Multi-step evolution strategy based DNA generic algorithm for parameters estimating

A multi-step evolution strategy based DNA genetic algorithm with the multi-step evolution strategy and a new random set crossover with new fitness function is proposed for solving the parameter estimation problem of chemical process. The algorithm adopts DNA encoding and operators. Three new kinds of crossover operators are designed which maintain the diversity of population. A new adaptive mutation rate is also applied to guarantee against stalling at local peak. A new fitness function is designed, which can compare individuals have high similarity. Besides, in order to release the dependence of the range of initial solution on experience set, strengthen the global and local search ability, the multi-step evolution strategy with interrupting genetic, simulated annealing algorithm and parameters interval evolution strategy are developed. Numerical experiment on four typical test functions and heavy oil thermal cracking parameter model are carried out show the efficiency and effectiveness of the proposed algorithms.

[1]  D. E. Goldberg,et al.  Genetic Algorithm in Search , 1989 .

[2]  Venkat Venkatasubramanian,et al.  A hybrid genetic algorithm for efficient parameter estimation of large kinetic models , 2004, Comput. Chem. Eng..

[3]  Behrang Mansoornejad,et al.  A hybrid GA-SQP optimization technique for determination of kinetic parameters of hydrogenation reactions , 2008, Comput. Chem. Eng..

[4]  Heng Wei,et al.  Study on continuous network design problem using simulated annealing and genetic algorithm , 2009, Expert Syst. Appl..

[5]  Heng Wei,et al.  Study on continuous network design problem using simulated annealing and genetic algorithm , 2009, Expert Syst. Appl..

[6]  Jili Tao,et al.  DNA computing based RNA genetic algorithm with applications in parameter estimation of chemical engineering processes , 2007, Comput. Chem. Eng..

[7]  Li Shu,et al.  Operational Rules for Digital Coding of RNA Sequences Based on DNA Computing in High Dimensional Space , 2003 .

[8]  Dirk Tomandl,et al.  A Modified General Regression Neural Network (MGRNN) with new, efficient training algorithms as a robust 'black box'-tool for data analysis , 2001, Neural Networks.

[9]  Ning Wang,et al.  A novel RNA genetic algorithm for parameter estimation of dynamic systems , 2010 .

[10]  Stephen M. Krone,et al.  The genealogy of samples in models with selection. , 1997, Genetics.

[11]  Ning Wang,et al.  DNA Double Helix Based Hybrid GA for the Gasoline Blending Recipe Optimization Problem , 2008 .

[12]  Zheng Qifu.  Modified holographic research strategy and its application to optimization in chemical engineering , 2006 .

[13]  Ning Wang,et al.  A DNA based genetic algorithm for parameter estimation in the hydrogenation reaction , 2009 .

[14]  Ehl Emile Aarts,et al.  Simulated annealing and Boltzmann machines , 2003 .

[15]  Jianwei Yin,et al.  A DNA-Based Genetic Algorithm Implementation for Graph Coloring Problem , 2005, ICIC.

[16]  Gheorghe Paun,et al.  DNA Computing: New Computing Paradigms , 1998 .