(60s) Synthesis of Indirect Multi-Plant Heat-Integrated Water Allocation Networks

The effective utilization of water and energy is of paramount importance for promoting the sustainable development of process industrial enterprises. Water is widely employed in various capacities such as a reactant, cooling utility, and mass separation agent for scrubbing and washing processes. The use of water and energy are highly interdependent, as water is used in the generation of steam for heating purposes, while many water-dependent processes require heating or cooling.

Thus, the efficient utilization of water and energy under a holistic framework gains interest from many researchers in the area of process integration and synthesis. According to the review by Kamat¹, there are more than 100 published articles about heat integrated water allocation networks (HIWANs) in 1998-2022. From an optimization perspective, the HIWANs problem belongs to MI/NLP type that is highly nonconvex and nonlinear and very hard to handle. The methods mainly adopted for this problem can be classified as sequential or simultaneous based on pinch analysis, mathematical programming, or a hybrid approach. The sequential method relies on the decomposition of problem with different optimization objectives, although greatly reduces the computational burden, the trade-offs among all factors are not fully explored. To address this concern, the simultaneous superstructure-based model can be used. Solving the optimization model can automatically result in a final process design, but the complexity of the problem requires careful consideration of solving strategies.

To date, a significant portion of research on HIWANs has focused on individual plants, but comparatively little attention given to synthesizing HIWANs from a multi-plant perspective which can offer further opportunities for energy and water saving. The major issues hindering this approach are the high cost of interplant pipelines and pumps, as well as the potential fluctuation of one plant influencing the others. Additionally, the scale of the problem is large, and an efficient method is necessary to guarantee solution quality. To address these issues, this work proposes a novel superstructure representation of the problem based on an indirect interplant connection mode inspired by the work of Liu et al.² It is assumed that a hub is located at the center of the industrial zone that can function as a storage or wastewater treatment system, the only interplant connections are pipelines between the plant and the center hub. Also, to simplify the interplant pipeline layouts, the exporter is settled inside each plant to collect streams from the plant and send them to the hub, and the importers are settled inside each plant to receive streams from the hub. The allowable heat integration positions are not only inside each plant but also inside the hub.

The resulting superstructure is more complex than the direct interplant connection one due to the additional nonlinearity introduced by importers, exporters, and the center hub. A new sequential approach is proposed in this work to ensure a high-quality solution can be obtained within a reasonable solving time. In the first step, the trade-offs are made between water network cost and the heat integration targets by combing interplant water allocation superstructure with an improved pinch location model³. Then, the interplant connections are fixed, the original problem is decoupled to several single-plant integration problems, and a simultaneous method is used for the single-plant HIWAN design. Several examples are adopted to demonstrate the effectiveness of the proposed method.

References：

1. Kamat S, Bandyopadhyay S, Foo D.C.Y, Liao Z. State-of-the-Art Review of Heat Integrated Water Allocation Network Synthesis. Comput Chem Eng. 2022;167;108003.
2. Liu L, Song H, Zhang L, Du J. Heat-Integrated Water Allocation Network Synthesis for Industrial Parks with Sequential and Simultaneous Design. Comput Chem Eng. 2018;108;408–424.
3. Quirante N, Caballero J.A; Grossmann I.E. A Novel Disjunctive Model for the Simultaneous Optimization and Heat Integration. Comput Chem Eng. 2017;96;149–168.