The new power system in agricultural parks faced the problems of poor system operation stability and low comprehensive economic efficiency due to power fluctuation on the generation side and diverse and uncertain load demand on the consumption side. For this reason, this paper constructed a model of optimized allocation of energy storage capacity based on load division, aiming to improve the stability and economic efficiency of the system. Firstly, a two-stage energy storage allocation tactic was determined in accordance with load segmentation. In this regard, the conventional load demand was preferentially supplemented by photovoltaic power, and the root-mean-square （RMS） envelope technique was harnessed to quell the unsteady power oscillations. Secondly, for the agricultural load demand, the net present value （NPV） was constructed as the optimization objective function. This was then conjoined with the time-of-use tariff strategy to impel the energy storage capacity to perform peak shaving, valley filling, and consumption compensation. Finally, by leveraging the summer and winter data of an agricultural park and deploying the particle swarm optimization algorithm, the outcomes of the two-stage allocation were attained. Namely, the overall power oscillations of the summer and winter loads are abated by around 32% and 27% respectively. The peak-to-valley ratios of the loads are lessened by 13.1% and 14.4% respectively. Moreover, the comprehensive economic return of energy storage is enhanced by 21% when compared with the direct allocation method. The exemplifications corroborated that the methodology proposed in this paper could efficaciously attenuate the microgrid power fluctuation quandary and augment the comprehensive economic benefits engendered by energy storage allocation.

PV power  /  power flow optimization  /  energy storage  /  agricultural park  /  load division  /  time-of-use tariff