Automated Design of Total Water Systems

Abstract

An automated method for the design of total water systems is developed in this paper. This approach considers simultaneously the optimal distribution of water to satisfy process demands and optimal treatment of effluent streams. Treatment can be for discharge to the environment or for regeneration of wastewater. The cases of regeneration reuse and regeneration recycle can be distinguished in the approach. It combines engineering insights with mathematical programming tools based on a superstructure model that results in a mixed-integer nonlinear programming problem. The approach features a fast and robust solution strategy. Complex tradeoffs involving operating, as well as capital, costs and other practical constraints have been included. In particular, piping and sewer costs, which are a major element in the capital cost of such systems, can be included. Network complexity is controlled by specifying the minimum permissible flow rates in the network, the maximum number of streams allowed at mixing junctions, and the inclusion of piping costs in the problem formulation. In addition to being able to solve the problem of total water system design, the approach is capable of designing water-using systems and effluent treatment systems when considered individually. Case studies are used to demonstrate the method.

On the use of systems technologies and a systematic approach for the synthesis and the design of future biorefineries

“…Such methods are plentiful on energy targeting (Gundersen, 1988;Briones and Kokossis, 1999a,b), water targeting (Wang and Smith, 1994a,b;Smith, 2005;Hallale and Fraser, 1998;Alva-Argaez et al, 1999;2007a,b;Gunaratnam et al, 2005), and mass-exchange networks (Halwagi, 1997(Halwagi, , 1990a(Halwagi, ,b, 2006(Halwagi, , 2008, but also in reactor and reaction engineering targeting (Halwagi and Srinivas, 1992;Marcoulaki and Kokossis, 1999;Kokossis, 1997, 1998;Kokossis and Floudas, 1990), simple and complex distillation Kokossis, 2001, 2002), solvent design (Papadopoulos and Linke, 2006a, b;Kazantzi et al, 2007, Kazantzi andHalwagi, 2005;Kokossis, 2000, 2001) and reaction-separation systems (Linke and Kokossis, 2003a,b, 2004Kokossis and Floudas, 1991). Instead, literature evidence indicates a predominant use of general flowsheeting technology for techno-economic and case-by-case analysis, using Aspen Plus for instance, as in the work of NREL on bioethanol production (Wooley et al, 1999) and that of Haas et al (2006) for biodiesel.…”

Optimum Design of Waste Water Treatment Network

“…As suggested through equation (19), T i is the cheapest prioritized cost based on the treatment pinch C pj . Due to change of pinch; the prioritized cost for each treatment unit can be changed.…”

“…Note that equation (21) can be derived using equations (19) and (20) through simple algebraic manipulation (not included due to brevity).…”

Review of Mixed‐Integer Nonlinear and Generalized Disjunctive Programming Methods

“…The third technique is sequential optimization, originally presented by Takama [212]. Some of the sequential optimization approach methods include: a relaxed NLP model [213], dividing and sequentially reducing concentration intervals for WUN [214], an MILP-LP technique for TWN [215], and an LP-NLP approach for the WWTN [216]. The fourth strategy is to solve the MINLP problem by using meta-heuristic algorithms [217][218][219].…”