中国电机工程学报
中國電機工程學報
중국전궤공정학보
ZHONGGUO DIANJI GONGCHENG XUEBAO
2014年
25期
4250-4258
,共9页
微网%可时移负荷%优化配置%混沌自由搜索算法
微網%可時移負荷%優化配置%混沌自由搜索算法
미망%가시이부하%우화배치%혼돈자유수색산법
microgrid%time-shifting load%optimal configuration%chaotic free search (CFS) algorithm
优化配置分布式电源容量,提高供电可靠性和经济性是微网规划的关键问题。文中在海岛微网规划中加入海水淡化类可时移负荷,以便更好地跟踪可再生能源发电,并解决海岛淡水需求困难。根据海岛用户年负荷需求和供电可靠性要求,结合不同岛屿气象条件、地理位置情况、海水淡化需求,构建风/光/储/柴等多类电源组合,采用混沌自由搜索算法,得到海岛微网分布式电源容量的最优配置方案,并用Matlab 编程实现。结果表明,海水淡化类可时移负荷的加入可降低海岛微网优化配置方案的投资冗余度,使资源利用最大化。在算法收敛特性及最优解求解两个指标中,与传统优化算法进行对比,证明了混沌自由搜索算法在快速收敛特性和全局寻优方面具有明显的优势。
優化配置分佈式電源容量,提高供電可靠性和經濟性是微網規劃的關鍵問題。文中在海島微網規劃中加入海水淡化類可時移負荷,以便更好地跟蹤可再生能源髮電,併解決海島淡水需求睏難。根據海島用戶年負荷需求和供電可靠性要求,結閤不同島嶼氣象條件、地理位置情況、海水淡化需求,構建風/光/儲/柴等多類電源組閤,採用混沌自由搜索算法,得到海島微網分佈式電源容量的最優配置方案,併用Matlab 編程實現。結果錶明,海水淡化類可時移負荷的加入可降低海島微網優化配置方案的投資冗餘度,使資源利用最大化。在算法收斂特性及最優解求解兩箇指標中,與傳統優化算法進行對比,證明瞭混沌自由搜索算法在快速收斂特性和全跼尋優方麵具有明顯的優勢。
우화배치분포식전원용량,제고공전가고성화경제성시미망규화적관건문제。문중재해도미망규화중가입해수담화류가시이부하,이편경호지근종가재생능원발전,병해결해도담수수구곤난。근거해도용호년부하수구화공전가고성요구,결합불동도서기상조건、지리위치정황、해수담화수구,구건풍/광/저/시등다류전원조합,채용혼돈자유수색산법,득도해도미망분포식전원용량적최우배치방안,병용Matlab 편정실현。결과표명,해수담화류가시이부하적가입가강저해도미망우화배치방안적투자용여도,사자원이용최대화。재산법수렴특성급최우해구해량개지표중,여전통우화산법진행대비,증명료혼돈자유수색산법재쾌속수렴특성화전국심우방면구유명현적우세。
ABSTRACT:The optimal sizing method for distributed resource to improve the power supply reliability and economy efficiency is a critical issue for an islanded microgrid. This paper presented the time-shifting desalination load in an island microgrid to track renewable energy generation better and provide fresh water. A novel chaotic free search (CFS) algorithm to solve for optimal sizing problem of distributed resource was proposed. According to the annual load and power supply reliability, combined with the annual meteorological conditions and residential water consumption, the optimal sizing problem of an island microgrid was resolved by Matlab programming. The optimal sizing result showed that considering time-shifting desalination load can reduce redundant investment to maximize resource utilization. Compared with traditional optimization algorithms, the simulation results verified that the proposed algorithm had the obvious superiority in the fast convergence property and the global optimization.