Evaluating the choice of radial basis functions in multiobjective optimal control applications

• DOI: 10.1016/j.envsoft.2023.105889
• Publication Date: 2023-11-15T17:24:19Z
• Authors: Jazmin Zatarain S...; Jan H. Kwakkel; Mark Witvliet
• Keywords: Pareto optimal

Evolutionary Multi-Objective Direct Policy Search (EMODPS) is a prominent framework for designing control policies in multi-purpose environmental systems, combining direct policy search with multi-objective evolutionary algorithms (MOEAs) to identify Pareto approximate policies. While EMODPS effective, the choice of functions within its global approximator networks remains underexplored, despite their potential significantly influence both solution quality and MOEA performance. This study conducts rigorous assessment suite Radial Basis Functions (RBFs) as candidates these networks. We critically evaluate ability map system states actions, assess on efficient apply this analysis two contrasting case studies: Conowingo Reservoir System, which balances competing water demands including hydropower, flows, urban supply, power plant cooling, recreation; The Shallow Lake Problem, where city navigates trade-off between economic objectives when releasing anthropogenic phosphorus. Our findings reveal that RBF substantially impacts model outcomes. In complex scenarios like reservoir control, critical, while simpler contexts, such less pronounced, though distinctive differences emerge characteristics prescribed strategies.