資源描述:
《自主式水下機(jī)器人推進(jìn)器及舵狀態(tài)監(jiān)測(cè)研究》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(kù)。
1、墮壘型型筌蘭墜:盟薹耋盜一。。一;。一;一;摘要自主式水下機(jī)器人(AUV)的研究是我國(guó)海上國(guó)防與海洋"發(fā)戰(zhàn)略的重要組成部分。為了保證AUV在研制丌發(fā)、設(shè)計(jì)制造中高效率、高可靠性,以及在水下實(shí)驗(yàn)和作戰(zhàn)使用完成運(yùn)動(dòng)控制、作戰(zhàn)任務(wù),并且能夠安全回收,必須對(duì)其系統(tǒng)各部件的工作狀態(tài)進(jìn)行監(jiān)測(cè)。本文通過對(duì)水F機(jī)器人的控制系統(tǒng)中執(zhí)行機(jī)構(gòu),運(yùn)動(dòng)模型,傳感器系統(tǒng)的分析,提出了基于智能方法的以診斷模型為核心的監(jiān)測(cè)系統(tǒng)結(jié)構(gòu),J宅i耍建立了自主式水下機(jī)器人的各推進(jìn)器、方向舵的滲斷模型,通過塒在巡航和搜索兩種航行方式下的工作狀況進(jìn)行監(jiān)測(cè).來診斷各部件是否出現(xiàn)故障,保證水下
2、機(jī)器人安全航行。在診斷模塊建模中采用模糊邏輯與神經(jīng)網(wǎng)絡(luò)結(jié)合的技術(shù),以模糊自適應(yīng)學(xué)習(xí)控制網(wǎng)絡(luò)為核一IL,,提出了一種簡(jiǎn)單可行的基于盛大權(quán)值矩陣的規(guī)則提取及基于浮點(diǎn)數(shù)編碼的遺傳算法的參數(shù)調(diào)整的,完善網(wǎng)絡(luò)結(jié)構(gòu)與性能的方法,并在狀念監(jiān)測(cè)過程中,通過對(duì)由控制器輸入的水下機(jī)器人運(yùn)動(dòng)控制量以及運(yùn)行狀態(tài)的模糊推理,得到執(zhí)行部件(推進(jìn)器或舵)的工作狀態(tài)優(yōu)劣程度,為保證水下機(jī)器人完成任務(wù)。安全返回提供控制依據(jù)。通過計(jì)算機(jī)仿真試驗(yàn),證明了該方法的可行性和監(jiān)測(cè)系統(tǒng)的有效性。關(guān)鍵詞:水F+,tL器人;狀態(tài)監(jiān)測(cè);神經(jīng)模糊網(wǎng)絡(luò);規(guī)則提取哈爾濱1‘群人學(xué)碩十學(xué)位論文Abstr
3、actTheresearchoftheautonomousunderwatervehicleisoneofthemostimportantpansofthenavalnationaldefenceandocealldevelopmentstratagem.Toguaranteethehigherefficiencyandreliabilityintheresearch.development,designandmanufacture,andfinishthemotioncontrol,missionandsafelyhomingintheund
4、erwaterexperimentsandactuaIcombat.itiSnecessarytomonitorthe,conditionofsomecomponentsintheAUVsystem.Inthisarticle,accordingtotheanalysisoftheexecutionmachine,motionmodelandsensorsystemintheAUv,astructureofconditionmonitoringsystemfocusedonthediagnosismoduleandbasedontheintel
5、ligentmethodispresented,whichmainlyconstructsthemodelsofthepropellersandruddersrespectively,andmonitorstherunningconditionoftheAUVincruisingandhunting,anddiagnoseswhethertherearesomefaultsoccurringinthepartsthatbeingmonitored,andensurethenavigationsecurity.Intheconstructingo
6、fthediagnosismoduleusingthetechnologyofthecombinationofthefuzzylogicandneuralnetwork.whichbasedonthefuzzyadaptivelearningcontrolnetwork,asimplekindofcapablemethodforconsummatethestructureandperformanceofnetworkisintroduced,whichincludestherulesextractionbasedonthemaximumweig
7、htsmatrixandtheparametersamendmentbasedongeneticalgorithmbyfloating—pointcoding.Duringthemonitoringofthepartscondition,theoutputoftheconditionmonitoringsystemshowsthegoodworkingconditionoftheexecutingagencybyfuzzilydeducingfromtheeontr01instructionsendbytheAUV‘Scontrollerand
8、motionstatus。andSOofferstheprooftocompletemissionandretumsafely.111eresulto