A Multi-Objective Optimal Design of Thermal Distribution Systems

Abstract

A multi-objective nonlinear optimal planning method is proposed to design thermal distribution systems used for district heating. The following three objective functions are considered which are conflicting mutually and noncommensurable with one another; that is, 1) to minimize the total size of piping system, 2) to minimize the pump power, and 3) to minimize the total size of heat exchangers. Adopting the weighting method in multi-objective optimization, the abovementioned multi-objective functions are optimized by using the generalized reduced gradient algorithm. A man-machine interactive optimal planning system is developed to determine the optimally preferred solution from the set of Pareto optimal solutions derived by the method mentioned above. The validity and effectiveness of the design method proposed here are ascertained through a numerical study for a thermal distribution system, and it is certified that much worthwhile information can be obtained by the optimal planning system developed in this study.