Unified scheduling and configuration optimization for power to gas system considering thermal dynamics and uncertainties
Fengyuan Zuo,Yuqiong Zhang,Qingsong Hua,Li Sun
DOI: https://doi.org/10.1016/j.seta.2024.103711
IF: 7.632
2024-03-12
Sustainable Energy Technologies and Assessments
Abstract:Power to gas (PtG) system, consisting of electrolytic hydrogen production and carbon capture from coal-fired power plant (CP), is emerging as the large-scale energy storage technology in addressing the fluctuations of the intermittent renewable energy power generation. Based on this, a coal-fired power plant based renewable power to gas system (CP-PtG) is constructed. However, the scheduling and configuration of the system is challenging due to the stochastic uncertainty of the renewable source and load demand. The difficulty is exacerbated by the slow dynamics of CP, which is generally neglected in the operational optimization for simplicity. To this end, this paper develops a two-layer stochastic optimization method. There are 9 given scenarios generated by PV-load joint temporal sequence strategy that preserves historical features, of which the upper layer utilizes the system life-cycle economic cost expectation from the lower layer as the fitness value. Fitness value is a necessary evolutionary criterion in the genetic algorithm searching for the optimal capacity; The lower layer is for the scheduling optimization, in which mixed integer linear programming (MILP) is used for the operational cost minimization of a typical day. The thermal dynamics is treated in this layer as the additional discretized state constraints of the optimization. The upper and lower layer are collaboratively optimized to obtain the optimal system capacity. The optimization results show that the capacity configuration of system is more sensitive to uncertainty, while its operational behavior is more sensitive to the dynamics of PGU. When the uncertainty is considered, the capacity of the key equipment increases by about 10%, which effectively improves the operation stability under intermittent and fluctuation. When the dynamic is considered, the actual output of PGU can better track its scheduling instructions by reducing about 7 times tracking deviation, which ensures system economical and stable operation.
energy & fuels,green & sustainable science & technology