Superstructure-based Optimization Research Articles

Hybrid Optimisation Model for the Synthesis of Centralised Utility System in Eco-Industrial Park

Eco-industrial park (EIP) is a community of manufacturing and service businesses collaborating to manage environmental and resource (e.g. water, material and energy) issues together to preserve the environment while bringing financial benefit to the individual party. One way to achieve this goal is to integrate the EIP network via a centralised utility system (CUS), which facilitates the material and energy exchange among participating companies (or individual actors, IAs) in EIP. This work presents a hybrid model that combines superstructure-based optimisation approach and insight-based, automated targeting approach to synthesise a sustainable CUS for an EIP with the consideration of variations. The superstructure encompasses tri-generation system, wastewater treatment system and a steam distribution network that consider the use of mechanical drive steam turbine (MDST). Meanwhile, automated targeting approach is adapted to provide valuable insight and detailed allocation of resources within the network. In this work, the synthesis of CUS is analysed based on the economic performance and the total potential environmental impact of the system. Since such objectives are frequently contradictory in nature, fuzzy optimisation approach is adapted in this work to synthesise a sustainable CUS which fulfils the two important objectives. In addition, multi-period optimisation approach is applied to address the multiple operational periods with the presence of variations in supply of raw materials, market price of raw materials, demand for desired end products, etc. To illustrate the proposed framework, a palm oil processing complex (POPC) case study is solved.

Integration of carbon capture and sequestration and renewable resource technologies for sustainable energy supply in the transportation sector

In this study, a new design for a sustainable energy system was developed by integrating two technology frameworks: the renewable resource-based energy supply and the conventional (fossil fuel) resource-based energy production coupled with carbon capture and sequestration. To achieve this goal, a new superstructure-based optimization model was proposed using mixed-integer linear programming to identify the optimal combination of these technologies that minimizes the total daily cost, subject to various practical and logical constraints. The performance of the proposed model was validated via an application study of the future transportation sector in Korea. By considering six different scenarios that combined varying crude oil/natural gas prices and environmental regulation options, the optimal configuration of the energy supply system was identified, and the major cost drivers and their sensitivities were analyzed. It was shown that conventional resource-based energy production was preferred if crude oil and natural gas prices were low, even though environmental regulation was considered. Environmental regulation caused an increase in the total daily cost by an average of 26.4%, mainly due to CO2 capture cost.

Optimization of multi-effect distillation with brine treatment via membrane distillation and process heat integration

The objective of this paper is to introduce an optimization approach to the design of an integrated system of multi-effect distillation (MED) and membrane distillation (MD). Brine from the MED system is further treated in the MD network to enhance the recovery of fresh water and to reduce brine disposal. The desalination system is thermally integrated with an industrial facility to use excess process heat to reduce the cooling load for the industrial process and to decrease the heating load for the desalination system. A superstructure-based optimization formulation is developed and solved to determine the key design and operating variables. A case study is solved to illustrate the usefulness of the proposed approach.

Design and operation of renewable energy sources based hydrogen supply system: Technology integration and optimization

The benefits of hydrogen as a clean and efficient fuel can be realized fully only when hydrogen is produced from renewable energy sources (RES). In this study, a new methodology is proposed to identify the optimal configuration and operation of an RES-based hydrogen supply system. The key is a superstructure-based optimization model using mixed-integer linear programming (MILP), which can integrate multiple resources and various technologies, such as electricity generation (e.g., wind turbine, photovoltaic panel, and dish-stirling power systems) and hydrogen production technologies (e.g., water splitting using alkaline electrolyzer and biomass gasifier), to minimize the total annual cost. The performance of the proposed methodology was validated through an application study on Jeju Island, Korea. In the case study, process economics and main cost drivers were analyzed. In addition, the effect of integration of technologies along with different resources on the economics was discussed.

Synthesis of integrated membrane desalination and salt production networks

The development of desalination processes that are coupled with simultaneous salt production opportunities are increasingly receiving attention by the scientific community. This work is the first to propose a superstructure-based optimization approach that enables a systematic identification of high performance designs for such systems. The approach considers hybrid seawater reverse osmosis (SWRO) and nanofiltration (NF) membrane desalination together with salt production processes (SPPs). The proposed mathematical formulation has been developed based on a membrane superstructure, which mainly consists of SWRO, NF, and SPP as its primary synthesis units. The various connectivity options between synthesis units has also been explored in this work. For instance, SWRO and NF membranes follow slightly different connectivity patterns due to NF membrane's lower rejections for monovalent ions. The objective function identifies the minimum cost network embedded in the superstructure. The proposed method is illustrated with an example involving SWRO and NF membrane units as well as two different SPP options. The example highlights potential savings in water production cost as a consequence of salt value extraction. It further illustrates the benefits of integrated design of membrane desalination systems coupled with salt production processes over sequential designs of desalination processes followed by end-of-pipe salt production.

Optimization of multi-stage membrane systems for CO2 capture from flue gas

This paper applies mathematical programming and superstructure-based optimization approaches to design multi-stage membrane systems for CO2 capture, with the aim to systematically determine the optimal number of membrane stages, membrane areas, power requirements, the location of recycle streams, and operating conditions that satisfy desired CO2 recovery−purity value pairs ranging from 90 to 98% at minimum total cost. A superstructure that embeds several candidate configurations is represented as a NLP model which is used to investigate how these ranges of target values affect the optimal number of stages, membrane areas, operating conditions, and total cost.The results indicate that the optimal number of stages depend strongly on the desired CO2 purity. For CO2 purity ranging from 90 to 93%, the optimal configuration involves two stages and one recycle stream; from 94 to 96%, the optimal configuration involves three stages and two recycle streams; and finally, from 97 and 98%, the optimal configuration involves all the four stages proposed in the superstructure keeping the two recycle streams.For the same design specifications, the obtained process configurations compete not only with reference cases of multi-stage membrane systems in terms of total cost, but also with absorption-based CO2 capture processes in terms of energy requirements.

An integrated knowledge-based and optimization tool for the sustainable selection of wastewater treatment process concepts

The increasing demand on wastewater treatment plants (WWTPs) has involved an interest in improving the alternative treatment selection process. In this study, an integrated framework including an intelligent knowledge-based system and superstructure-based optimization has been developed and applied to a real case study. Hence, a multi-criteria analysis together with mathematical models is applied to generate a ranked short-list of feasible treatments for three different scenarios. Finally, the uncertainty analysis performed allows for increasing the quality and robustness of the decisions considering variation in influent concentrations. For the case study application, the expert system identifies 5 potential process technologies and, using this input, the superstructure identifies membrane bioreactors as the optimal and robust solution under influent uncertainties and tighter effluent limits. A mutual benefit and synergy is achieved when both tools are integrated because expert knowledge and expertise are considered together with mathematical models to select the most appropriate treatment alternative.

A superstructure model of an isolated power supply system using renewable energy: Development and application to Jeju Island, Korea

In this study, we aim to develop a superstructure-based optimization model using mixed integer linear programming (MILP) to determine the optimal combination and sizing for a hybrid renewable energy system to be used in an isolated area. The developed model has a three-layered energy structure to reflect the current reality in which energy production and consumption sites are generally separate. A variety of economic factors, including distance between facilities and an installation area, are considered for a more accurate estimation of the total annualized cost. Two types of optimization models, i.e., with and without a battery, are proposed to evaluate the economic and technical effects of a storage device to resolve operation issues caused by intermittent resources. An application case study on Jeju Island, Korea, confirms that the proposed model is suitable for decision making at the planning stage of a renewable energy system.

Model-based optimisation of biodiesel production from microalgae

This work presents a superstructure-based optimisation model to optimise the microalgae to biodiesel production flowsheet for the minimum net annualised life cycle cost (ALCC) of biodiesel. The model includes the important processing steps of converting microalgae into biodiesel, viz. microalgae growth, harvesting, lipid extraction, and transesterification of lipid. Different options to perform these steps are considered. The mass and volumetric balance for each process and equipment, and the equipment capacity limitations constitute the important model constraints. The decision variables include growth duration, medium, as well as the techniques and specifications to be followed in each of the downstream steps. The mixed integer linear programming model was applied to a case study of producing 30,000kg/d biodiesel from Chlorella. The minimum ALCC was US $ 13.286/l for the flowsheet and equipment details recommended by the model. Sensitivity analysis showed that lipid extraction was the most crucial step in the flowsheet.

Industrial wastewater treatment network based on recycling and rerouting strategies for retrofit design schemes

The advent of complex industrial water/wastewater management problems points to a need for effective systematic design for a sustainable solution. The objective of this work is to extend the research in the area of systematic design of water/wastewater management by further developing and extending a generic model-based synthesis and design framework for retrofit wastewater treatment networks (WWTN) of an existing industrial process. The developed approach is suitable for grassroots and retrofit systems and adaptable to a wide range of wastewater treatment problems. A sequential solution procedure is employed to solve a network superstructure-based optimization problem formulated as Mixed Integer Linear and/or Non-Linear Programming (MILP/MINLP). Data from a petroleum refinery effluent treatment plant together with special design constraints are employed to formulate different design schemes based on recycling and rerouting strategies focusing on completely splitting system and zero liquid discharge (ZLD) opportunity. The base case design of the existing process has been verified against the refinery data, while the grassroots and the retrofit options are generated and compared with the existing process. The network design solutions obtained with effectively computational time from the case study shows an improvement in the reduction of a total annualized cost (TAC) and wastewater discharge rate (WWDR) as a result of water recycling and rerouting options. Pareto plot (trade-off solution graph) for the analysis of such optimal solutions has been applied to implicitly verify the optimality of the solution based on all possible scenarios. Superior retrofit alternatives have been identified based on their performance including cost and environmental impacts and can be used as efficient design guidelines for the future development of the existing wastewater treatment process.

Automated process design of acid gas removal units in natural gas processing

This study looks at the design of the acid gas removal unit (AGRU) for natural gas processing. For the purpose of enhancing energy efficiency a number of different structural options are considered including multiple feeds, semi-lean and pump around modifications in addition to modification of operational parameters. Previous studies in this area have considered the comparison of different individual configurations but there has been a lack of research considering the simultaneous optimization of equipment configuration. Hence, in this study a superstructure-based optimization approach is used to simultaneously identify the most appropriate arrangement and operating conditions while the maximum energy recovery potential is also realized with the aid of energy composite curves (ECC). This methodology is applied to a case study where it is shown that the optimal configuration contains a combination of pump around and semi-lean process modifications allowing a 15.9% reduction of utility costs.

A hybrid approach for the efficient synthesis of renewable energy systems

An efficient synthesis method for renewable energy systems is presented that exploits synergies between heuristic- and optimization-based approaches. For this purpose, the RenewIslands method has been integrated into a superstructure-based optimization approach. The resulting hybrid approach consists of two steps: First, heuristic-based equipment preselection identifies a set of promising candidate technologies. Next, the preselected technologies are employed in superstructure-based optimization to determine the optimal renewable energy system. The heuristic preselection systematically avoids excessively large superstructures, while the subsequent optimization ensures that the optimal solution is selected. The proposed method is applied to the case of Mljet Island, Croatia. A renewable energy system is synthesized that requires 59% less investment costs compared to the solution derived by classical simulation. At the same time, the optimization problem is solved in less than 2min. The proposed hybrid method is shown to provide an efficient route to the synthesis of renewable energy systems.

Toward a Computer-Aided Synthesis and Design of Biorefinery Networks: Data Collection and Management Using a Generic Modeling Approach

Recent research into biorefineries resulted in many competing concepts and technologies for conversion of renewable biobased feedstock into an array of promising products including fuels, chemicals, materials, etc. The topic of this study is collection and management of the complex biorefinery data that are needed among others to support the superstructure-based optimization studies. To this end, we first formulate an integrated thermochemical and biochemical biorefinery superstructure and then use a generic modeling approach to represent each processing technology in the superstructure. The generic model parameters includes reaction yield, utility consumption, and separation efficiency among others, which are identified on the basis of input–output data (generated from rigorous models) collected from detailed biorefinery case studies reported in the open literature. The outcome is a verified database for the extended biorefinery networks combining thermochemical and biochemical platforms that represents ...

A hybrid optimisation model for the synthesis of sustainable gasification-based integrated biorefinery

Depletion of fossil fuels and increasing public awareness of environmental issues has stimulated the search for alternative energy sources. Biofuels are recognised as one of the most promising alternatives to fossil fuels, as they can be produced from various types of feedstock. The efficiency and sustainability of biomass-based production can be maximised by producing biofuels along with other valuable coproducts in a “biorefinery”. This concept was proposed to make the production of biofuels and biochemicals more economically viable by taking advantage of opportunities for process integration and waste recovery. In this work, a novel hybrid optimisation model that combines superstructure-based optimisation approach and insight-based automated targeting for the synthesis of a sustainable integrated biorefinery is presented. In addition, fuzzy optimisation is also adapted to synthesize such integrated facility with the simultaneous consideration of both economic and environmental performance. Note that the proposed approach is a generic synthesis strategy that can be applied even without detailed modelling of individual processes.

Optimization of Water Systems with the Consideration of Pressure Drop and Pumping

Novel automated design method has been developed to consider impacts of pressure drop and pumping arrangement in the water network in which water reuse is exploited to reduce freshwater consumption in process industries. A superstructure-based optimization framework has been developed to systematically consider all the key design issues simultaneously, including flow rate constraints, pressure-drop constraints, and other operational constraints, as well as to fully accommodate rigorous economic trade-off between freshwater cost, piping cost, and pumping cost. The proposed optimization study enables the identification of the optimal distribution of water within the network, together with the most appropriate location and capacity of pumps required for water systems. A robust and reliable solution strategy has been developed to effectively deal with computational difficulties associated with solving a mixed-integer nonlinear programming problem, with the aid of the effective application of physical insights gained from conceptual understanding of the design problem. The network complexity has been readily controlled by imposing design constraints in the optimization. Examples have been provided to prove the effectiveness of the design method developed in this paper and to demonstrate the importance of considering pressure drop constraints in practice.

Integrated water networks optimisation under uncertainty

An automated design method for the water networks under uncertain process conditions has been developed. To deal with the uncertainty problem, the installation of buffer tank and the installation of supplementing pipelines are systematically considered in the design of water networks. The cost-effective and feasible distribution and reuse of water within the network is identified from the superstructure-based optimisation model, with which all the operational constraints and fluctuating operating characteristics for water-using operations are fully and simultaneously reflected. The optimisation framework provides rigorous economic evaluation of the network through trade-off between operating and capital costs under uncertainty. The conceptual insight related to the design of water networks is incorporated in developing a reliable solution strategy which can effectively deal with the difficulty in solving the mixed-integer nonlinear programming problem. Case studies are given to validate the proposed model and demonstrate the importance of considering uncertain process conditions in the design and optimisation of water networks.

A superstructure model for the synthesis of single-contaminant water networks with partitioning regenerators

This paper presents a novel superstructure-based optimization model for the synthesis of industrial water networks with partitioning regenerators. Such regenerators function by splitting a contaminated water stream into a regenerated lean stream and a low-quality reject stream. Membrane separation-based processes are examples of these types of regenerators. The optimization model presented in this work integrates a single, centralized partitioning regenerator with a source–demand superstructure under the assumption that the processes within the plant are of the fixed flow rate type. The formulation is non-linear as a result of the presence of bilinear terms in the regenerator balance equations, but global optimal solutions can be found using commercial software. The features of the model are illustrated by solving case studies from the literature. It is notable from these examples that considerable design flexibility exists in networks of this type, since potentially both the lean and reject streams from the partitioning regenerator can be reused/recycled within the plant.