Development of a new self-adaptive F-NSGA-III algorithm with fuzzy structure for designing urban water distribution networks

ABSTRACT The present study establishes an optimization procedure in several sample networks with the objective of reducing the cost and eliminating the pressure deficit throughout the entire network. The optimization mechanism is implemented by coding the novel self-adaptive F-NSGA-III algorithm with a fuzzy structure in the MATLAB environment and integrating it with the EPANET framework. Initial definition of the cost function is based on the correlation between the cost, diameter, and length of the pipes. Subsequently, the program is ran through 10,000 and 20,000 iterations. In order to enhance the rate of convergence, the cost resulting from the breach of the permissible pressure range (minimum: 30 m and maximum: 60 m) is included into this function. Subsequently, the program is executed once more to get the optimal solution. The results indicate that the F-NSGA-III algorithm, has superior speed in identifying optimal solutions as compared to NSGA-II. In this method, the utilization of fuzzy structure and the consideration of the cost of violating the permissible pressure in each iteration result in the fastest attainment of the optimal response, as previously achieved by other researchers, for sample networks. Therefore, this structure yields more optimal solutions with fewer iterations and substantially reduces the time required to achieve convergence.