A mathematical model for water and energy networks

Abstract

Mathematical programming is one the most used techniques in process integration, especially in water and energy network designs. Unlike conceptual and graphical approaches, mathematical programming is a better option in dealing with complex industrial water and energy systems, involving multiple contaminants and mass transfer based and non-mass transfer based operations. This thesis presents the development of a mathematical model for minimum water and energy networks considering direct heat transfer. The approach optimizes a superstructure which represents a set of all potential water minimisation arrangements together with direct heat transfer options and water and energy network configurations in a process system. The model has been set to minimize fresh water and energy consumption, cost applied to the system and wastewater discharged from the system. The model formulation is a mixed integer nonlinear program (MINLP) that is used to optimize an existing design. It considers all stages of water management hierarchy (i.e. elimination, reduction, reuse, outsourcing and regeneration) and operating cost factors simultaneously to bring about the lowest total cost. In this work fresh water contaminant concentration can be assumed as either zero or non-zero. The constraint for waste water temperature has been considered in the model. The model has been tested with a case study of a paper mill plant for retrofit case. The results show a minimization of 20.3% in annual operating costs which is roughly a 5 million dollar savings per year for the plant. The model showed that 97.96% reduction in wastewater generation and 60.2 % in utility consumption is achievable in compare with the previous graphical method. This shows that the model is very beneficial for the retrofit of industrial water and energy networks.