粒子群算法的研究与应用-计算力学专业论文.docxVIP

Abstract Abstract In the pr.oblem .of allocating resources，.optimizati.on means 阳 meet 也eωm往回nts .ofh.ow ωall.ocate resources neededωmake the systemωst 由.e leωt， acωss 阳由e m.ost econ.omic benefi臼.In由is paper ec.on.omic dispatch pr.oblem is a typiω1 class，according ωthe number .of difIerent .optin血ati.on .0均ectives川le pr.oblem can be divided inωtw.o typesωmposed .of single .0时ective and multi咄jective econ.omic dispatch pr.oblem. Chaos has the 讪itial value sensi咀vity，rand.omness，erg.odicity and .0吐ler characteristics. As a new research meth.od，it has been success践Illy ∞mbined wi由a variety .of alg.orithms wi由 its excellent characteristics in 由is paper. 咀山 combinati.on is mainly 也.e use .of the erg.odic chaotic，rand.omly search a be忧町 perf.ormanlω 叫伽M 伽 sta臼.Di能rent searching me由.01也 have ditIerent possibi1ityωfind the be忧.er perf.ormance. n巾p叩町 impr.oves a Tent mapping ωmake the chaotic sequenωis m.ore unif.orm. Then， a new chaotic 脱即ching .of linear decreasing the radius ，by comparing the experimen阳1 results sh.ow the new way has high町 e:fficient 由an traditi.onalωrri町. As a heuristic，Particle swarm .optimizati .on alg.ori伽m is a kind .of intelligent alg.orithms simulated the social behavi.or .of bird .or fish. As an .optimizati .on 怡。1，PSO has been widely applied ω 也.e single-objective and multi-objective .optimizati .on pr.oblem. In由e multi-.objective .optimizati .on process，Traditi.onal particle swarm is easily to have the phen.omen.on .of stagnati.on during the later 到age .of ev.oluti.on.In the same time，P.opulati.on may fall inωl.ocal .optimum.Also In the multi咱bjective .optimizati .on 严。ωss，P.opulati.on may fall int.o l.ocal .optimum ω.o.The paperωme .ovl町由.e bugs .over 也r.ough integrati.on .of the new chaotic search meth.od. In additi.on， in multi咱bjective .optimizati .on prlωess，using .of 1.0ωl chaotic search meth.od can alωenhance 由.e diversity and ωnvergence .of soluti.ons in 也is pr.oωss.Finally，Applicatingω 也e single-.objective .of 饵.on.om