強(qiáng)化學(xué)習(xí)算法在電力市場(chǎng)中的應(yīng)用.pdf

1、華中科技大學(xué)碩士學(xué)位論文強(qiáng)化學(xué)習(xí)算法在電力市場(chǎng)中的應(yīng)用姓名：黃金紅申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：概率論與數(shù)理統(tǒng)計(jì)指導(dǎo)教師：周曉陽20061001華中科技大學(xué)碩士學(xué)位論文IIAbstractWiththeestablishingofelectricitymarketthepowercompany’scompetitionhasincreasinglyattractedpublicattention.Intheelectricitymarkete

2、nvironmentthepowercompanies’taskchangefromthetraditionallycompletingthetaskofgeneratingtobiddingthepowerloadtoobtainmaximumprofitsitsessentialpurposebehavihaschanged.Thepowercompaniesispritodeclarethepriceofelectricityot

3、herinfmationservices.Withpowercompany’sgeneratingcapacitythespiritof“fairjustopen“theBusinessInfmationCentearrangesfthecompanies’generatingcapacity.Howtousetheinfmationreasonabllyregisteringtoobtainmaximumprofitsisthecom

4、pany’sfirstquotationstosolve.Inanalysisofthepowercompanieshowtoobtainthemaximumprofitaneffectivetoolisgamethey.Classicalgamethey’sbasicassumptionsisperfectrationalitybutinfactwhenreturningtheelectricitypricesthemanufactu

5、rerscouldnotfullygraspofmarketinfmationitisimpossibletoputalltherealityvariousfactsintoaccount.Therefethemodelbasedontheboundedrationalityismerealisticitallowspolicymakerscanbestreamlinedmisunderstinglackofcapacitymiscal

6、culationsfgetestimatingdependingonthesurfaceunrelatedtotheissueofwhatconstitutesthedetailsofthings.Thegamemodebasedonboundedrationalityismeattentiontoexplainhowpeoplelearnadaptevolutiontobalance.Thispaperapplicatespartic

7、ipants’boundedrationalityintothepowercompaniesbiddingstrategystudy.Befethepowercompanychoosesadecisiontheyhopetobegrantedaprofitusethisprofitcomparingwiththeactualprofitstoconstantlymodifythestrategy.Duetoincompleteinfma

8、tionpowercompany’saspirationprofitlevelisempiricalsubjectivewhenmanyindividualsusetheaspirationprofittostudytheinteractionofpowercompany’ssubjectivedesireswillbringwhatimpacttothemacroscopicmarket.Whichisthethemeofthispa

9、per.Fthesethepaperputsreinfcementlearningapproachbasedonaspirationprofitwhichcanbetterreflecttheirownboundedaspirationprofitapproximationtoreallevel.Thepapersimulatsthereinfcementlearningalgithmwhichbasedontheaspiration.

10、Alongwiththesimulationgeneratkeepscrectingtheirstrategiesatlastthestrategywillconvergencetoabalancedpositionwemayexperienceavarietyofmarketequilibrium.Keywds：ElectricityMarketGameTheyReinfcementLearningBoundedRationality