Investigation and optimization of PEMFC-CHP systems based on Chinese residential thermal and electrical consumption data

The proton exchange membrane fuel cell based combined heat and power (PEMFC-CHP) system can recover the waste heat and obtain high energy efficiency. In the present study, a residential PEMFC-CHP system is modeled and optimized. To obtain reliable thermal and electrical consumption data of Chinese residence a questionnaire is designed based on the bottom-up approach which considers the effects of family size, seasons as well as weekends and weekdays. With the data as the input, effects of key parameters including tank setting temperature and volume, PEMFC unit number and operating modes on the CHP efficiency and matching degree of the PEMFCCHP system are investigated for different family sizes in different seasons. The results show that big family has higher energy consumption, especially thermal consumption, leading to higher CHP efficiency but lower matching degree. Increasing the tank setting temperature and volume results in higher CHP efficiency, but a too big tank also generates low matching degree. Increasing the PEMFC unit number can get higher electric efficiency but lower matching degree. Compared to electrical-following mode, constant load operating mode can achieve higher CHP efficiency and matching degree. Besides, a genetic algorithm is coupled with ensemble learning model to optimize the PEMFC-CHP system, leading to improvement of system performance, with 96.91% (96.01%) of CHP efficiency and 99.49% (99.60%) of matching degree for big (small) family. The present study provides reliable thermal and electrical consumption data of Chinese residence and is helpful for the design of residential PEMFC-CHP system.