基于梯度提升决策树的含光伏接入微电网调度优化  

Dispatching Optimization Including Photovoltaic Access to Microgrid Based on Gradient Improvement Decision Tree

作　　者：李春华 雷艾虎 郭庭哲 王家祥 王昆能 LI Chunhua;LEI Aihu;GUO Tingzhe;WANG Jiaxiang;WANG Kunneng

机构地区：[1]国网甘肃省电力公司白银供电公司,甘肃白银730900

出　　处：《电力系统装备》2024年第10期49-51,共3页Electric Power System Equipment

摘　　要：针对常规的含光伏接入微电网调度约束固定速度慢的问题,提出对基于梯度提升决策树的含光伏接入微电网调度优化。通过设定多层级目标函数,构建优化模型,并采用自适应修正处理实现调度优化。测试结果表明,与传统优化方法对比,文章设计的梯度提升决策树含光伏接入微电网调度优化方法得出的电网调度频率平均优化比相对较高,针对性也较强,在不同的背景下,对于微电网的调度优化效果明显提升,具有较强的针对性和稳定性。Aiming at the problem of fixed constraints and slow speed in conventional scheduling of microgrids with photovoltaic access,this paper proposes optimization of scheduling of microgrids with photovoltaic access based on gradient boosting decision trees.By setting multi﹣level objective functions,constructing an optimization model,and implementing scheduling optimization through adaptive correction processing.The test results show that compared with traditional optimization methods,the gradient boosting decision tree designed this time,which includes photovoltaic access to microgrid scheduling optimization method,has a relatively high average optimization ratio of grid scheduling frequency and strong targeting.In different backgrounds,the optimization effect on microgrid scheduling is significantly improved,with strong targeting and stability.

关 键 词：梯度提升决策树 含光伏背景 微电网 电网调度 调度优化 供电模式 

分 类 号：TP391.92[自动化与计算机技术—计算机应用技术]