Pareto-based distributed Q-learning optimization for fed-batch fermentation process
Intelligent algorithm used in multi-objective optimization is in rapid development. A design of a Pareto-based distributed Q-learning (PDQL) optimization strategy is presented in this paper. Q-learning algorithm makes desired solutions approximating the actual Pareto front, while the Pareto sorting method is joined to generate the nondominated solution set. Many agents working together share the experience to improve the performance of Q-learning optimization, as well as the parallel-searching capability. It generates much larger solution set which is in better distribution characteristics and is more close to the Pareto front directions. An example of the lysine fed-batch fermentation process optimization problem is taken to prove the effectiveness of this strategy. The result of PDQL optimization shows significantly better than PSO with the aggregated function method. The proposed method gives an alternative intelligent optimization strategy in this area.
Reinforcement-learning optimization Pareto-based optimization fed-batch fermentation
Dazi Li Tianheng Song Tianwei Tan Qibing Jin
College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 10 College of Life Science and Technology, Beijing University of Chemical Technology, Beijing 100029, C
国内会议
厦门
英文
1-6
2012-08-01（万方平台首次上网日期，不代表论文的发表时间）