Integrated Synthesis and Design of Process and Water Networks

Abstract

This work presents a systematic framework for a simultaneous synthesis of process and wastewater treatment network using superstructure-based optimization method. The overall superstructure is composed of i) the process network ii) the wastewater treatment network that connected by converter intervals. In this approach, the problem is generally expressed as a mixed-integer nonlinear programming (MINLP), which is solved to identify the optimal configurations for the process and water network, among a set of feasible alternatives, according to selected performance criteria. A solution strategy to solve the multi-network problem accounts explicitly the interactions between the networks by selecting suitable technologies in order to transform raw materials into products and produce cleaned water to be reused in the process. The features of the developed synthesis method has been demonstrated on conceptual and bioethanol production case studies.

Introduction

Process synthesis offers an attractive framework for undertaking various design problems through a systematic framework, either using sequential or simultaneous optimization approach. For the former, the overall process system is decomposed into different subsystems for ease of analysis. However, these subsystems (i.e., resource conservation network, heat exchanger network, water network etc.) may not guarantee a truly optimized system as they are synthesized separately after the process flowsheet is obtained. On the other hand, the simultaneous approach gives a better solution as all interactions and economics trade-offs are taken into consideration explicitly [1-3]. The solution strategies of an integrated network problem are able to find the optimal raw material, product portfolio, process and wastewater treatment technology by simultaneously screening various alternatives included in the search space, while minimizing fresh water intake and satisfying environmental regulations [4]. In addition, various water minimization strategies (i.e. reuse, recycle, regeneration and treatment-reuse) are considered in order to reduce the amount of freshwater intake.

Systematic Framework

The framework is supplemented by a software infrastructure based on Ms EXCEL for gathering required input data and General Algebraic Modeling System (GAMS) for the solution of the formulated optimization problem. The framework has been developed in an earlier work and the details of the framework can be found in Handani et al. [4]. The systematic framework consists of four main steps.
After defining the goal and scope of optimization problem as well as objective function in the first step, one can define a superstructure that consist of various alternatives for the process and wastewater treatment networks are specified with respect to raw materials, technologies and products. The treated wastewater in the treatment tasks can be either discharged into the environment and/or it can be recycled for use in the same processes or in neighboring processes. Then, generic models based on mass input-output describing each of the elements of the superstructure are developed. Generic model parameters require to represent the activity of each interval includes chemical and utility consumption, reaction conversion, split fraction and separation efficiency are collected from various sources i.e. open literature, simulation and technical report. Finally, the optimal process and wastewater treatment network are then formulated as MI(N)LP and solved under different process synthesis-design scenarios.

Supervisors

Prof. Georgios M. Kontogeorgis
Prof. Kim Dam-Johansen

To see figures and references for this PhD study, please download the Graduate School Yearbook.

The yearbook provides an overview of most of the PhD theses currently being written at DTU Chemical Engineering and is updated every year in February.