Molecular Design of Fuel Additives

This chapter presents a case study that focuses on fuel-additive that falls under the category of an engineering material. This study discusses computer aided molecular design (CAMD) of fuel-additives. Fuel-additives are one example in the ever-growing class of engineering materials. The idea behind the approach presented here is the identification and modeling of design-relevant building blocks. These structural (and functional) elements tie together the forward and inverse problems. They reflect in a transparent fashion the design process undertaken by an expert formulator in the area. Fuel-additives are a class of performance modifiers that are added to gasoline to enhance certain properties and/or to provide additional properties not present in the gasoline. Fuel additives are used as combustion modifiers, antioxidants, corrosion inhibitors, and deposit control detergents. The hybrid model for intake-valve deposits (IVD) prediction was a black-box model, because of the use of a neural network. The essential idea and indeed the power behind this computer-aided molecular design (CAMD) approach is the identification of the most sensitive set of design/decision variables based on a first-principles understanding the way the structure relates to performance.

[1]  Robert L. Mason,et al.  SwRI-BMW N.A. Intake Valve Deposit Test - A Statistical Review , 1992 .

[2]  Paul I. Lacey,et al.  A Laboratory-Scale Test to Predict Intake Valve Deposits , 1997 .

[3]  Venkat Venkatasubramanian,et al.  Design of Fuel Additives Using Neural Networks and Evolutionary Algorithms , 2001 .

[4]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[5]  B. Schenker,et al.  Cross-validated structure selection for neural networks , 1996 .

[6]  Allan F. M. Barton,et al.  CRC Handbook of solubility parameters and other cohesion parameters , 1983 .

[7]  Thomas J. McAvoy,et al.  Nonlinear PLS Modeling Using Neural Networks , 1992 .

[8]  Gautam Kalghatgi,et al.  Deposits in Gasoline Engines - A Literature Review , 1990 .

[9]  Antonis C. Kokossis,et al.  Molecular design synthesis using stochastic optimisation as a tool for scoping and screening , 1998 .

[10]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[11]  Igor I. Baskin,et al.  Inverse problem in QSAR/QSPR studies for the case of topological indexes characterizing molecular shape (Kier indices) , 1993, J. Chem. Inf. Comput. Sci..

[12]  B. Kowalski,et al.  Partial least-squares regression: a tutorial , 1986 .

[13]  Rafiqul Gani,et al.  Computer aided molecular and mixture design with specified property constraints , 1993 .

[14]  Teri A. Crosby,et al.  The Impact of Intake Valve Deposits on Exhaust Emissions , 1992 .

[15]  Luke E. K. Achenie,et al.  Novel Mathematical Programming Model for Computer Aided Molecular Design , 1996 .

[16]  Venkat Venkatasubramanian,et al.  Computer-Aided Molecular Design Using Neural Networks and Genetic Algorithms , 1996 .

[17]  C. Maranas.  Optimal Computer-Aided Molecular Design: A Polymer Design Case Study , 1996 .

[18]  Venkat Venkatasubramanian,et al.  Computer-aided molecular design using genetic algorithms , 1994 .

[19]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[20]  R. Gani,et al.  A group contribution approach to computer‐aided molecular design , 1991 .

[21]  Venkat Venkatasubramanian,et al.  Genetic Algorithmic Approach for Computer-Aided Molecular Design , 1995 .

[22]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[23]  Elizabeth A. Schiferl,et al.  Variability of intake valve deposit measurements in the BMW vehicle intake valve deposit test , 1997 .

[24]  Lars Renberg,et al.  Non‐linear modelling with a coupled neural network — PLS regression system , 1996 .