基于灵活储氢和蓄电池联合储能的混合可再生能源系统规划运行协同优化

MULTI STAGE PLANNING AND OPERATION OPTIMIZATION OF HYBRID RENEWABLE ENERGY SYSTEMS WITH FLEXIBLE HYDROGEN-BATTERY STROAGE

提出一种包含灵活储氢和蓄电池联合储能的混合可再生能源系统,并提出一种考虑极端运行工况的规划运行多阶段协同优化方法,第1阶段以最小化净负荷为目标,优化混合系统中风电与光伏的容量;第2阶段在正常运行、极端全充电、极端全放电3种典型工况下对联合储能系统进行容量运行双层协同优化。上层规划模型以平准化储能度电成本最小为目标优化储能容量,下层调度模型以功率偏差最小为目标优化储能系统运行策略,并采用元启发式优化算法和混合整数线性规划进行求解。研究表明:1)考虑极端全充电和全放电运行工况时,储氢系统的容量配置灵活性可以使系统更经济地应对极端工况,体现了氢储能的优越性;2)基于联合储能的混合可再生能源系统比单一储能系统具有更高的经济效益,其平准化储能度电成本相较于氢储能降低了3.97%,相较于蓄电池储能降低了8.25%;3)相比于基于运行规则的容量优化方法,该文提出的规划运行多阶段协同优化方法使系统平准化储能度电成本降低了13.7%。

In this paper,a hybrid renewable energy system containing flexible hydrogen storage and battery storage is proposed,along with a multi-stage co-optimization method for planning and operation.In the first stage,the capacity of WT and PV in the hybrid system is optimized to minimize the net load;In the second stage,a bi-level co-optimization of capacity operation is conducted for the hybrid energy storage system under three typical operating conditions,namely normal operating conditions,extreme full charging conditions and extreme full discharging conditions,where the upper-level model optimizes the storage capacity based on minimizing the cost of levelized cost of storage(LCOS)and the lower-level model optimizes the operation strategy to minimize the power deviation.The optimization problem is solved by a meta-heuristic algorithm coupled with mixed-integer linear programming.The study shows that.1)When considering extreme full-charge and full-discharge operating conditions,the flexible combination of the components of the hydrogen storage system allows the system to cope with extreme operating conditions more economically,reflecting the superiority of hydrogen storage.2)The hybrid energy storage system has higher economic efficiency than the single energy storage system,and its LCOS is reduced by 3.97%compared to the single hydrogen storage and 8.25%compared to the single battery storage.3)The multi-stage co-optimization method of planning and operation proposed in this paper reduces the LCOS by 13.7%compared with the rule-based capacity optimization method.