2813

2729

كارشناسي ارشد

شيمي تجزيه

اصفهان: دانشگاه صنعتي اصفهان، دانشكده شيمي

1383

دوازده، 86، [I]ص.: مصور، جدول، شكل، نمودار

تقي خياميان

جهانبخش قاسمي

محمد كاظم اميني، حسن قاضي عسگر

1396/10/26

كتابنامه

شيمي

شيمي

ID2729

به فارسي :قابل رويت در قسمت نسخه ديجيتالي

Abstract In the first step of this thesis a QSPR study is performed to develop a model relating the structure of 94 Gemini surfactantsto their critical micelle concentrationsforthe first time A whole number of descriptorswere calculated with Dragon software and asubsetof calculated descriptors was selectedfrom 18classesof Dragon descriptors with aforward stepwise MLR method The data set was randomly divided into three data sets calibration 69 molecules prediction 15 molecules and tl st set 10 molecules Theselected descriptors were used as input neurons in a wavelet neural network WNN model The WNN architecture and its parameters vere optimized simultaneously Theperformance of the model was investigated by the test set and the averageabsolute errorwas 0 105 for the test set The capability of the model as compared with the MLRmodel It was demonstratedthat the WNN is superiorto the MLR model In the second stepof this thesis the acidity constantsof the PAR were determined byprincipal componentanalysis artificial neural net orks using simulated and spectraldata Triprotic acid massbalance equations and correspondingsp ctral profiles generatedby aGaussian model were used to simulate all required absorbance pH data A triangularexperimental design was used to select and produce the different simulated acidityconstants The effects of white noise at different levels were also studied to check theprediction ability of the model A fully experimental data set photometric titration dataof PAR at pH l 5 13 range was used as a test set The obtained acidity constantsare in agood agreementwith previously reported values using DA TAN software d 0 II 0 o w I JJj 1f IA Jt Aq

تقي خياميان

جهانبخش قاسمي

محمد كاظم اميني، حسن قاضي عسگر