Procedure for equipment selection in mechanical systems based on capital cost minimization and optimal loading

Abstract

Providing maximum performance at minimum cost is a major challenge in the design of mechanical systems. One of the key problems is determining the optimal schedule of a set of units operating in parallel (such as chillers or turbines) that minimizes the overall energy consumption. This is a widely studied problem in the literature known as “economic dispatch” (in power systems) or as “optimal loading problem” (in air-conditioning systems). However, the loading problem requires the units to be previously selected. This work proposes a procedure to solve the selection problem and the loading problem at the same time. The proposal is a novel procedure that allows determining, in a rigorous manner, the units that have to be purchased and the corresponding operation schedule. It was applied to two case studies: i) selection of chillers in cooling plants, and ii) selection of the utility plant in oil and gas offshore platforms. Both cases are analyzed using two alternative optimization approaches. First, a mono-objective optimization of a single cost-based function and, second, a multi-objective optimization of capital cost and energy consumption. The results in all cases showed that the total nominal capacity of the selected units is not necessarily closed to the peak load, which is a common rule-of-thumb guideline for equipment selection. In the case of the cooling plant, the nominal capacity of the chillers selected was up to 1.6 times the peak load demand. Likewise, the best solutions in the case of the utility plant consisted of very different-sized models. These counter-intuitive results demonstrate the importance of using a systematic selection procedure.