水泥行业清洁生产审核中的模糊数学和计算机工具运用

重庆大学硕士学位论文水泥行业清洁生产审核中的模糊数学和计算机工具运用姓名：张洁申请学位级别：硕士专业：环境工程指导教师：熊文强20051201I,IIABSTRACTCleanerproduction(CP)isaneffectiveapproachtocontrolindustrypollution;itcouldbeconsideredasthebestmodeforbotheconomicandenvironmentalbenefitsforenterprise;indeed,theproblemexistsincementplantscan’tbesolvedjustbythetraditionalpollutiondisposal.CPassessmentisakeyfactorinCPprocess,it’ssensibleforustosearchforeffective,scientificCPassessmentmethod.ItshowsthatFuzzyMathmethodcouldbeintroducedtoCleanerProduction(CP).It’smuchbetterthanthesingleindicatormethodthatpopularlyusedinCPassessments.Itcaneasilytransformtheundeterminedproblemintoasimplemathpattern,andhelptheCPspecialisttofindouttheweakfactorofcemententerprises.Meanwhile,anotherproblemexistsincleanerproductionassessmentisthatwehaven’tmuchrelatedexperienceandadequateexperts.So,seekingfortheassistantofcomputertoolsisurgent.Totheleast,theANN(artificialneuralnetwork)anddatabaseisworthtofurtherresearch.Furthermore,takingChongqingXiaoNanHaicementplantsastypicalcasestudy,usingfuzzymathmethodtodiscerntheCPopportunity,andassessthelevelofthiscementplant.Thecasestudyresultsareeduced:theabove-mentionedestimationsystemandfuzzymathmethodisconcise,feasibleandpracticable,itseemstobeoneoftheapproachestostatusanalysisanddiagnosisforcemententerprises;It’shelpfultochecktheCPleveloftechnologies,equipments,managementetc.andit’susefultoseekmorepotentialofCP.Thedatabaseconstructionismainlydesignedtosatisfytheinquiryandthetechnicalsupportfunction,theANNutilizationincludes:CPexpertdecisionsystemwithANNandMATLABANNtoolbags.ThispaperanalysisthefeasibilityandthelimitationofthetoolsinCPassessment,anddesignedtherelatedtechnicalplanontheory.Keywords:CPassessment,CementEnterprise,Fuzzymath,ANN,Database1111.11050%7[1]700080%2000200411.120002004Table1.1CementemissiondustinchinaofYear2000-2004t(t)(%)kg/t20004635597008093913.62001479266100551468.32002482272500524307.52003506886300647387.52004504297000728407.51.11090,,,,,12(),,10%80[2],,1.21.2.11.2.2181.2.3132422.11996:P2,GreenBusiness(Eco-efficiency)(WasteMinimization)[3]2.2801989,(UNEP),1992“21”,UNEP,[45],[5],,4019933000102519982003112.3Chandak[6]A.K.A.Rathi[7]one-manfirms2.42.4.1901992[3]2.4.2,,,,,262.4.3[6][6]()2715%--30%2.1,1);2),,;3);4)2.4.4[6]2.1Fig.2.1cleanerproductionassessmentscontent28“”1)2)3)4)5)6)7)2.4.5,2.2Fig.2.2cleanerproductionassessmentprocess292.5;3CPA103CPA3.13.1.1[8]3.1,(),3.1Fig.3.1Dustpointincementproduction3CPA113.1.2[8]WilhelmSchramm[9]SO23.1.323%70%4000t/d(2000t/d)2000t/d(1000t/d)[8]3.1.4“———3CPA12——”“————”[12]2/3,3,,[10]2530[13]CO2SO2NOXCO2SO2NOX3CPA133.2CPAcleanerproductionassessment,CPA,,,,CPA3.1CPATable3.1ThefunctionofCPACPA,CPA,CPACPA,3CPA143.2.1CPACPACPA3.2CPA3.2CPA3.3CPA3.2CPAFig3.2TheprincipleoftheCPAindicators3CPA153.2.2CPA3.2CPASO23.3Fig.3.3CPperformanceestimationindicatorssystemofcementcorporations3CPA163.2CPATable3.2CPAindicatorsstandardofcementcorporations1.0-0.80.8-0.60.6-0.40.4-0.20.2-0.0a.12t/d20001500120010008003100959085804t/(t)50004000300020001000b.5t/0.30.61.01.21.5690858075707kwh/t901051151251408kcal/kg7201100150018002000915%12%10%7%0%c.kg/t100.080.180.250.350.50110.050.070.100.150.20120.0160.0320.040.080.10130.82.03.04.05.0140.0030.0070.050.100.1515SO20.080.321.001.502.503CPA17d.1617181007050%30%20%3.2103.2.3CPA3.43.3CPA3.4Fig3.4Cleanerproductionperformanceestimationprocedure3CPA18CPA41944.1,,4.1.1U,,,,U={U1,U2,…,Ui}u1,u2,…,ii,V={V1,V2,…,VJ}V1,V2,…,VJUk,≥−−≤≤=)2.(....................0)21.........(121)0.........(..........1VUiVUiVvVVUiViUiRij(4.1)UV,U,()()R=ijiiijjijRRRRRRRRRRRRR..............................................................................................3.21223.22211131211(4.2)420,ij,,j,(),ViUiWi=(4.3),iUi;iV,∑==niViUiViUiWi1//(4.4),,WR,Y=W×R(4.5)WR,,Y4.1.22000/ABA2000/B1999/ABABABA1B0.99421“”4.2,,,,,,,,,,,4.2.1BPBPD.E.RumelhartJ.L.McCelland1986BPSigmoid4.2.2BPBPBP422[0,1],MATLABPrestd18Vρ,sigmoid()1(1exp())fxx=+(4.6)1)MATLABRAND5[0,1],184.15123452)(Uρ,Vρ),Uρ18Vρ3),4)Uρ0,Vρ0MATLABRAND[0,1]p=rand(18,5)1)1,2,3...........1,2,3............1,2........()aaabbbddUuuuuuuuu=r();fxx=1,minmaxmin()/()iiiuUruuuu∈=−−maxminuu4232)t=[4,3,2,4,3]3)traingdmnet=newff(minmax(p),[18,1],{'tansig','purelin'},'traingdm');4)inputWeight=net.IW{1,1}5)inputbias=net,b{1};6)p=0.12100.17080.96690.46080.41990.45080.99430.66490.45740.75370.71590.43980.87040.45070.79390.89280.34000.00990.41220.92000.27310.31420.13700.90160.84470.25480.36510.81880.00560.36780.86560.39320.43020.29740.62080.23240.59150.89030.04920.73130.80490.11970.73490.69320.19390.90840.03810.68730.65010.90480.23190.45860.34610.98300.56920.23930.86990.16600.55270.63180.04980.93420.15560.40010.23440.07840.26440.19110.19880.54880.64080.16030.42250.62520.93160.19090.87290.85600.73340.33520.84390.23790.49020.37590.65550.17390.64580.81590.00990.3919inputbias=-1.0412-5.96650.75093.30391.17780.36880.1353-0.3059-3.9057-3.7626-2.98714.48611.4425-4.02700.2596-0.15070.63760.39564247)[net,tr]=train(net,p,t);8)[0,1]518518771e-3,0.8-0.9,0.14.2.34.3,,,TR