Application of Hybrid Genetic Algorithm Routine in Optimizing Food and Bioengineering Processes
Optimization is a crucial step in the analysis of experimental results. Deterministic methods only converge on local optimums and require exponentially more time as dimensionality increases. Stochastic algorithms are capable of efficiently searching the domain space, however convergence is not guaranteed. This article demonstrates the novelty of hybrid genetic algorithm which combines both stochastic and deterministic routine for improved optimization results. A new hybrid genetic algorithm developed is applied to the Ackley benchmark function as well as case studies in foods, biofuels, and biotechnology processes. For each case study the hybrid genetic algorithm found a better optimum candidate than reported by the sources. In the case of food processing the hybrid genetic predicted 5% higher anthocyanin yield. Optimization of bio-oil production using HGA resulted in a 4.73 % higher yield. In enzyme production process HGA predicted 2.18 IU/gds of higher xylanase yield. In conclusion, hybridization of genetic algorithm with a deterministic algorithm resulted in maximum optimum compared to the statistical methods.
- Research Organization:
- Idaho National Laboratory (INL), Idaho Falls, ID (United States)
- Sponsoring Organization:
- USDOE Office of Energy Efficiency and Renewable Energy (EERE)
- Grant/Contract Number:
- AC07-05ID14517
- OSTI ID:
- 1470309
- Report Number(s):
- INL/JOU-16-39900-Rev000; FOODBV
- Journal Information:
- Foods, Vol. 5, Issue 4; ISSN 2304-8158
- Publisher:
- MDPICopyright Statement
- Country of Publication:
- United States
- Language:
- English
Web of Science
Similar Records
Neotropical termite microbiomes as sources of novel plant cell wall degrading enzymes
Tree Growth--Hybrid Genetic Algorithm for Predicting the Structure of Small (TiO₂)n, n = 2-13, Nanoclusters