Synthesis of integrated processing clusters
Introduction
Process industries are challenged to become more sustainable. To a large extent, this requires better management of material and energy resources with the goal of achieving improved circularity of processing systems. Consequently, there is an increasing research focus in the field of process systems engineering aimed at enhancing processing for a Circular Economy (CE) (Avraamidou et al., 2020) and the development of eco-industrial parks (EIPs) (Susur et al., 2019). Walmsley et al. (Walmsley et al., 2019) have explained the importance of Process Integration (PI), i.e. holistic process systems design to minimise resource and energy consumption, in achieving efficient resource management across the levels of EIP and CE, and have observed that the overall benefits from PI increase the more “dimensions” (e.g. resources and processes) are considered for integration. This paper presents a PI method to holistically design integrated processing clusters that can account for candidate processes and all materials and energy resources involved.
PI has its roots in systematically minimizing process heat requirements and methods have evolved over the past decades to address optimisation of energy (heat, power) and material use efficiency. PI research has been extensively reviewed (Klemeš and Kravanja, 2013) and future perspectives provided (Klemeš et al., 2018). Of particular relevance to the synthesis of processing clusters are developments of PI methods for integration across processes, such as Total Site methods or inter-process process integration methods, which have been proposed for EIP design. Available PI methods for energy have evolved from considering heat integration across processes through steam exchanges (Stijepovic and Linke, 2011), steam and power integration (Hassiba et al., 2017) to consider multiple forms of energy from fuel to heat and power by integration through central utility systems (Klemeš et al., 1997) or decentrally (Stijepovic et al., 2012). PI methods for the integration of material resources have long focused on the integration of single resources across processes such as water (Alnouri et al., 2018), hydrogen (Alves and Towler, 2002), or CO2 (Al-Mohannadi and Linke, 2016).
Recent works have seen a shift to extend the dimensions of analysis by considering multiple resources simultaneously. Such works include methods for inter-process integration considering water and energy simultaneously (Gabriel et al., 2016), and a recently proposed multi-objective EIP water-energy network optimisation method using an analytic hierarchy process (Leong et al., 2017). Further simultaneous integration approaches have been developed for both CO2 and energy integration (Hassiba and Linke, 2017) and natural gas and CO2 (Al-Mohannadi et al., 2017). Another contribution has introduced an optimisation approach to identify minimum cost of mass and heat integration networks (Ghazouani et al., 2017). While these emerging approaches have moved away from considering single resources towards considering the integration of multiple resources simultaneously, they remain limited to only a few resources. Recognizing the need to more holistically consider rich sets of material resources, a more general high-level integration approach to C-H-O symbiosis has been proposed with a focus on enabling the balancing of Carbon, Hydrogen, and Oxygen atoms in integrated processing (Noureldin and El-Halwagi, 2015). The work has since been expanded and applied to synthesise networks with reduced CO2 emissions (Panu et al., 2019) and an EIP for CO2 conversion to achieve a maximum ROI (Al-Fadhli et al., 2018). In the C-H-O representation, the material exchanges are facilitated through source to sink connections, with stream compositions and conditions specified based on specific source and sink characteristics. The resulting optimization formulations take the form of (mixed integer) non-linear programs. Energy integration options to enable synthesis of EIPs considering the mass-energy nexus have not yet been considered in the approach. A number of recent works have analyzed and optimized process networks at the macro-level while considering the allocation of multiple feedstocks across chemical processes to yield diverse products. Derosa and Allen (Derosa and Allen, 2015) have explored national processing networks for the United States by optimizing the resource flows across a large number of over 800 processing plants in order to minimize the total industry cost in terms of production costs of important intermediates and end products. Kadampur and Kotecha (Kadambur and Kotecha, 2016) propose a piece-wise linearized model for production planning in the petrochemical industry to maximize overall profits by altering individual production levels. Calvo-Serrano and Guillén-Gosálbez (Calvo-Serrano and Guillén-Gosálbez, 2018) propose a superstructure network representation of the chemical industry to explore processing setups that minimize the total variable network cost in terms of raw materials and utility and to quantify their Life Cycle Assessment (LCA) impact. The same authors (Calvo-Serrano et al., 2019) proposed to optimize bio-process network (BPN) for utilizing biomass to make fuels, chemicals (ethylene), and electricity whilst considering economic and environmental constraints. Similar to the C-H-O optimisation representation, these process network optimisation methods employ simple input–output process models. Compositions and conditions of products and intermediates are not specifically accounted for in the optimisation models.
While recent works have significantly have enabled the identification of integrated processing network designs, there remain limitations in their ability to simultaneously explore configurations of all relevant material and energy resources across the many possible combinations of candidate processing options. One gap in the current state-of-the-art is the lack of simultaneous consideration of both material and energy resources in EIP design even in the most advanced C-H-O symbiosis network approaches (Noureldin and El-Halwagi, 2015). In addition, the resources to be considered in an integrated design effort would need to include all possible raw materials, intermediates, wastes and emissions, which goes beyond chemical compounds containing Carbon, Hydrogen, or Oxygen as considered in C-H-O symbiosis networks and in the advanced process network optimisation models (Derosa and Allen, 2015). In addition, methods to optimize EIPs do not consider quality specifications on products and intermediates in exchanges across multiple plants, which are important for enabling exchanges across processes in a cluster that are often owned by different companies. This paper addresses these gaps and shortcoming by introducing a new system representation and corresponding optimisation model that considers quality specifications on material and energy resources through their exchange between plants through resource lines. A network synthesis model is formulated based on the new representation that can determine the optimal design of clusters of processes by determining the most profitable combinations of processes and their optimal integration in terms of material and energy resource exchanges. In contrast to previous work, the proposed method is not limited by the number of material and energy resources that can be represented and allows to track all relevant resources across processing facilities of the cluster, including raw materials, intermediates, products, waste, emissions, heat carriers and power. The optimisation approach is enabled through the proposed compact cluster synthesis representation, which is detailed in the next section. This section is followed by the mathematical model formulation to enable optimal network synthesis to maximize overall profitability subject to constraints on emissions and production requirements. The use of the resulting optimisation method is illustrated with a case study in the synthesis of CO2 utilisation and storage clusters.
Section snippets
Processing cluster synthesis representation
This section outlines the proposed representation, which allows the tracking of multiple resources throughout a cluster and the selection of processes. Resources to be considered include material flows, including feedstock, products, intermediates and wastes/emissions, as well as energy flows, including heat carriers (e.g. steam) and power. In general, resources can enter the cluster as raw materials, leave as valuable products or wastes/emissions, or be exchanged between processes in the
Network optimisation model
The network optimisation model takes the form of a mixed integer linear program (MILP) with an economic objective and a number of equality and inequality constraints. Let R = {r1, r2, …., rn}, n ∈ ℕ be a set of resources, P = { p1, p2, …., pn }, n ∈ ℕ be a set of processes and C = { c1, c2, …., cn }, n ∈ ℕ be a set of components. The processing cluster network balances are developed considering both resource lines and processes. A resource line may receive external inputs, send or receive flows
Comments on model parameters
The proposed optimisation model requires multiple parameters to be specified by the user for a given case study. These include resource line specifications, resource prices, as well as performance and cost parameters for the processes involved.
For each resource line containing materials, the required resource composition (), pressure () and temperature () needs to be specified (Fig. 4). For resource lines corresponding to raw materials, the available raw material qualities are typically
Case study
The developed methodology is applied to a case study in CO2 utilisation and sequestration (CCUS). The design of integrated CCUS systems is becoming increasingly important in the context of global efforts to curtail CO2 emissions. A cluster is to be synthesised that receives a feed of up to 120,000 t/y of CO2, which is either to be converted to value added products and/or sequestrated. The requirement for the processing cluster is to convert at least 90% of the maximum available CO2 feed. Any
Conclusions
A method for synthesis of processing clusters has been presented. The method is built around a novel representation to allow the tracking of materials across processes through ‘resource lines’ associated with quality, temperature and pressure specifications. A resource line is included into the representation for each resource of interest in a particular study, including raw materials, intermediates, wastes/emissions and different energy resources. Processes are represented through input–output
CRediT authorship contribution statement
Razan Ahmed: Software, Validation, Formal analysis, Writing - original draft. Shaza Shehab: Software, Validation, Formal analysis, Writing - original draft. Dhabia M. Al-Mohannadi: Methodology, Validation, Writing - review & editing, Supervision. Patrick Linke: Conceptualization, Methodology, Validation, Writing - original draft, Writing - review & editing, Supervision, Project administration.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (45)
- et al.
On the synthesis of carbon constrained natural gas monetization networks
J. Clean. Prod.
(2017) - et al.
On the systematic carbon integration of industrial parks for climate footprint reduction
J. Clean. Prod.
(2016) - et al.
Accounting for central and distributed zero liquid discharge options in interprocess water network design
J. Clean. Prod.
(2018) - et al.
Methanol production via direct carbon dioxide hydrogenation using hydrogen from photocatalytic water splitting: process development and techno-economic analysis
J. Clean. Prod.
(2019) - et al.
Comparison between two methods of methanol production from carbon dioxide
Energy
(2014) - et al.
Potential and economic viability of green hydrogen production by water electrolysis using wind energy resources in South Africa
Int. J. Hydrogen Energy.
(2019) - et al.
Technology review and data analysis for cost assessment of water treatment systems
Sci. Total Environ.
(2019) - et al.
Energetic, exergetic and economic assessment of oxygen production from two columns cryogenic air separation unit
Energy
(2015) - et al.
An MILP model for simultaneous mass allocation and heat exchange networks design
Chem. Eng. Sci.
(2017) - et al.
On the simultaneous integration of heat and carbon dioxide in industrial parks
Appl. Therm. Eng.
(2017)
Carbon dioxide and heat integration of industrial parks
J. Clean. Prod.
Optimal production planning in a petrochemical industry using multiple levels
Comput. Industrial Eng.
Targeting and design methodology for reduction of fuel, power and CO2 on total sites
Appl. Therm. Eng.
Forty years of Heat Integration: Pinch Analysis (PA) and Mathematical Programming (MP)
Curr. Opin. Chem. Eng.
New directions in the implementation of Pinch Methodology (PM)
Renew. Sustain. Energy Rev.
Multi-objective optimization for resource network synthesis in eco-industrial parks using an integrated analytic hierarchy process
J. Clean. Prod.
CO2 footprint reduction via the optimal design of Carbon-Hydrogen-Oxygen SYmbiosis Networks (CHOSYNs)
Chem. Eng. Sci.
Power-to-fuels through carbon dioxide Re-Utilization and high-temperature electrolysis: a technical and economical comparison between synthetic methanol and methane
J. Clean. Prod.
Methanol synthesis using captured CO2 as raw material: Techno-economic and environmental assessment
Appl. Energy
Future cost and performance of water electrolysis: an expert elicitation study
Int. J. Hydrogen Energy.
Optimal waste heat recovery and reuse in industrial zones
Energy
Targeting and design of industrial zone waste heat reuse for combined heat and power generation
Energy
Cited by (24)
Development of a multi-leader multi-follower game to design industrial symbioses
2024, Computers and Chemical EngineeringOptimization of low-carbon hydrogen supply chain networks in industrial clusters
2023, International Journal of Hydrogen EnergyIncorporating negative emissions technologies with policy instruments for net-zero emissions
2023, Computer Aided Chemical EngineeringOptimization of low-cost negative emissions strategies through multi-resource integration
2022, Journal of Cleaner ProductionResource integration of industrial parks over time
2022, Computers and Chemical EngineeringCitation Excerpt :If an approach exists that can integrate more than one resource simultaneously, it is limited in the number of resources it can integrate or restricted to use within a specific type of industrial park. One work that addresses the multi-period integration of multiple resources is that of Abraham et al. (2021), which extends the approach developed by Ahmed et al. (2020) to multiple periods. The MILP proposed here considers capital investments to be a function of the operational capacity in each period.