An improved sampling strategy based on trajectory design for application of the Morris method to systems with many input factors

摘要

In this paper, a revised version of the Morris approach, which includes an improved sampling strategy based on trajectory design, has been adapted to the screening of the most influential parameters of a fuzzy controller applied to WWTPs. Due to the high number of parameters, a systematic approach has been proposed to apply this improved sampling strategy with low computational demand. In order to find out the proper repetition number of elementary effects of each input factor on model output (EE,) calculations, an iterative and automatic procedure has been applied. The results show that the sampling strategy has a significant effect on the parameter significance ranking and that random sampling could lead to a non-proper coverage of the parameter space.