Conference Agenda

The water-energy nexus has been introduced on the scope of the most relevant sustainability policies of each region of the world. This concept deals with all potential interdependencies between water and energy, and the full understanding of the inherent aspects is fundamental for the promotion of simultaneous water and energy use-related benefits (Gabbar and Abdelsalam, 2020; Souza et al., 2023; Walsh et al., 2015).

The concept of Water and Energy Integration Systems (WEIS) was recently introduced by Castro Oliveira (2023). It consists of physical systems encompassing a set of water and energy-using units in a site, and the water and energy recirculation between these units, with the goal of achieving economic and environmental benefits. The innovative concept was developed to expand the Total Site Integration (TSI) and Combined Water-Energy Integration (CWEI) concepts, and to attend to their limitations.

While TSI only focus on energy consumers (despite considering all types of energy forms involved in an energy system), CWEI focus on water consumers, with a limited focus on energy efficiency (only though the reduction of hot/ cold utility input in a water system).On the other hand, the WEIS concept focus on energy and water consumers as a whole, adapting conceptual elements from both TSI and CWEI and introducing new ones. These include technologies for energy recovery from water/ wastewater units (such as electrolysis) and the supply of thermal energy (direct waste heat) or electric energy (indirectly produced in thermodynamic cycles) to fulfill already existing energy requirements (hot utilities, in this case) and additional ones (wastewater treatment units). These practices necessarily generate new stream recirculation, for instance, additional fuel streams (such as hydrogen from electrolysis) to combustion-based processes and waste heat streams to several points in the water system.

This work details the methodology behind the concept of WEIS, by elaborating two conceptual superstructure configurations: a standard one (case 1) based on a steady-state perspective (continuous processes only) and a dynamic one (case 2) considering energy storage units (both continuous and batch processes). To prove the innovation and real-life adequacy potential associated to the concept, two representative plants, set within the Portuguese ceramic industry (to which economic and environmental viability have been secured) were used as case studies to apply both conceptual configurations. For case 1, it was estimated an eco-efficiency promotion of 6.46%, 8.57% total energy savings and 23.71% total water savings. For case 2, it was estimated an eco-efficiency promotion of 4.00%, 6.88% total energy savings and 38.57% total water savings.

References

M. Castro Oliveira, 2023. Simulation and Optimisation of Water and Energy Integration Systems (WEIS): An Innovative Approach for Process Industries.

H.A. Gabbar, A.A. Abdesalam, 2020. Energy—water nexus: Integration, monitoring, KPIS tools and research vision, Energies, 13.

R.G. Souza, A. Barbosa, G. Meirelles, 2023, The Water–Energy Nexus of Leakages in Water Distribution Systems, Water, 15.

B.P. Walsh, S.N. Murray, D.T.J. O’Sullivan, 2015. The water energy nexus, an ISO50001 water case study and the need for a water value system, Water Resour. Ind., 10, 15–28.

Industrial symbiosis has emerged as a paradigm to enhance industrial objectives (economic, environmental, etc.) by promoting collaboration between facilities that share resources such as heat and power, but also other tangible or intangible assets, like intermediate products or client satisfaction. In this context, a shift from intra-company to inter-company integration has emerged, e.g., in the form of eco-industrial parks. However, the proper management of these networks requires to reach consensus among multiple stakeholders with conflicting objectives [1]. To provide a transparent and practical approach, this work develops a novel price fixation strategy for resources exchanged within industrial symbiosis networks. By directly linking economic benefits to resource prices, the strategy enables companies to understand the value of each exchange. This approach simplifies bookkeeping, fosters stakeholder trust, and facilitates the formation of symbiotic networks.

Building upon the Process Integration Multi-Actor Multi-Criteria Optimization (PI‑MAMCO) framework [2], which balances multiple objectives and derives fair and stable benefit allocations based on stakeholder preferences and contribution, the proposed strategy begins by identifying all intermediate resources exchanged within the network, including materials and utilities. Price ranges for these resources are established using reference values from market data, ensuring that prices accurately reflect the economic context. The fair prices for these intermediate assets (price fixation policy) are based on the formulation of the economic balances for each company, incorporating the allocated economic benefits derived from the PI-MAMCO framework. Then, the core of the proposed solution approach involves selecting an optimization criterion, such as the minimization of the deviation of internal prices from reference market prices, while satisfying the economic balances for each company. These prices can then form the basis for contractual agreements within the symbiotic network, providing a clear and equitable mechanism for economic benefit allocation without the need for side payments.

A case study involving the formation of a palm oil–based industrial complex [3] is presented to demonstrate the approach. The complex comprises three companies that exchange materials, utilities, and effluents. The case study illustrates how the described price fixation strategy effectively allocates economic benefits among the companies, and enhances transparency in the value of resource exchanges, to support the successful establishment of the symbiotic network. This price fixation strategy offers a practical approach for companies considering participation in industrial symbiosis networks.

[1] M. Hiete, J. Ludwig, F. Schultmann (2012). Journal of Industrial Ecology.
[2] F. Lechtenberg, L. Aresté-Saló, A. Espuña, M. Graells (2024). Applied Energy.
[3] Y. D. Tan, J. S. Lim, S. R. Wan Alwi (2022). Energy.

Due to safety, economic and logistical challenges, flaring of small to medium volumes of natural gas remains a widespread practice, despite global focus on carbon emissions reduction. This detrimental practice results in climatic decline and waste of valuable resources. The global effort to curb gas flaring has led to increasing innovative technologies for the valorization of these small - medium (stranded) NG volumes (Global Gas Flaring Reduction Partnership, 2019).

This study presents a decision-making support tool that allows for fast evaluation of sustainable, viable alternatives for utilization of small to medium-volume natural gas over flaring. The proposed tool first computes the flare intensity and toxicity of NG profile, including gas composition and volume, provided by the user, to establish an environmental baseline. It then allows the user to select a desired objective between economic, environmental, or with a tradeoff. Based on the selected objective, optimization is performed, and the analysis of performance indices for the utilization processes are presented. The performance indices include the levelized cost of product, return on investment, payback period, capital, and operating cost, for the economic baseline, while the environmental baseline includes CO₂ produced, emitted, or avoided from the process, as well as product CO₂ emission potentials. The technical performance indices include energy consumption, energy efficiency and feed efficiency. The tool involves the integration of python with Aspen Plus to develop a user-friendly interface that facilitates decision making. The optimization framework focuses on maximizing economic profit with strict adherence to environmental constraints. Different scenarios from best to worst are tested to using the proposed tool. Focusing on sustainable processes, existing and new processes with low-zero carbon emission, as well as new integrated processes with reduced carbon emissions are presented. The data used for this study would be obtained from the world bank gas flaring data, the global gas flaring reduction partnership report (Darani et al., 2021), and other literature sources for process operating conditions. The proposed tool enables stakeholders to evaluate available clean technologies and the possible concessions required to reduce environmental impact of flaring and optimize natural gas valorization.

Prospective lifecycle design of emerging technologies combined with Life Cycle Assessment (LCA), Material Flow Analysis (MFA) and Input-Output Analysis (IOA) plays an important role in the design of sustainable societies and business strategies. However, the prospective lifecycle design tends not to be seamlessly linked to technological development. In the example of air conditioners (ACs), state-of-the-art development is taking place for each component, such as heat exchangers for indoor and outdoor unit, expansion valves, compressors, piping and refrigerants. However, these development data are too detailed to provide for lifecycle design to build a business strategy. On the other hand, useful indicators such as COP (Coefficient of Performance) show no relationship among design parameters, making it difficult to support decision making. In this study, computer-aided process modeling was positioned as a function that links technology development and the system level analyses, and a fundamental model combining ACs process simulation, MFA and LCA was developed.

In the process modeling, a process flow diagram including heat exchangers for indoor and outdoor, a compressor and an expansion valve was modelled, and isenthalpic expansion and isentropic compression were assumed in the expansion valve and the compressor, respectively. In the heat exchangers for indoor and outdoor units, heat exchanging process between refrigerants and ambient air and indoor air were displayed on Temperature-enthalpy diagram, then UA (Overall heat transfer coefficient) values of each heat exchanger were quantified to calculate the relative change in heat exchanger size. By using this process modeling, operating power and size of each component for conventional and natural refrigerants could be defined. In the MFA, statistic data on past and estimated future cooling demand were firstly incorporated, then shipping amount of air conditioners was calculated by balancing of the installed stock and the waste amount estimated by Weibull distribution. Annual power consumption was estimated using performance calculated by the process modeling, then a cradle to grave LCA of ACs was conducted until 2050.

Case studies of replacing conventional refrigerants with natural refrigerants show that if air conditioners are designed to maintain their performance, the size of the heat exchanger will increase, and the share of the impact of air conditioner manufacturing in the total environmental burden in 2050 will be larger. On the other hand, if the air conditioner is designed to maintain the size of the heat exchanger, its performance will deteriorate drastically and its electricity consumption will increase, making its environmental impact strongly dependent on the power grid inventory.

Biomass sugars from crops are the typical substrates used in the microbial production of useful substances. These bioproduction processes have been shown to be more environmentally friendly than conventional petrochemical processes. However, the Earth's biomass production capacity is limited by its biophysical boundaries, and it cannot meet the enormous demand for fuel and chemical production. Additionally, it is important that the expansion of industrial agriculture has negative aspects, such as vast land use and massive consumption of depletable resources like water, nitrogen, and phosphorus. Thus, for biomanufacturing technology to be fully implemented, concerns regarding the supply of biomass sugar need to be addressed.

Against this background, research is being conducted to synthesize sugars through an agriculture-independent catalytic process. The chemical sugar synthesis is achieved by integrating (1) the formation of formaldehyde through CO₂ reduction and (2) the sugar synthesis using formaldehyde as a substrate. One of the main issues to be addressed is the development of a catalyst that can selectively synthesize sugars from formaldehyde. Consequently, we have been developing catalysts to improve the selectivity of this sugar synthesis. In this study, we have achieved a significant improvement in sugar selectivity by using metal oxometalates as catalysts to suppress side reactions¹. Furthermore, by using Corynebacterium glutamicum, bioproduction using chemically synthesized sugars has been realized for the first time globally².

To implement such emerging technologies in society, it is essential to work in tandem with technological development to design future life cycles that will enable system design and evaluation based on the assumption of future performance improvements and infrastructure development. As a further effort in this study, we have applied the knowledge gained from research and development using actual materials to system design to create a system that can produce sugar at maximum efficiency while considering technical, economic, and social feasibility. More specifically, we have designed a system that includes a series of processes from the procurement of raw materials to the refining of the produced sugar, and we have calculated the material and energy balance using a process simulation. Subsequently, a life-cycle assessment will be conducted based on the inventory data obtained, and guidelines for future technological development will be presented.

The environmental impact of the excessive expansion of biomass sugar utilization has been pointed out in the past. However, sugar production is currently dependent on agriculture, and there have been no solutions to replace it with other methods. In contrast, chemical sugar synthesis is much faster than the agricultural process on which biomass sugars depend, and it consumes far fewer resources, such as land and water. This study quantitatively demonstrates the social significance of supplementing sugar production with chemical synthesis processes. Consequently, it is expected to provide new methodologies and perspectives for the field of biomanufacturing, which has been based on the use of biomass sugars, and ultimately for the entire carbon resource recycling system, including food production.

1. H. Tabata, et al., Chem. Sci. 2023,14, 13475-13484

2. H. Tabata, et al., ChemBioChem 2024, 25, e202300760

Bilevel optimization is an active area of research within process systems engineering due to its ability to capture the interdependencies between two levels of decisions. This becomes particularly valuable in decentralized supply chain optimization, where participants have conflicting interests and compete against each other. Bilevel models excel in representing such leader-follower relationships by incorporating the follower’s dynamics as a constraint in the leader’s upper-level problem. While continuous bilevel problems are already computationally challenging, the presence of discrete decisions—such as production recipes, technology selection, and capacity levels—further complicates their solution (Jeroslow, 1985).

State-of-the-art optimization solvers like Gurobi and CPLEX now offer extensions to handle mixed-integer linear bilevel problems directly, implementing branch-and-cut algorithms inspired by the original work of Moore and Bard (1990). However, current approaches suffer from scalability issues when applied to large-scale real-life instances (Yue and You, 2017), which evidences the need for innovative strategies, such as neural networks, to better handle these complexities.

In this work, we leverage the approximation abilities of neural networks to develop a single-level reformulation. By properly defining the linking constraints—those that map the neural network’s inputs and outputs to the follower’s variables present in the leader’s problem—just one neural network embedding can be used to represent all possible discrete decisions of the follower and replace the lower-level problem by a set of linear constraints that express the forward pass of the trained neural network. This approach eliminates the need for separate neural network models for each discrete solution, significantly simplifying the reformulation while ensuring that the follower’s decision space is fully captured within the upper-level optimization.

A case study on the optimization of a two-echelon supply chain is solved to demonstrate the viability of the proposed methodology and compare its performance against conventional solution techniques. Our approach allows a significant reduction in problem size compared to classical enumeration methods, where single-level reformulations require enumerating all feasible integer lower-level solutions and incorporating their collection of KKT conditions to the upper-level problem (Yue and You, 2016). Moreover, we proof that solution accuracy is maintained if and when the neural network is properly trained.

This data-driven approach offers a scalable and flexible alternative to conventional KKT-based methods and can be applied to a wide range of bilevel problems without requiring assumptions on the relative response property, thereby expanding the scope of traditional solution strategies in the field.

References:

• Jeroslow, R.G., 1985. The polynomial hierarchy and a simple model for competitive analysis. Math Program 32, 146–164. https://doi.org/10.1007/BF01586088
• Moore, J.T., Bard, J.F., 1990. The Mixed Integer Linear Bilevel Programming Problem. Oper Res 38, 911–921.
• Yue, D., You, F., 2017. Stackelberg-game-based modeling and optimization for supply chain design and operations: A mixed integer bilevel programming framework. Comput Chem Eng 102, 81–95. https://doi.org/10.1016/j.compchemeng.2016.07.026
• Yue, D., You, F., 2016. Projection-based Reformulation and Decomposition Algorithm for A Class of Mixed-Integer Bilevel Linear Programs. Computer Aided Chemical Engineering 38, 481–486. https://doi.org/10.1016/B978-0-444-63428-3.50085-0