In the process industries, cooling capacity is an important enabler for the facility to manufacture on specification product. The cooling water network is an important part of the overall cooling system of the facility. In this paper a cooling water circuit consisting of 3 cooling towers in parallel, 2 cooling water pumps in parallel, and 11 heat exchangers in parallel, is modelled. The model developed is based on first principles and captures the dynamic, non-linear nature of the plant. The modelled plant is further complicated by continuous, as well as Boolean process variables, giving the model a hybrid nature. Energy consumption is included in the model as it is a very important parameter for plant operation. The model is fitted to real industry data by using a particle swarm optimisation approach. The model is suitable to be used for optimisation and control purposes.

The successful operation of any petrochemical plant is dependent on the use of several utilities which may include electricity, steam, compressed air, cooling media, refrigeration media, nitrogen, condensate and fuel gas. These utilities form a significant portion of the fixed cost associated with running a plant. Utility optimisation has not received much attention until recently, driven by rising energy costs, stricter environmental policies, more competitive markets, and the threat of climate change. The generation, preparation, and transportation of utilities require energy and therefore should be optimised to reduce losses and improve operating efficiency. One example of such a utility is a cooling water system. This paper describes the modelling of a dual circuit induced draft cooling water system for control and optimisation purposes. The derived model is verified with plant data indicating promising results. The model is represented in a steady-state algebraic form as well as a dynamic state-space form.This provides a convenient basis for simulation studies and controller/optimiser design.

Various process utilities are used in the petrochemical industry as auxiliary variables to facilitate the addition/removal of energy to/from the process, power process equipment and inhibit unwanted reaction. Optimisation activities usually focus on the process itself or on the utility consumption though the generation and distribution of these utilities are often overlooked in this regard.Many utilities are prepared or generated far from the process plant and have to be transported or transmitted, giving rise to more losses and potential inefficiencies. To illustrate the potential benefit of utility optimisation, this paper explores the control of a dual circuit cooling water system with focus on energy reduction subject process constraints. This is accomplished through the development of an advanced regulatory control (ARC) and switching strategy which does not require the development of a system model, only rudimentary knowledge of the behaviour of the process and system constraints. The novelty of this manuscript lies in the fact that it demonstrates that significant energy savings can be obtained by applying ARC to a process utility containing both discrete and continuous dynamics. Furthermore, the proposed ARC strategy does not require a plant model, uses only existing plant equipment, and can $ This work is based on research partly supported by the National Research Foundation of South Africa (Grant No. 90533 be implemented on control system hardware commonly used in industry. The simulation results indicate energy saving potential in the region of 30% on the system under investigation.