中国电机工程学报
中國電機工程學報
중국전궤공정학보
ZHONGGUO DIANJI GONGCHENG XUEBAO
2014年
4期
587-595
,共9页
500 kV变电站%静止无功补偿器%低压电容/电抗器%多目标优化%协调控制
500 kV變電站%靜止無功補償器%低壓電容/電抗器%多目標優化%協調控製
500 kV변전참%정지무공보상기%저압전용/전항기%다목표우화%협조공제
500 kV substation%reactive power compensator%low-voltage capacitors/reactors%multi-objective optimization%coordinated control
针对目前500 kV变电站中无功补偿装置所采用的单独补偿控制方式，提出一种多目标协调控制方式来克服无功补偿装置缺乏协调且损耗较大的不足。该方法将变电站内低容/低抗装置纳入SVC的控制体系，并考虑站与站之间无功补偿装置的相互影响，以节点电压偏差和无功补偿装置总损耗最小为目标建立多目标无功协调控制模型。根据无功协调控制中变量敏感度不同、局部搜索能力不足的特点，将控制变量划分为敏感变量和非敏感变量，采用具有二级搜索的改进NSGA-II算法求取其Pareto最优解集。对南方电网中电压耦合较强的2个变电站在3种不同负荷水平下进行无功协调控制仿真，结果表明优化结果能够根据控制需要为决策者提供多种最优的协调控制策略。与常规 NSGA-II 算法和法线边界交叉算法的对比，表明改进 NSGA-II 算法得到的Pareto解集具有更优的收敛曲线及分布性。
針對目前500 kV變電站中無功補償裝置所採用的單獨補償控製方式，提齣一種多目標協調控製方式來剋服無功補償裝置缺乏協調且損耗較大的不足。該方法將變電站內低容/低抗裝置納入SVC的控製體繫，併攷慮站與站之間無功補償裝置的相互影響，以節點電壓偏差和無功補償裝置總損耗最小為目標建立多目標無功協調控製模型。根據無功協調控製中變量敏感度不同、跼部搜索能力不足的特點，將控製變量劃分為敏感變量和非敏感變量，採用具有二級搜索的改進NSGA-II算法求取其Pareto最優解集。對南方電網中電壓耦閤較彊的2箇變電站在3種不同負荷水平下進行無功協調控製倣真，結果錶明優化結果能夠根據控製需要為決策者提供多種最優的協調控製策略。與常規 NSGA-II 算法和法線邊界交扠算法的對比，錶明改進 NSGA-II 算法得到的Pareto解集具有更優的收斂麯線及分佈性。
침대목전500 kV변전참중무공보상장치소채용적단독보상공제방식，제출일충다목표협조공제방식래극복무공보상장치결핍협조차손모교대적불족。해방법장변전참내저용/저항장치납입SVC적공제체계，병고필참여참지간무공보상장치적상호영향，이절점전압편차화무공보상장치총손모최소위목표건립다목표무공협조공제모형。근거무공협조공제중변량민감도불동、국부수색능력불족적특점，장공제변량화분위민감변량화비민감변량，채용구유이급수색적개진NSGA-II산법구취기Pareto최우해집。대남방전망중전압우합교강적2개변전참재3충불동부하수평하진행무공협조공제방진，결과표명우화결과능구근거공제수요위결책자제공다충최우적협조공제책략。여상규 NSGA-II 산법화법선변계교차산법적대비，표명개진 NSGA-II 산법득도적Pareto해집구유경우적수렴곡선급분포성。
Being different from the separate compensation control mode used widely in 500 kV transformer substations currently, the multi-objective reactive power coordination control model was presented in this paper to overcome the drawback that reactive power compensation devices lack in coordination and have high active power loss. In this model, the minimum bus voltage deviation and total loss of reactive power compensation devices were taken as objective functions by bringing the lower-voltage reactors and capacitors into the SVC control system and considering the reactive power compensation devices interaction between different transformer substations. Due to the different sensitivity of control variables in control model and lack of the local search ability, control variables are divided into sensitive variables and non-sensitive variables, and then, the improved NSGA-II algorithm with secondary search ability was used to search the Pareto optimal solution set. Two substations in large power grid with strong voltage coupling were coordinated in different load level, the results obtained by proposed NSGA-II algorithm can provide a variety of optimal control strategies. Compared to the conventional NSGA-II algorithm and normal boundary intersection method, the improved NSGA-II algorithm has better convergence curve and distribution of the Pareto solution sets.
