(87a) Optimal Performance Management of Clean Water Treatment Processes

A major challenge currently facing a sustainable future is the ever increasing demand for clean water of adequate quality and quantity. With growing populations and industrial growth, the need for cost-effective water purification processes has become more urgent than ever. The processes used for potable water treatment can vary greatly depending on the sources of raw water and the associated water quality. One of the biggest operational risks to water companies arise from their ability to control the day-to-day management and optimisation of their water treatment systems; and with emerging pressures to remain competitive within the global market, companies are looking for better solutions to be able to make predictions on how their treatment processes should be managed and operated.

Work in literature has so far focused solely on performance optimisation of individual water treatment unit operations, due both to the complexity of the overall process and to the interactions between unit operations. Such an approach will inevitably lead to sub-optimal performance as the operation of one unit operation has an impact on the next, and their operations can therefore not be considered in isolation. A plant-wide approach to performance optimisation will therefore be of significant benefit to the water industry by increasing the overall process efficiency and thus decreasing overall plant costs. By considering the plant as a whole, the interactions between individual processing units can be taken into account and properly balanced. A design and operating procedure that considers an optimal series of several unit operations, including units which may run in parallel, whilst at the same time reducing the number of steps required to achieve a given product quality, will therefore improve the overallplant efficiency.

Optimization based on plant superstructures has proven to be an effective tool for the synthesis of chemical engineering process flowsheets and for overall plant performance optimization. The design of the process, and the operating parameters for each individual piece of equipment, can thus all be determined optimally, but more importantly, simultaneously.

This work addresses the current gap in clean water performance research by developing an approach based on a superstructure concept for the synthesis of clean water treatment works through the application of mixed integer optimisation techniques. A systematic framework is presented for the representation of superstructures and derivation of optimisation models for process synthesis. The state task network (STN) and state equipment network (SEN) are proposed as the two fundamental representations of superstructures used to describe the overall plant operation including multiple sources of raw water and water treatment processes.  The proposed framework is illustrated by several examples to demonstrate that the approach can provide valuable guidance in clean water treatment process design and operation, thus providing a tool to achieve better day-to-day performance management.