(455e) Using Multiobjective Optimization and Life Cycle Assessment for the Design of More Sustainable National and International Energy Systems

Sabio, N^a*., Volkart, K^b., Densing, M^b., Strachan, N^a.

^a UCL Energy Institute, University College London, 14 Upper Woburn Place, WC1H 0NN

^b Paul Scherrer Institut, Energy Economics Group, 5232 Villigen, Switzerland

Nowadays, an increasing energy demand driven by population and economic growth, and the need to reduce green-house gas emissions are the two major axis along which energy systems need to evolve. Not in vain, the design of sustainable energy systems has also been classified as a major challenge in the area of process systems engineering [1,2].

Energy systems models were developed after the first oil crisis using the methods of linear programming for informing policy makers concerned with the importance of energy systems planning and design at the time [3]. These models are still nowadays the main decision-support tools used for informing energy policy.

Although they were initially conceived for analysing energy security and costs, climate regulation has turned environmental factors in the major focus of energy systems analysis. For instance, after the UK committed to reduce its greenhouse gas emissions by 80% in 2050 [4], the Committee of Climate Change produced a set of carbon budgets based on energy systems analysis that outline the pathways and expected changes in the energy system needed to achieve the target [5].

Conflicting objectives lie at the heart of any design problem. Multi-objective optimization (MOO) frameworks were pervasive in PSE since the early days, the combination of multi-objective optimization frameworks combined and life-cycle analysis [6] has gained more recent attention [1,7,8]. Nevertheless, to the extent of our knowledge, this framework has not been applied to energy systems models.

One reason for this might be that multi-criteria analysis is not seen as a popular method in energy systems modelling community [9]. Although the advantages of multi-objective optimization have already been more widely recognised when applied to distributed generation problems involving a variety of stakeholders [10]. However, the challenges for the implementation of life cycle assessment in a whole energy systems model has been subject of more extensive attention in the literature [11].

In the present work we propose a framework for simultaneously optimizing several life cycle environmental impact indicators and the economic performance in energy systems models. For that purpose we use the classical ε-constraint deterministic multi-objective method that a side from being suitable for convex Pareto-sets, it allows to exploit the problem structure showing the different objective spaces without the need to articulate a priori preferences (i.e., weights), and avoiding the numerical problems arising from normalization.

Our results evaluate the effect of simultaneously optimizing several life-cycle environmental impact metrics against the more traditional total system cost optimisation with CO₂ emissions constrained for a newly developed national energy systems model of the UK (UKTM-UCL) [12] and for an international multi-regional energy systems model (GMM) [13].

*Corresponding author: n.sabio@ucl.ac.uk

References

[1] Grossmann, I.E., Guillén-Gosálbez, G. (2010) Scope for the application of mathematical programming techniques in the synthesis and planning of sustainable processes, Computers and chemical engineering, 34: 1365-1376

[2] Grossmann, I.E.(2012) Advances in mathematical programming models for enterprise-wide optimization, Computers and Chemical Engineering, 47: 2-18

[3] Stefan Pfenninger, Adam Hawkes, James Keirstead, Energy systems modeling for twenty-first century energy challenges, Renewable and Sustainable Energy Reviews, Volume 33, May 2014, Pages 74-86

[4] Acts of Parliament. Climate Change Act 2008. The Stationary Office Limited, UK

[5] The Committee of Climate Change (CCC) The 5th Carbon Budget: The next step towards a low carbon economy, November 2015

[6] Guinee JB, Goree M, Heijungs R, Huppes G, Kleijn R, de Koning A, van Oers L, Wegener Sleeswijk A, Suh S, Udo de Haes HA, de Bruijn H, van Duin R, Huijbregts MAJ. Handbook on Life Cycle Assessment. Operational Guide to the ISO Standards. Dordrecht: Kluwer Academic Publishers, 2002.

[7] Azapagic A, Clift R. (1999) The application of life cycle assessment to process optimisation. Computers and Chemical Engineering, 10:1509â??1526.

[8] N Sabio, C Pozo, G Guillén�Gosálbez, L Jiménez, R Karuppiah, Vasudevan, V., Sawaya, N., Farrell, J.T. Multiobjective optimization under uncertainty of the economic and life�cycle environmental performance of industrial processes, AIChE Journal, 60 (6): 2098-2121

[9] Hall, M.H. L., Buckley, A.R. (2016) A review of energy systems models in the UK: Prevalent usage and categorisation. Applied Energy, 169: 607-628

[10] Alarcon-Rodriguez, A., Ault, G., Galloway, S. (2010) Multi-objective planning of distributed energy resources: A review of the state-of-the-art. Renewable and Sustainable Energy Reviews, 14 (5): 1353-1366

[11] Volkart K. (2014).Development of a New Methodology for the Integration of LCA and Energy-Economic System Modelling, LCA XIV, San Francisco, USA

[12]UCL Energy Institute Models Website, Available at: https://www.ucl.ac.uk/energy-models/models/uktm-ucl [10 May 2016]

[13] Turton H., Panos E., Densing M., Volkart K. (2013). Global Multi-regional MARKAL (GMM) model update: Disaggregation to 15 regions and 2010 recalibration, PSI Bericht 13-03