摘要啤酒工业是国家社会经济发展的重要产业部门，啤酒企业耗能高是制约企业发展的主要因素。提高资源能源利用效率，实现环保与经济效益的双赢，是啤酒企业实现可持续发展的必由之路。在啤酒生产过程中，寻找并应用节能降耗的方法和技术，不仅可以使企业自身受益，而且对社会的可持续发展将具有重要的作用。因此对啤酒酿造车间的能耗分析对于探求节能技术非常有必要。本文首先是数据的预处理，数据的预处理主要除去无效的数据并且对原始数据逐行作差，然后采用主成分分析方法，对原始数据降维。可以有效的找出数据中最“主要”的元素和结构，去除噪音和冗余，将原有的复杂数据降维，揭示隐藏在复杂数据背后的简单结构。最大程度的保持了原有数据的信息。接着对经过降维处理的数据采用聚类分析的方法进行分类。在聚类分析之后采用模糊方法来对每一类进行建模，这里采用MATLAB模糊逻辑工具箱建立模糊模型，并用SIMULINK验证模型。该方法具有简便、直观、高效的特点。关键词：主成分分析；降维；聚类分析；模糊建模；能耗AbstractBeerindustryisveryimportantinthenationalsocio-economicdevelopmentsectors,highenergyconsumptionofbeercompaniesarethemainfactorsrestrictingthedevelopmentofenterprises.Improvingenergyefficiency,environmentalandeconomicbenefitstoachieveawin-win,isthebeerbusinesstheonlywaytoachievesustainabledevelopment.Inthebeerproductionprocess,tofindandapplyenergysavingmethodsandtechnology,notonlytobenefittheenterprisesthemselves,butalsosustainabledevelopmentofsocietywillhaveanimportantrole.Thereforetheenergyconsumptionofbeerbrewingworkshoptoexploreenergy-savingtechnologiesisverynecessary.Firstly,wecanuseprincipalcomponentanalysismethodtoreducetheoriginaldata.Thiscaneffectivelyfindthedatawhichisthemost"major"elementsandstructures,toremovenoiseandredundancy,thecomplexityoftheoriginaldatadimension,revealinghiddenbehindthesimpleandcomplexdatastructures.Maximummaintainedtheoriginaldata.Thenafterreducingthedimensionofthedataherehadusedclusteranalysismethodsareclassified.Herethedatawerenormalizedafternormalizationtreatment.Afterusingoffuzzyclusteringanalysismethodtomodeleachclass,thenwecanusefuzzylogictoolboxwithMATLABfuzzymodelandusingSIMULINKtoverifythemodel.Themethodissimple,intuitiveandefficientfeatures.KeyWords：PCA;Dimensionreduction;clusteranalysis;FuzzyModeling;energyconsumption目录摘要........................................................................................................................1Abstract...........................................................................................................................2引言........................................................................................................................11绪论.............................................................................................................................21.1课题的研究背景及意义...................................................................................21.2国内外研究现状...............................................................................................21.3研究内容..........................................................................................................22原始数据的预处理....................................................................................................42.1数据预处理的意义...........................................................................................42.2数据预处理的主要方法....................................................................................
