(440g) Sustainable Synthesis and Capacity Expansion Planning for the Integrated Multi-Network Process Under Endogenous Uncertainty

With the targets for peak CO₂ emissions and carbon neutrality in mind, using advanced optimization methods to deal with international challenges associated with climate change and sustainability is of great significance. Sustainable development has received extensive attention in recent years, and new requirements have been placed on the design, planning, and scheduling of renewable energy conversion systems and sustainable chemical processes. A multi-network integrated optimization framework including process network, water network, and energy network is one of the critical steps to achieve sustainable chemical process design. In addition, due to the vast existence of various types of uncertainties, industrial processes generated by the deterministic optimization approach are prone to a lack of robustness and optimality. Therefore, decision-makers need to make optimal decisions under uncertain conditions to improve the reliability and economy of complex process systems. In summary, this paper focuses on sustainable design and capacity expansion planning for the integrated chemical process, aiming to achieve zero-carbon emissions and sustainable development under uncertainty.

Aiming at a sustainable integrated multi-network framework that integrates chemical process, water supply, wastewater treatment, heat generation, power generation, and carbon capture, utilization and storage (CCUS) network, this research takes both economic benefits and environmental implications into account, which provides specific theoretical and practical guidance for the synthesis, design, and capacity expansion planning of sustainable chemical processes under endogenous uncertainty.

Compared with exogenous uncertain factors related to external markets, such as price fluctuations of raw materials and product demand changes, we here mainly consider the potential impact of endogenous uncertain parameters, such as the reaction conversion rate of crucial steps. Consequently, a multi-stage stochastic programming model is constructed to deal with endogenous uncertainty. In order to efficiently solve the formulated large-scale mixed-integer linear programming model (MILP), a lagrangian decomposition algorithm is proposed. The framework mentioned above and the optimization approach are implemented into the design of the sustainable production process of xylitol, where a couple of case studies are carried out. Results illustrate that with the help of multi-stage stochastic programming, the trade-off between the optimality and robustness of process operations as well as the reliability of the optimized system can be significantly improved when compared to the deterministic model under the same conditions. At the same time, the results show that the multi-network model can find the optimal solution with high economic benefits and low carbon emissions under different scenarios, proving the proposed method's applicability and advantages.