Non-interactive approach to solve multi-objective thermal power dispatch problem using composite search algorithm

[Display omitted] •A non-interactive approach is employed to solve multi-objective problem.•A Chaotic differential evolutionary and Powell’s method based approach is proposed.•Two CDEPS variants: Gauss map and Tent map based are proposed.•Tested on generalized test functions and multi-objective load...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Applied soft computing Ročník 65; s. 644 - 658
Hlavní autoři: Singh, Nirbhow Jap, Dhillon, J.S., Kothari, D.P.
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 01.04.2018
Témata:
Chaotic differential evolution algorithm
Multi-objective optimization
No preference approach
Powell’s pattern search
Surrogate worth function
Thermal power load dispatch
Thermal power load dispatch
Multi-objective optimization
Surrogate worth function
Powell’s pattern search
Chaotic differential evolution algorithm
No preference approach
ISSN:1568-4946, 1872-9681
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:[Display omitted] •A non-interactive approach is employed to solve multi-objective problem.•A Chaotic differential evolutionary and Powell’s method based approach is proposed.•Two CDEPS variants: Gauss map and Tent map based are proposed.•Tested on generalized test functions and multi-objective load dispatch problem.•Wilcoxon signed rank test is employed to analyze the performance. This paper deals with a non-interactive approach to solve multi-objective thermal power load dispatch (MTPLD), where either decision maker is not involved or preference information is available in prior. To reduce the computational complexities due to generation of Pareto-front and selection of satisficing solution, this paper adopts no-preference approach. A satisficing function to resolve the conflict of non-commensurable objectives is proposed, which reformulates the MTPLD problem as scalar thermal power load dispatch (MTPLD) problem. Owing to ambiguous or vague in objectives, the proposed method exploits fuzzy theory. MTPLD problems’ satisfying solution is obtained by implementing hybrid chaotic differential evolution algorithm and Powell’s pattern search algorithm (CDEPS). The chaotic differential evolution algorithm is responsible for the diversification of feasible solutions and provides global solution. Whereas Powell’s pattern search, method improves the exploitation by performing local search. The paper investigates the performance of two CDEPS variants based on Gauss map and Tent map, respectively. The performance of the proposed solution procedure is analyzed using generalized benchmark test functions and complex MTPLD problems. The exhaustive analysis using non-parametric significance test and descriptive statistics shows that the Tent map based CDEPS solution procedure has better ability to generate quality generation schedule and faster convergence rate.
ISSN:1568-4946
1872-9681
DOI:10.1016/j.asoc.2018.01.044