博弈视角下电-气互联综合能源系统多目标协同优化

Multi-objective Cooperative Optimization of Electricity-Gas Interconnected Integrated Energy System from Game Perspective

考虑到现有综合能源系统多目标优化存在主观性,文中提出一种基于非合作博弈理论的单主体多目标协同优化方法。首先,建立耦合配电网络、配气网络和能源站的电-气互联综合能源系统模型。为使优化调度结果更贴近现实,在配气网模型中采用可变压缩比的压缩机模型,并考虑其非线性耗量特性。其次,基于非合作博弈理论,将能源站单一利益主体的经济性目标和环保性目标视为完全平等且理性的虚拟博弈者,其策略空间由系统内各设备约束组成。再次,为确保博弈模型存在纳什均衡解,在新增辅助变量和约束作用下将收益函数(目标函数)转为拟凸的可微函数。采用凸松弛方法处理模型中非凸非线性约束,并提出非线性迭代策略加速松弛收紧。为便于求解出纳什均衡,使用Nikaido-Isoda函数重构收益函数,将原模型转变为全局最优问题。最后,通过算例验证了所提方法的有效性。

Considering the subjectivity in multi-objective optimization for existing integrated energy systems,this paper proposes a multi-objective cooperative optimization method of single entity based on the non-cooperative game theory.First,an integrated electricity-gas system model coupling the distribution network,gas distribution network,and energy station is established.To make the optimization dispatch results closer to reality,a compressor model with variable compression ratio is employed in the model of gas distribution network.Its nonlinear consumption characteristic is considered.Then,based on the non-cooperative game theory,the economic objective and environmental objective of the energy station with a single stakeholder are treated as completely equal and rational virtual gamers.The strategy space is composed of the constraints of various devices in the system.Moreover,to ensure the existence of Nash equilibrium solutions in the game model,auxiliary variables and constraints are used to transform the payoff functions(objective functions)into pseudo-convex and differentiable functions.The method of convex relaxation is applied to handle non-convex and nonlinear constraints in the model,and a non-linear iterative strategy is proposed to accelerate the relaxation tightening process.In order to facilitate the solving of Nash equilibrium,the Nikaido-Isoda function is used to reformulate the payoff function,transforming the original model into a global optimization problem.Finally,the validity of the proposed method is verified by cases.