考虑风电高阶不确定性的电气综合系统分布鲁棒协同优化调度

A Distributionally Robust Coordinated Optimization Scheduling of Integrated Electricity and Natural Gas Systems Considering Higher-Order Uncertainty of Wind Power

电力系统中大规模风电功率的接入和燃气机组装机的不断扩展使得电力网络和天然气网络联系日益密切,电力-天然气双网耦合系统的协同运行为风电消纳提供了新的途径。针对历史样本数据不准确性及其概率分布建模不完备性导致风电功率呈现出的高阶不确定性问题,提出考虑风电概率分布不确定性的电-气综合能源系统分布鲁棒协调优化调度模型。首先,基于风电出力概率密度函数置信带构造不确定量的模糊集,在此基础上,结合双重因子仿射可调策略建立电力系统两阶段分布鲁棒经济调度主问题,并将上述具min-max形式的含有不确定量的优化问题转化为混合整数线性规划问题;其次,建立考虑燃气机组备用配置有效性的天然气系统运行可行性检验子问题;再次,提出主、子问题迭代求解的协同框架,通过耦合元件决策变量及不可行约束集进行信息交互博弈,实现电力、天然气系统的协调优化调度。最后,通过3个不同规模测试系统的算例验证所提模型和方法的有效性。

Large-scale wind power integration in power system and the continuous expansion of gas-fired unit installation has intensified the interdependency between power grid and natural gas network increasingly. The cooperative operation of electricity-gas dual-network coupled system has provided a new way for wind energy accommodation. In view of higher-order uncertainties of wind power caused by the inaccuracy of historical sample data and the incompleteness of probability distribution modeling, a distributionally robust coordinated optimization scheduling model considering the uncertainty of wind power probability distribution was proposed for integrated electricity and gas energy system. Firstly, an ambiguity set of uncertainties was constructed based on the confidence bands for probability density functions of wind power output. On this basis, a two-stage distributionally robust economic dispatch master problem for power system was established by combining the affine adjustable strategy with dual factors. Moreover, the above optimization problem with uncertainties in the form of min-max was transformed into a mixed integer linear programming problem. Secondly, the operation feasibility check subproblems for natural gas system considering the availability of gas-fired unit reserve configuration were built. Thirdly, a cooperative framework for iteratively solving of master problem and subproblems was developed, in which information interaction and game was carried out by the decision variables of coupling components and infeasible constraint set to achieve the coordination optimization scheduling of power and natural gas systems. Finally, the simulation results on three test systems with different scales demonstrated the effectiveness of the proposed model and approach.