Evaluation of optimal capacity of hot water tank in PEM cogeneration system for residential energy demand profiles

Various residential cogeneration systems have been studied and developed as parts of distributed energy supply system. A residential fuel cell cogeneration system is composed of three units: a fuel cell unit, a thermal storage unit and a back-up boiler. Those three-parts are necessary to compensate a gap between system output with a rigid heat-to-power ratio and heat and power demand. It is difficult to operate the fuel cell system with frequent starts and stops. The peak thermal load is far greater than the maximum thermal output capacity of the fuel cell unit alone. Generally, a cogeneration system performs effectively when all of the recovered heat was utilized in a day. The common issues in promoting residential fuel cell cogeneration systems are the unit price and size. The size must be reduced for use in apartment houses, which constitute about half of the households in Japan. The objectives of this study are to evaluate the rational hot water tank capacity for individual households in order to provide a guide for hot water tank capacity sizing. Sensitivity analysis on the hot water tank capacity from energy saving and economic viewpoints based on an optimal operational planning problem was carried out. Additionally, residential energy demand profiles with different climate conditions are analysed. Our findings show that, in Japan, commutative duration of hot water demand is about 10% or less of the time in one year, and domestic hot water demand rarely observed above 4 kWh in any quarter of an hour. Annual primary energy consumption and operating costs rarely fell when the hot water tank capacity was 7 kWh or more. Additionally, reducing the hot water tank capacity to 5 kWh had only a small effect on the objective values. This suggests that the hot water tank could be smaller than that currently in use. The daily hot water demand gives an excellent estimation of the maximum stored heat of the hot water tank. It was found that the hot water tank capacity can be estimated from the regression lines, or from the fraction of daily hot water demand.