Martin Holeňa
Lukáš Bajer
ABSTRACT
The paper presents a case study in an industrially important application domain the optimization of catalytic materials. Though evolutionary algorithms are the by far most frequent approach to optimization tasks in that domain, they are challenged by mixing continuous and discrete variables, and especially by a large number of constraints. The paper describes the various kinds of encountered constraints, and explains constraint handling in GENACAT, one of evolutionary optimization systems developed specifically for catalyst optimization. In particular, it is shown that the interplay between cardinality constraints and linear equality and inequality constraints allows GENACAT to efficienlty determine the set of feasible solutions, and to split the original optimization task into a sequence of discrete and continuous optimization. Finally, the genetic operations employed in the discrete optimization are sketched, among which crossover is based on an assumption about the importance of the choice of sets of continuous variables in the cardinality constraints.
- M. Baerns and M. Holeňa. Combinatorial Development of Solid Catalytic Materials. Design of High-Throughput Experiments, Data Analysis, Data Mining. World Scientific, Singapore, 2009. Google Scholar
Digital Library
- L. Baumes, A. Blanché, P. Serna, A. Tchougang, N. Lachiche, P. Collet, and A. Corma. Using genetic programming for advanced performance assessment of industrially relevant heterogeneous catalysts. Materials and Manufacturing Processes, 24:282--292, 2009.Google ScholarCross Ref
- L. Baumes, P. Jouve, D. Farrusseng, M. Lengliz, N. Nicoloyannis, and C. Mirodatos. Dynamic control of the browsing-exploitation ratio for iterative optimisations. In Knowledge-Based Intelligent Information and Engineering Systems, pages 265--270. Springer Verlag, Berlin, 2003.Google Scholar
- C. Breuer, M. Lucas, F. Schuetze, and P. Claus. Implementation of the multi-channel monolith reactor in an optimisation procedure for heterogeneous oxidation catalysts based on genetic algorithms. Combinatorial Chemistry and High Throughput Screening, 10:59--70, 2007.Google ScholarCross Ref
- S. Cheng, H. Chen, H. Chang, C. Chang, and Y. Chen. Multi-objective optimization for two catalytic membrane reactors - methanol synthesis and hydrogen production. Chemical Engineering Science, 63:1428--1437, 2008.Google ScholarCross Ref
- F. Clerc, M. Lengliz, D. Farrusseng, C. Mirodatos, S. Pereira, and R. Rakotomata. Library design using genetic algorithms for catalysts discovery and optimization. Review of Scientific Instruments, 76:062208, 2005.Google ScholarCross Ref
- A. Corma, J. Serra, and A. Chica. Discovery of new paraffin isomerization catalysts based on SO4
2-/ZrO2 and WOx/ZrO2 applying combinatorial techniques. Catalysis Today, 81:495--506, 2003.Google ScholarCross Ref
- A. Corma, J. Serra, P. Serna, S. Valero, E. Argente, and V. Botti. Optimisation of olefin epoxidation catalysts with the application of high-throughput and genetic algorithms assisted by artificial neural networks (softcomputing techniques). Journal of Catalysis, 229:513--524, 2005.Google ScholarCross Ref
- O. Gobin and F. Schüth. On the suitability of different representations of solid catalysts for combinatoral library design by genetic algorithms. Journal of Combinatorial Chemistry, 10:835--846, 2008.Google ScholarCross Ref
- G. Grubert, E. Kondratenko, S. Kolf, M. Baerns, P. van Geem, and R. Parton. Fundamental insights into the oxydative dehydrogenation of ethane to ethylene over catalytic materials discovered by an evolutionary approach. Catalysis Today, 81:337--345, 2003.Google ScholarCross Ref
- M. Holeňa and M. Baerns. Computer-aided strategies for catalyst development. In G. Ertl, H. Knözinger, F. Schüth, and J. Eitkamp, editors, Handbook of Heterogeneous Catalysis, pages 66--81. Wiley-VCH, Weinheim, 2008.Google Scholar
- M. Holeňa, T. Cukic, U. Rodemerck, and D. Linke. Optimization of catalysts using specific, description based genetic algorithms. Journal of Chemical Information and Modeling, 48:274--282, 2008.Google ScholarCross Ref
- M. Holeňna, D. Linke, and U. Rodemerck. Evolutionary optimization of catalysts assisted by neural-network learning. In Simulated Evolution and Learning. Lecture Notes in Computer Science 6457, pages 220--229. Springer Verlag, Berlin, 2010. Google Scholar
- M. Holeňa, D. Linke, and U. Rodemerck. Generator approach to evolutionary optimization of catalysts and its integration with surrogate modeling. Catalysis Today, 159:84--95, 2011.Google ScholarCross Ref
- K. Keyvanloo and J. Towfighi. Comparing the catalytic performances of mixed molybdenum with cerium and lanthanide oxides supported on hzsm-5 by multiobjective optimization of catalyst compositions using nondominated sorting genetic algorithm. Journal of Analytical and Applied Pyrolysis, 88:140--148, 2010.Google ScholarCross Ref
- G. Kirsten and W. Maier. Strategies for the discovery of new catalysts with combinatorial chemistry. Applied Surface Science, 223:87--101, 2004.Google ScholarCross Ref
- The Mathworks, Inc. Global Optimization Toolbox 3, 2010.Google Scholar
- A. McLeod, M. Johnston, and G. L. F. Development of a genetic algorithm for molecular scale catalyst design. Journal of Catalysis, 167:279--285, 1997.Google ScholarCross Ref
- A. McLeod and G. L. F. Heterogeneous catalyst design using stochastic optimization algorithms. Journal of Chemical Information and Computer Science, 40:981--987, 2000.Google ScholarCross Ref
- A. Ohrenberg, C. Törne, A. Schuppert, and B. Knab. Application of data mining and evolutionary optimization in catalyst discovery and high-throughput experimentation - techniques, strategies, and software. QSAR and Combinatorial Science, 24:29--37, 2005.Google ScholarCross Ref
- R. Pereira, F. Clerc, D. Farrusseng, J. Waal, and T. Maschmeyer. Effect of genetic algorithm parameters on the optimization of heterogeneous catalysts. QSAR and Combinatorial Science, 24:45--57, 2005.Google ScholarCross Ref
- U. Rodemerck, M. Baerns, and M. Holeňa. Application of a genetic algorithm and a neural network for the discovery and optimization of new solid catalytic materials. Applied Surface Science, 223:168--174, 2004.Google ScholarCross Ref
- J. Serra, L. Baumes, M. Moliner, P. Serna, and A. Corma. Zeolite synthesis modelling with support vector machines: a combinatorial approach. Combinatorial Chemistry and High Throughput Screening, 10:13--24, 2007.Google ScholarCross Ref
- J. Serra, A. Chica, and A. Corma. Development of a low temperature light paraffin isomerization catalysts with improved resistance to water and sulphur by combinatorial methods. Applied Catalysis A: General, 239:35--42, 2003.Google ScholarCross Ref
- Y. Watanabe, Umegaki, T., Hashimoto, M., Omata, K., and M. Yamada. Optimization of Cu oxide catalysts for methanol synthesis by combinatorial tools using 96 well microplates, artificial neural network and genetic algorithm. Catalysis Today, 89:455--464, 2004.Google ScholarCross Ref
- D. Wolf, O. Buyevskaya, and M. Baerns. An evolutionary approach in the combinatorial selection and optimization of catalytic materials. Applied Catalyst A: General, 200:63--77, 2000.Google ScholarCross Ref
Index Terms
- Case study: constraint handling in evolutionary optimization of catalytic materials
Recommendations
An algorithm for linear least squares problems with equality and nonnegativity constraints
We present a new algorithm for solving a linear least squares problem with linear constraints. These are equality constraint equations and nonnegativity constraints on selected variables. This problem, while appearing to be quite special, is the core ...
Original Contribution: Solving inequality constrained combinatorial optimization problems by the hopfield neural networks
The Hopfield neural networks are extended to handle inequality constraints where linear combinations of variables are lower- or upper-bounded. Then by eigenvalue analysis, the effects of the inequality constraints are analyzed and the following results ...
Linear optimization with bipolar max-min constraints
We consider a generalization of the linear optimization problem with fuzzy relational (in)equality constraints by allowing for bipolar max-min constraints, i.e. constraints in which not only the independent variables but also their negations occur. A ...
Comments