eBooks View all | By AtoZ | By category
Multiobjective evolutionary algorithms knowledge acquisition system for renewable energy power plants
By Burak Omer Saracoglu
Orhantepe Mahallesi, Tekel Caddesi, Istanbul, Turkey
Published: May 17, 2019 | pg. no: 1-21
Abstract: Engineers have to present the best of best solutions amongst the best solutions to engineering problems or engineering design problems (EP, EDP, EPs, EDPs). Hence, EPs or EDPs are indeed Multiobjective Optimization Problems (MOPs). Although all EPs or EDPs are MOPs in reality, only a few of them can be modeled as MOPs, some of them can be modeled as Single-Objective Optimization Problems (SOPs) and most of them cannot even be modeled as MOPs or SOPs, because of the difficulties of EPs or EDPs and optimization studies. According to these basic facts, a multiobjective evolutionary algorithm knowledge acquisition system for renewable energy power plants (MOEAs-KAS-F-REPPs) is proposed to deal with those difficulties. The proposed MOEAs-KAS-F-REPPs will help engineers in the renewable energy field to work with the most appropriate and satisfactory MOPs in their daily work routine. The proposed knowledge acquisition system in its Research, Development, Demonstration, Deployment, and Diffusion (RD3&D) stages are explained in a concise style. A representative example based on some experimental test MOPs with some linear, quadratic, polynomial functions is also presented with a brief descriptive way to show how the proposed knowledge acquisition system will operate after its RD3&D stages. According to the proposed MOEAs-KAS-F-REPPs design, the representative example has selective and elective proposed standard objectives and constraints (as test objectives and constraints). A standardized MOP is developed and saved into its own console for a virtual small hydropower plant design and investment (VSHPDI). The Pareto Optimal solutions are found by only one algorithm (NSGA-II) in the Scilab 6.0.1 on a desktop computer configuration (Windows 10 Pro, Intel(R) Core(TM) i5 CPU 650 @3.20 GHz, 6,00GB RAM with internet connection). The algorithm run-times of the current applications are between 29,489 and 50,666 seconds. All data and information are stored for the next applications and improvements according to the RD3&D philosophy of the proposed MOEAs-KAS-F-REPPs.
View eBook