DOE Method for Optimizing Desalination Systems

Abstract

Fresh water production is one of the main concerns in the new century. Population grows fast and potable water resources are decreased. In the other hand energy crises would also be another issue that must be well addressed by the politicians and also scientists. Developing desalination plant with using renewable energy (particularly solar energy) is one of the important options to overcome these concerns. Thus many researchers have been working on different desalination plants to find the best conditions and to realize the most efficient performances for different cycles. Different approaches have been used to achieve the most efficient conditions or to find the optimum operation and design conditions. Some of the researchers used parametric study approach while many other adopted different conventional optimization algorithms for these tasks. The algorithms such as gradient based algorithm, genetic algorithm, search and pattern algorithm and neural network method have been used in the field of desalination. For instance; Ophir and Lokiec (2005) described the design principles of a MED plant and various energy considerations that result in an economical MED process and plant. Kamali and Mohebbinia (2007) showed that parametric study as one of the optimization methods on thermo-hydraulic data strongly helps to increase GOR value inside MED-TVC systems. Shamel and Chung (2006) used parametric study to find the optimum condition of a Reverse Osmosis (RO) system for sea water desalination. Metaiche et al (2008) developed optimization software, Desaltop, for RO system for water desalination. They used genetic algorithm to find suitable operating parameters and also to find appropriate type of membrane. Al-Shayji (1998) used neural network method for optimization of large-scale commercial desalination plants. Djebedjian et al. (2008) used genetic algorithm for optimization of a reverse osmosis desalination system. Mussati et al. (2003) used an evolutionary algorithm for the optimization of Multi Stage Flash (MSF) system. Finding the optimum conditions is a major challenge on the desalination plant studies. The plant performance depends on several different variables and constraints that need exhausting efforts to find the optimum conditions. This chapter introduces Design of Experiment (DOE) method as a statistical tool for optimization of desalination systems. Thus two different desalination plants; Multi-Effect Desalination (MED) system and solar desalination using humidification–dehumidification cycle (SDHD) have been considered to show the ability of DOE method for optimizing such systems. These both desalination plants could use the low graded heating energy sources