A multi-objective optimization method based on discrete bacterial algorithm for environmental/economic power dispatch

• Published in: Natural computing Vol. 16; no. 4; pp. 549 - 565
• Main Authors: Tan, Lijing, Wang, Hong, Yang, Chen, Niu, Ben
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.12.2017; Springer Nature B.V
• Subjects: Algorithms; Artificial Intelligence; Bacteria; Complex Systems; Computer Science; Economics; Evolutionary algorithms; Evolutionary Biology; Fitness; Multiple objective analysis; Optimization; Power dispatch; Processor Architectures; Theory of Computation; Discrete bacterial-based algorithm; Environmental/economic power dispatch; Multi-objective optimization
• ISSN: 1567-7818; 1572-9796
• DOI: 10.1007/s11047-017-9620-7

Multi-objective optimization is an interesting and hot topic in the literature involving the conflicting objectives to be solved simultaneously. In this study, a new multiple optimization method based on a discrete bacterial algorithm is developed to address the multi-objective economic-environmental dispatch problem with non-linear, non-convex, and complexity constraints. In the proposed multi-objective bacterial based algorithm, the existence of bacteria complies with a fitness survival mechanism, in which a health sorting approach is operated to control the chances of reproduction as well as elimination. The performances of bacteria have been recorded and sorted for health evaluation, which can help to group the individuals according to their search capability and improve the overall quality of the population. To speed up the convergence rate and avoid local minima to some extent, a comprehensive learning strategy is embedded to enable the communication exchanges between the bacteria and external archive. The standard IEEE 30-bus, 6-generator test system is adopted to illustrate the efficiency of the proposed method by making the comparison with the other multiple bacterial-based algorithms as well as six other well developed evolutionary algorithms. The effectiveness of the propose method is well validated in experiments by providing similar or superior solutions to environmental/economic power dispatch issues considering the various constraints.