考虑源荷双重不确定性的电气互联综合能源系统分布鲁棒优化调度

针对电气互联综合能源系统中不确定因素对运行调度带来的风险问题,提出一种考虑新能源负荷双重不确定性的电气互联综合能源系统分布鲁棒优化调度模型。以系统运行总成本最小为目标函数,建立电气互联综合能源系统随机规划模型,采用矩不确定分布鲁棒优化方法(DRO-MU),构建风电出力和电力负荷的矩不确定集合,在源荷双重不确定集合下将随机规划模型转化为矩不确定分布鲁棒优化模型,通过拉格朗日对偶原理将鲁棒模型转化为确定性的半正定规划模型进行求解。仿真结果表明,与不考虑或仅考虑新能源、负荷不确定性相比,考虑源荷双重不确定性的DRO-MU模型运行成本有所增加,但方案降低了不确定性因素给系统运行带来的风险,且更加符合实际。与随机优化方法和传统鲁棒方法相比,DRO-MU方法既保证了调度策略的鲁棒性,又克服了其过于保守的问题;随着源荷矩不确定集范围的增大,系统运行成本增加,通过合理设置二者矩不确定集范围,实现系统经济性与鲁棒性的平衡。

Robust Optimal Scheduling of Power-Gas Interconnected Integrated Energy System with Double Uncertainty of Source and Load

Aiming at the risk of uncertainties to operation and scheduling of the electric-gas interconnected integrated energy system, a robust optimal scheduling model for this system is proposed by considering double uncertainties of new energy and load. A stochastic programming model for the integrated energy system with Electricity-Gas interconnection is established with the total minimal operating cost as the objective function. The moment uncertainty distributed robust optimization method (DRO-MU) is used to construct the uncertainty set of wind power output and load. Under the source-load double uncertainty set, the stochastic programming model is transformed into a moment-indeterminate distributed robust optimization model, which is transformed into a deterministic semi-definite programming model by Lagrangian duality principle. The simulation results show that the operating cost obtained by using the DRO-MU model considering the dual uncertainty of source and load is increased compared without considering new energy or load uncertainty. However, the new scheme reduces the operation risk brought by uncertainties, and is more realistic. Compared with the stochastic optimization method and the traditional robust method, the DRO-MU method not only ensures the robustness of the scheduling strategy, but also overcomes its overly conservative problem. As the range of source-load moment uncertainties increases, the system operation cost increases. By setting the range of source-load moment uncertainties reasonably, the system economy and robustness can be balanced.