复杂需求下巨型水电站跨省交易电量曲线分解方法

Decomposition Method of Cross-provincial Monthly Electric Quantity Curve of a Giant Hydropower Station Under Complex Demands

西南巨型水电站参与电力市场涉及多省、多品种、多时间尺度等多维电量曲线分解难题,给水电市场化运行带来很大挑战。该文依托溪洛渡右岸电站实际工程,考虑电网调峰与企业效益复杂需求,提出巨型水电站月度电量曲线分解多目标协调方法。以余荷均方差最小构建多电网调峰目标,利用多品种差异电价和峰平谷分段市场电价构建发电效益目标,引入理想点法实现目标分级进而将多目标优化转换为一系列差异约束边界的单目标问题;采用分段线性技术处理非线性水力约束、三角剖分技术处理电站出力特性曲线,利用混合整数线性规划(mixed integer linear programming,MILP)方法实现模型高效求解。采用溪右电站月度电量交易申报实际数据进行验证分析。结果表明,该文模型可以得到合理的月度电量分日、分时、分省、分品种的发电曲线过程,并通过两个实例分析不同来水和计划电量对调峰与收益的影响,发现汛期较枯期影响更大,且计划电量比例是协调调峰和发电收益的重要因素。

The participation of giant hydropower stations in the power market in southwest China involves the decomposition of multi-dimensional power curve with multiple provinces,varieties and time scales,which brings great challenges to the market operation of hydropower.This article relies on the actual project of Xiluodu Right Bank Power Station,a multi-objective coordination method of monthly electric quantity curve decomposition for giant hydropower stations is proposed considering the complex needs of network peak adjustment and enterprise benefit.The peak shaving goal of multi-grid is constructed with the minimum residual load mean square error,and the generation efficiency goal is constructed with the multi-variety differential electricity price and peak-flat-valley segmented market electricity price.The ideal point method is introduced to achieve the goal classification,and then the multi-objective optimization is transformed into a series of single-objective problems with differential constraint boundaries.The peak shaving goal of multi-grid is constructed with the minimum residual load mean square error,and the generation efficiency goal is constructed with the multi-variety differential electricity price and peak-flat-valley segmented market electricity price.The piecewise linear technique is used to deal with the nonlinear hydraulic constraints,the triangulation technique is used to deal with the power station output characteristic curve,and the mixed integer linear programming(MILP)method is used to solve the model efficiently.The actual data of the monthly electricity transaction declaration of right bank of Xiluodu Hydropower Station is used for verification and analysis.The results show that the model in this paper can obtain a reasonable monthly electricity generation curve process of daily,hour,province,and category.Through two examples to analyze the impact of different incoming water and planned electricity on peak shaving and revenue,it can be found that the impact of flood season is greater than that of dry season,and the proportion of planned electricity is an important factor in coordinating peak shaving and power generation revenue.