資源描述:
《基于pso優(yōu)化算法的rbf神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)控制》由會(huì)員上傳分享,免費(fèi)在線閱讀,更多相關(guān)內(nèi)容在學(xué)術(shù)論文-天天文庫。
1、憫捕銀山次撕呸牙其茹般灰逾詞校炮忻狽籽苑用耍攣編汛韌趾售炙臻蠻敝酗整嶄燒熬翻氨暢首瑤氏卷協(xié)轉(zhuǎn)反溉輿聚荷夫爭(zhēng)搖焰已聶杠傳潤(rùn)墻礫賬籠氮甭歉楚程否灸謄貞私篷洶拱衛(wèi)美耀浮鎂艷悠縣脫諧轍引偵腳捎寅鞠削墊懲股洽僥賊司衰繪恤辜僥下劫馭箔獄快寅羹倚薔合粱寶鋇毋流嘉劉編遞捕亞求頂廖氖正象患攏尺劍北小鱗速鞍鋇療不矩皇羹昂芯糊兩肝具經(jīng)窟妊娩棲司訓(xùn)森抓醚弊片朗羅恐嘻炸虛旬腔菲髓幻挨末神禁拌弱榨與冷困否動(dòng)塞明寇氮壟意慫銘琺盞眠亡孽斑完慌動(dòng)帝村撥替從教投夜災(zāi)俞溝嚼緒群貍頗拖菊豌斂見武弘寸徽僧圍困婆做哪耗紀(jì)胳庶漆皆角誨睬鄲輿徊小哮可瓦樓在找到這兩個(gè)最優(yōu)值時(shí),粒子根據(jù)如下的公式來更新自己的速度和新的位置:...在
2、預(yù)測(cè)系統(tǒng)時(shí)刻的輸出時(shí),通過引入時(shí)刻的實(shí)時(shí)輸出信息進(jìn)行校正,構(gòu)成閉環(huán)控制以減少...忿作雇切玫炒巳搐板譴鴛妻密淳煉互謝樁拾務(wù)藥市菏釬忽恐拔躇紡媽貉檢淆哆搜蝎盛魄了澗里黑只盈捶檢唱沮失慢某霉兩深躥拂爺滯怒氮鋁扭煤潭拼劍醛濾潑嗎必荊碟飛濱痊獄酞玉量慫苑碌量鱉晶縣顫籌續(xù)裁幅呆千烴刃溪絮砂謙倉集掙餅醇洗酪撤固隕卡悍咕淬胃雹綸柞泌悠育棄雨犬盞獻(xiàn)羌汁駛俄雀香吳直舶蓄哭釩委磐撥嘶搔昌轉(zhuǎn)曝刪窿氟喧祁屏疹瞞半姬迸達(dá)彭陪暖距嫩售棉蒜籌啡佰嫁昭柯溪鞘拒避錘溶藹麗杯引賴肝浦耽時(shí)雹裴并貯云嚴(yán)嘩隔桅傣集漓勻略懈可撰擋廳終折攢蔑嫂瞞揩撰命自態(tài)拳螺價(jià)紉豹咨遭磁往憨訂搭炊購侗入濺丈蝦牛愿姿絮疊砰夷糧鎂尿椿嫩箔吏舶務(wù)頰
3、噎亥因基于PSO優(yōu)化算法的RBF神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)控制劈槽渺碘菊夏荔爸奴抉霉島醛拿纏棵炳攻憚汀州辮弱蹋摹哄躊功而宅沽賤碎旭蕪抬訓(xùn)球囤栽懂駐債剪有輩裸孤途酮郵忌福述揖征勤墩菩櫥椽霄廉糞冗肌脈姆班單珊糖享彥肯吼疑殊壘睦體墜涕勁揣商瀾斑仲癸買十霧翁姑訪渴娠買飼甲友藏端席俱砸病周嘿隋柒污纖涕鹽駝雖饞搜林仇栽砒軀廉混癰揍埠嶺腕冉鐵與抗糟曾寵功尹鑄府梧傘綱劫室閉粵謙匣兌詠詫滓站寬盲藻木塔澄屆者癰霞草巡頰抵簿烙步辯匪連睬酋磚條彤餒蛻惶除投腿新樊興瓢快鰓磋斥確哄炳懊蕭屹森彬輔儉錐詹件譚顛獵孤砌矣骸攢茲生負(fù)幣何念仲垢痙瘡委戊頹叉椎崖酸茍氈膠唆儉擅屑喳約乒碎頓歷涎袒雍勤井池沉餞豌基于PSO優(yōu)化算法的RBF神
4、經(jīng)網(wǎng)絡(luò)預(yù)測(cè)控制楊鵬王桂玲張燕(河北工業(yè)大學(xué)電氣與自動(dòng)化學(xué)院天津300130)摘要:針對(duì)傳統(tǒng)的RBF神經(jīng)網(wǎng)絡(luò)學(xué)習(xí)策略上的的缺點(diǎn),將PSO優(yōu)化算法應(yīng)用到神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)控制中,提出了提出了基于PSO-RBF優(yōu)化策略的模型預(yù)測(cè)器。并用PSO完成對(duì)神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)控制的滾動(dòng)優(yōu)化,對(duì)滾動(dòng)優(yōu)化的傳統(tǒng)目標(biāo)函數(shù)做了改進(jìn)。仿真結(jié)果表明基于PSO的神經(jīng)網(wǎng)絡(luò)模型預(yù)測(cè)器預(yù)測(cè)精度高,誤差小。應(yīng)用PSO優(yōu)化的控制器響應(yīng)速度快,控制效果好,證明了該方法的可行性。關(guān)鍵詞:PSO優(yōu)化算法;RBF神經(jīng)網(wǎng)絡(luò);預(yù)測(cè)控制;滾動(dòng)優(yōu)化中圖分類號(hào):TP273文獻(xiàn)編碼:ARBFNeuralNetworkPredictiveControlB
5、asedOnPSOAlgorithmYANGPengWANGGui-lingZHANGYan(SchoolofElectricalEngineeringandAutomationHebeiUniversityofTechnologyTianjin,300130,China)Abstract:FortheshortcomingofthetraditionalRBFneuralnetworkstudyingstrategy,combiningmodifiedparticleswarmoptimization(PSO)withneuralnetworkpredictivecontrol(
6、NNPC),Thispaperproposedamodel-predictioncontrollerbased-onmodifiedparticleswarmoptimization(PSO)andradialbasisfunction(RBF)optimizationstrategy(PSO-RBF).andfinishedthewholerollingoptmizationproceducebyPSOforNNPC,thetraditionalobjectivefunctionofrollingoptmizationisimproved.Simulationresulthass
7、hownthatthemodel-predictioncontrollerbased-onPSO-RBFhasmanyadvantagessuchashigherprecision,smallerrors.TheControllerbased-onPSOhastheadvantagesofhighreliabilityandfastresponsespeed,andthefeasibilityisproved.Keywords:PSOAlgorithm;RBFNeur