【工程方案】基于人工神经网络技术的注塑成型工艺参数优化

:2001-07-27:(59975032):(1977),,:1001-4934(2001)06-0001-041,2,2,1,1(11,430074;21,266101):,,,CAEHsCAE3DRF,:;CAE;;:TQ320166+2:AAbstract:TheapplicationoftheANN(ArtificialNeuralNetwork)technologyinoptimizationforinjectionmoldprocessparametersisstudied.Byconstructingtheinjectionpressureandtempera2turemodelbasedonANNmethod,optimizationofprocessparameterisrealized.ThismethodhasbeenappliedininjectionmoldCAEsystemdevelopedbyHUSTStateKeyLabofMold&DieTechnologyandprovedtomakegreateffect.Keywords:artificialneuralnetwork;injectionmoldCAE;processparameter;optimization0,,CAE,[1],,CAECAE,,,,CAE(),,1,,(),,,[2],:,,:F(P,Tmax,Tmin)=w1P+w2(Tmax-Tmin)+f1(Tmax)+f2(Tmin)(1)F120011No.6PTmaxTminf1Tmax,;,f2Tmin,;,w1,w2,,,2211(ArtificialNeuralNetwork,ANN),,,[3]BP(Back-PropagationNetwork,BP),,,BP212,(),,,,,,[4][5],:P=L05p5ldl(2)P5p5lL5p5lSS=b0z2dz(3)b=01+01-n(4)0n0=BeTb/Tep(5)TBTbp,,,112DieandMouldTechnologyNo.62001CAE,,,,,,22213,HsCAE3DRF,,600,160214MatLab5.2,,3,BP,375mm50mm16mm,ABS,230:4,,5,3BP4-,U,320011No.654,6,()(),,,,CAE65CAE,,,,,HsCAE3DRF,CAECAE:[1]L.T.Manzinone.ApplicationofCAEinInjectionMold2ing[M].Munich:HanserPublishers,1987.[2]1[M]1:,19961[3]1MATLAB[M]1:,19981[4]TatianaPetrova,DavidKazmer.HybridNeuralModelsforPressureControlinInjectionMolding[J].AdvancesinPolymerTechnology,1999,18(1):19311[5]1[D]1:,200114DieandMouldTechnologyNo.62001