資源描述:
《深潛救生艇動(dòng)力定位系統(tǒng)控制方法研究》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、深潛救生艇動(dòng)力定位系統(tǒng)控制方法研究摘要隨著人們對(duì)海洋的開(kāi)發(fā)向深海進(jìn)軍,對(duì)深潛救生艇動(dòng)力定位系統(tǒng)的需求越來(lái)越大,要求也越來(lái)越高,這就促使科技工作者進(jìn)行新的技術(shù)研究和開(kāi)發(fā),以滿足在軍用領(lǐng)域和民用領(lǐng)域的需求。在國(guó)外,動(dòng)力定位技術(shù)已經(jīng)發(fā)展了三十多年,形成了高科技的產(chǎn)業(yè),促進(jìn)了相應(yīng)科學(xué)研究的發(fā)展。本文以深潛救生艇為研究對(duì)象,以提高深潛救生艇定位精度、提高深潛救生艇工作的穩(wěn)定性、增強(qiáng)動(dòng)力定位系統(tǒng)的抗干擾能力及降低能耗為出發(fā)點(diǎn),研究深潛救生艇動(dòng)力定位系統(tǒng)新的控制技術(shù)。目前CMAC神經(jīng)網(wǎng)絡(luò)在深潛救生艇操縱性能的研究中已有一些報(bào)道,還沒(méi)有在深潛救生艇動(dòng)力定位系統(tǒng)中應(yīng)用的研究。
2、本文主要進(jìn)行以下幾個(gè)方面的研究工作:利用模塊化建模的思想,對(duì)深潛救生艇動(dòng)力定位系統(tǒng)建立模型,并且建立了仿真程序,包括深潛救生艇運(yùn)動(dòng)數(shù)學(xué)模型、推力器模型和環(huán)境(海流)數(shù)學(xué)模型,通過(guò)仿真實(shí)驗(yàn)驗(yàn)證模型的正確性。研究了前饋神經(jīng)網(wǎng)絡(luò),提出基于CMAC神經(jīng)網(wǎng)絡(luò)的PID復(fù)合控制算法,并且驗(yàn)證其控制方法的有效性。建立CMAC神經(jīng)網(wǎng)絡(luò)與常規(guī)PID相結(jié)合的復(fù)合控制器。通過(guò)C++編寫(xiě)動(dòng)力定位系統(tǒng)仿真程序,對(duì)常規(guī)PID控制器、CMAC神經(jīng)網(wǎng)絡(luò)與常規(guī)PID相結(jié)合的復(fù)合控制器進(jìn)行仿真,通過(guò)仿真結(jié)果驗(yàn)證了所研究技術(shù)的合理性和實(shí)用性。本文研究了深潛救生艇動(dòng)力定位系統(tǒng)的新的控制方法,該方法可
3、以在以后的動(dòng)力定位系統(tǒng)設(shè)計(jì)中進(jìn)行應(yīng)用。關(guān)鍵詞:深潛救生艇;動(dòng)力定位;CMAC;神經(jīng)網(wǎng)絡(luò)哈爾濱一l:程人學(xué)碩卜學(xué)位論文ABSTRACTAtpresent,peopleneedincreasingforhi曲qualityvesselswithdynamicpositioning(DP)systems,withexploitingsea’Sresourcesintodeepwater.Itimpelsresearcherstostudyanddevelophi—newtechniquecontinuously,SOthatthedemandissatisfied
4、incivilandmilitaryfields.Thedynamicpositioningtechniquehasbeendevelopedformorethan30yearsandhasformedahighscientificindustrythatpromotesthecorrespondingscientificresearchesabroad.Thispaper,targettingthefour—fredom—degreedeepsubmergencerescuevehicle,studiesthenewcontroltechnologyont
5、hedeepsubmergencerescuedynamicpositiontoenhencethedeepsubmergencerescuevehicleallocationaccuracyanddeepsubmergencerescuevehicleworkingstabilityandtostrengthentheanti—interferenceabilityandthereductionofenergyconsumptionofthedynamicpositionsystem.Nowthestudyingofdeepsubmergencerescu
6、evehicleoperatingperformancehassomereportsusingCMACneuralnetwork.Thereportsthatthestudyingmethodswithmystudyingaresimilarornearhasnotbeenfoundathomeandabroaduptonow.Themainresearchingworksinthepaperaresummarizedasfollows:Makeuseofmodularizationtobuildthemodel,builtthemodelofdeepsub
7、mergencerescuevehicle。Sdynamicpositionsystem,andbuilttheemulationprocedure,includingthatthedeepsubmergencerescuevehiclemovingmathematicmodel,Propellermodelandenvironmentmathematicmodel(oceancurrent).Bysimulating,verifiesthemodelcorrectness.Firstly,studiedfeed—forwardneuralnetworkan
8、dputforwardanewalgorithmon