Development of a framework for the optimum design of a cooling water system with a time constrained background process

Abstract

Cooling water systems research aims to investigate the interaction between the cooling water network and cooling tower performance. Historically, it has been assumed that the background process operates continuously. This dissertation presents a mathematical framework for the synthesis and optimisation of a cooling water system consisting of multiple cooling towers with a batch background process. The cooling water network is formulated as a series con guration, which is characterised by reuse of cooling water between di erent cooling water using operations. In achieving this, an overall superstructure is developed which allows for all di erent combinations between the cooling towers and cooling water network, as well as opportunities to reuse water between cooling operations to be explored. It should be noted that in previous research on cooling water systems, reuse opportunities were constrained only by temperature. Due to the nature of batch processing, reuse opportunities are subject to both temperature and time constraints in this investigation. The mathematical model is developed as a mixed integer nonlinear programming (MINLP) problem in the GAMS platform. The objective function is the maximisation of total annual pro t. The MINLP is solved using the BARON solver. The optimal system con guration, operating conditions and design parameters for the cooling towers and cooling water using operations in the network are determined as part of the optimisation of the mathematical formulation. The model was applied to two illustrative examples, a xed schedule problem and a exible schedule problem. The use of a sequential optimisation approach was also compared to an integrated optimisation approach. Results from the illustrative examples demonstrated the superiority of the integrated approach, which led to a reduction in the number of cooling towers required from 3 to 2. Total annual pro t was increased by 0.45 %, recirculating cooling water was reduced by 42.1 %, average overall cooling tower e ectiveness was increased by 25.3 %, and the makeup water required throughout the time horizon was reduced by 6.3 %.