基于场景树和机会约束规划的含风电场电力系统机组组合

Unit commitment with wind farms using scenario tree and chance-constrained programming

为了解决风电的随机波动性给含大规模风电场电力系统机组组合问题求解带来的影响,采用马尔科夫链原理描述风速变化的规律,并将它与场景树技术相结合,对风电的不确定性进行数学建模。同时基于机会约束规划建立了含风电场机组组合问题的随机数学模型,包含外层机组启停状态优化和内层机组间负荷经济分配两层优化子问题。在求解模型时,将离散粒子群算法(DPSO)与等微增率准则相结合,对两层优化问题进行交替迭代求解;同时提出开停机调整策略改善解的特性。对一个含风电场的10常规机组系统进行算例分析,验证了所提出数学模型和求解方法的合理性和有效性。

In order to cope with the difficulties brought by the volatile and intermittent nature of wind power when solving the unit commitment problem with large-scale wind farms, the basic principles of Markov chain are adopted to describe the regularity of the change of wind speed, and used to model the uncertainty of wind power combining with scenario tree. And this paper presents a stochastic programming model based on chance-constrained programming, and the unit commitment problem is decomposed into two embedded optimization sub-problems： the unit on/off status schedule problem and the load economic dispatch problem. The two problems are solved alternately and iteratively by discrete particle swarm optimization （DPSO） and the equal incremental principle, and an adjusted strategy of units＇ on/off status enhances the algorithm＇s optimization performance. The results on a system with 10 thermal units and wind farms demonstrate the feasibility and effectiveness of the proposed model and algorithm.