A Two-Step Optimization Approach for Retrofit of Water Network System

• Main Author: Zulfan Adi Putra, Zulfan
• Format: Thesis
• Published: 2008
• Subjects: TP Chemical technology
• Online Access: http://utpedia.utp.edu.my/1034/

Many industrial plants have been using an enormous amount of water to run their processes. The scarcities of water supply, increasing cost of providing freshwater as well as treating wastewater, and the stricter environmental regulations have driven more efficient water usage. In this regard, the most common way is the application of water reuse, regeneration, and recycle (W3R concept). Even though the W3R concept is not new, the systematic approach to select the optimum combination of the W3R is relatively new. Thus, the latter has drawn more attention of many researchers. Researchers have addressed common practical constraints such as forbidden/compulsory piping connections, minimum permissible water flowrate, and geographical constraints in developing their water network. However, most engineers challenge the researchers to provide a tool that can be controlled manually (user-interactive tool) during the designing process, especially in retrofit scenarios. This ability will be used to direct the solutions to become more realistic and practical featuring achievable targets. The tool must also able to produce multiple options simultaneously to give a large picture of what the solutions are going to be when more modifications are taken. These challenges were missed by previous works that are commonly based on graphical and automated mathematical programming methods, especially when their grassroots methods are used in retrofit designs. A two-step optimization approach is presented in this work to provide a new procedure to design multiple options of water network. The presented approach incorporates manual control (user-interactive tool) over the optimization process to consider the aforementioned practical constraints. The approach consists of structural targeting step utilizing Mixed Integer Linear Programming (MILP) and parametric optimization step using Non Linear Programming (NLP), which can be used to grassroots and retrofit design of water network. The considered water networks are water-using operations with water reuse, regeneration, and recycle (W3Rs) concept. wastewater treatment systems, and total water systems. Several water problems are solved and compared with their corresponding solutions obtained by previous work. The comparison shows that the presented approach can provide the same optimum solutions for each water problems. The approach can also provide multiple realistic solutions incorporated in "a class of good solutions" and the ability to see large picture of the developments of water system design ("project road map of achievable target and design"). These good solutions are to be further chosen by engineers based on their considerations or their company preferences. In the retrofit case study, one of the solutions has achieved a reduction of 59% of freshwater consumption. The total annual cost for the solution of retrofit water network system is reduced until 29%.