useable for other applications and for analyzing the influence of e.g. the operating strategy.