基于Lotka-Volterra模型的双群协同竞争粒子群优化算法.PDF

绗?5 鍗 绗?1 鏈 鎺 鍒 涓 鍐 绛 2010 骞 11 鏈 Vol. 25 No. 11 Control and Decision Nov. 2010 鏂囩珷缂栧彿: 1001-0920 (2010) 11-1619-06 鍩轰簬Lotka-Volterra 妯″瀷鐨勫弻缇ゅ崗鍚岀珵浜夌矑瀛愮兢浼樺寲绠楁硶 1 1 2 鍚村绁 , 閮疂榫 , 鐜 濞 (1. 瑗垮畨鐢靛瓙绉戞妧澶у鏈虹數宸ョ▼瀛﹂櫌锛岃タ瀹?10071 锛?. 绌哄啗宸ョ▼澶у鐢佃宸ョ▼瀛﹂櫌锛岃タ瀹?10077) 鎽 瑕? 閽堝绮掑瓙缇や紭鍖栫畻娉曟槗鍑虹幇鏃╃啛鏀舵暃鐨勯棶棰? 鎻愬嚭浜嗗熀浜嶭otka-Volterra 妯″瀷鐨勫弻缇ゅ崗鍚岀珵浜夌矑瀛愮兢浼樺寲绠 娉?LVPSO). LVPSO 绠楁硶鍊熼壌绉嶇兢鐢熸€佸涓憲鍚嶇殑Lotka-Volterra 鍙岀兢鍗忓悓绔炰簤妯″瀷, 璁ㄨ浜嗕袱绉嶇缇ゅ崗鍚岀珵浜夋柟妗? 閫氳繃缇ゅ唴鍜岀兢闂寸珵浜夊鍔犵矑瀛愮殑澶氭牱鎬? 鎻愰珮浜嗙缇ゆ憜鑴卞眬閮ㄦ瀬鍊肩殑鑳藉姏. 瀵? 涓吀鍨嬪熀鍑嗘祴璇曞嚱鏁拌繘琛屼紭鍖栧疄楠 琛ㄦ槑, LVPSO 鍦ㄦ敹鏁涢€熷害鍜屼紭鍖栫簿搴︽柟闈㈠潎鏈夎壇濂界殑琛ㄧ幇. 鍏抽敭璇岤 绮掑瓙缇や紭鍖栵紱鏃╃啛鏀舵暃锛汱otka-Volterra 妯″瀷锛涘鏍锋€ 涓浘鍒嗙被鍙耳 TP18 鏂囩尞鏍囪瘑鐮? A Lotka-Volterra model based particle swarm optimization 1 1 2 WU Xian-xiang , GUO Bao-long , WANG Juan (1. School of Electromechanical Engineering锛孹idian University 锛孹i鈥檃n 710071 锛孋hina 锛?. Telecommunication Engineering Institute 锛孉ir Force Engineering University 锛孹i鈥檃n 710077 锛孋hina. Correspondent: WU Xian-xiang, E-mail 锛歸uxianxiang@163.com) Abstract: A particle swarm optimization algorithm inspired by Lotka-Volterra model(LVPSO) is proposed to avoid the premature convergence problem. The famous Lotka-Volterra model in ecology is introduced into basic particle swarm optimization algorithm, and two different cooperative-competitive schemes are discussed. The diversity of particles are increased by intraspeci铿乧 and interspeci铿乧 competition, and the ability of particles breaking away from the local extremum is improved remarkably. The experimental results show that the proposed LVPSO algorithm can converge in higher speed and higher precision b