Content uploaded by Wouter Nijs
Author content
All content in this area was uploaded by Wouter Nijs on Aug 19, 2015
Content may be subject to copyright.
A preview of the PDF is not available
Treating Uncertainty and Risk in Energy Systems with Markal/TIMES
Abstract and Figures
The objective of the TUMATIM project is to develop the MARKAL/TIMES modelling framework for a better assessment of energy and climate change policies. Climate change and security of supply, along with sustainable development have remained high on the agenda of the policy makers and this was reinforced by recent oil price crisis. The energy sector and the development and implementation of new technologies are crucial elements for the achievement of the EU and Belgian targets regarding sustainable development. The contribution of MARKAL/TIMES on these issues can therefore be important.
Uncertainty and risk linked to energy technologies, energy supply and climate change increases the difficulty of defining appropriate policies. The first objective of TUMATIM is to explore the portfolio approach to integrate uncertainty about fuel prices in the evaluation of energy and environmental policies and to implement it in the MARKAL/TIMES model. The starting point was to build a small portfolio technology model for the electricity sector. The quantification of the uncertainty regarding energy prices and climate change was done through probability distribution functions. The portfolio approach contributes to a better evaluation of the risk associated with specific technology choices. Its integration in MARKAL/TIMES allows to examine other dimensions than the currently implemented stochastic version which allows to define hedging strategy for waiting till the disclosure of information on climate change risk or on technology breakthrough.
A second part of TUMATIM relates to the price elasticities in the demand components of MARKAL/TIMES. These elasticities are another important source of uncertainty and the objective is to use modern insights of econometrics to estimate price elasticities of service demand or useful energy demand. “Useful energy” or “energy service” can be defined as the service that comes from the consumption of energy in various aspects of daily life, assuming no (or only very limited) preferences for the technological choices. Price increase causes energy efficiency improvements. Consequently energy services or useful energy decrease will be less than energy consumption.
Figures - uploaded by Wouter Nijs
Author content
All figure content in this area was uploaded by Wouter Nijs
Content may be subject to copyright.
... Regarding electricity markets, Musgens and Neuhoff [86] use stochastic optimisation to analyse the impact of the daily wind feed-in on dispatch decisions and the value of updating wind forecasts. Benoot et al. [87] develop the MARKAL/TIMES modelling framework to integrate uncertainty about fuel prices, climate policy and price elasticity of demand in the comprehensive assessment of energy and climate change policies in the EU. Fürsch, Nagl, and Lindenberger [88] use a multi-stage stochastic programming approach to optimise power plant investments along uncertain renewable energy development paths. ...
This dissertation is a compilation of four self-contained research articles that focus on selected subjects in the field of energy economics. The first article focuses on the competitiveness of offshore wind in mature markets. In this work, we harmonise auction results based on the auction design features. We show that offshore wind power generation can be considered commercially competitive in mature markets without subsidy. Furthermore, once auction results are harmonised, we observe similar expected revenue streams of wind farms across countries. This finding means that different auction designs can fairly reflect the actual costs of developing wind farms and thus translate cost reductions into lower bids. The second article explores the impacts of uncertainty in integrated electricity and gas system optimization models. We address the trade-off that the energy research community faces on a daily basis, i.e., whether to neglect uncertainty when constructing an energy system model and accept a suboptimal solution or to incorporate uncertainty and increase model complexity. Our research aims to bring a systematic understanding of which parametric uncertainties most substantially affect long-term planning decisions in energy system models. In the third article, we focus on seasonal flexibility in the European natural gas market. We develop a market optimization model to simulate the operation of the gas market over a long period. This allows us to explore structural trends in market development, which are driven by changing supply and demand fundamentals. Our work contributes to the methodological question of how to measure the contributions of different flexibility options. Finally, the fourth article investigates the value of Projects of Common Interest—gas infrastructure projects supported by EU public funds—in maintaining gas system resilience amid cold-winter demand spikes and supply shortages. For this purpose, we develop the first application of adaptive robust optimization to gas infrastructure expansion planning. The model endogenously identifies the unfortunate realizations of unknown parameters and suggests the optimal investments strategies to address them. We find that (i) robust solutions point to consistent preferences for specific infrastructure projects, (ii) the real-world construction efforts have been focused on the most promising projects, and (iii) most projects are unlikely to be realized without financial support.
There are different options to reduce CO 2 emissions: efficiency improvement, structural changes to low carbon or carbon-free technologies, sequestration and demand reduction. While the decarbonisation of secondary energy carriers and technological changes have an important role in climate change mitigation, this paper addresses the role of demand reduction. For this analysis, the elastic demand version of the TIAM-UCL global energy system model is used under different long-term low carbon energy scenarios through to 2100 at low, medium and high values for demand elasticity. The role of demand reduction is examined by means of decomposition analysis at regional and global level. The results of the emission decomposition indicate that a reduction in the demand for energy services can play a limited role, contributing around 5% to global emission reduction in the 21 st century. A look at the sectoral level reveals that demand reduction is very different from sector to sector with demand reduction contributing around 16% in the transport sector and about 2% in the residential sector at a global level. Analysis also finds that demand reduction can significantly affect the regional emission mix.
After an analysis of the causes of the present environmental crisis, this paper presents the tools available to solve this problem and the criteria according to which they must be chosen. These instruments are: voluntary restraint, obligations (including interdictions and limitations), taxation, subsidies, the selling or buying of rights by the public sector, direct exchanges and agreements between authors and victims of the nuisance, internalization of the externality. The criteria for the choice of means of public intervention are: 'social efficiency', knowledge of relevant data, welfare distribution, effects on public finances, practicability and implementation costs, and morality of the method. These tools are appraised and compared according to these criteria. In particular, the implicit or explicit allocation of environmental rights is emphasized, along with the importance of accompanying specific measures with compensatory financial transfers.
Evidence suggests that when the cost of travel changes, people take some time to adjust their behaviour and find a new equilibrium. Then the process of adaptation can be interpreted as a search for choices that increase the utility to be gained from cost reductions and reduce the disutility imposed by cost increases. The search for higher utility alternatives underpins the empirical observation that longer-run demand elasticities tend to be higher than in the short run. However, there is another implication, that there will be a bias if consumer surplus is estimated by the use of a static model. This paper shows that there is a special case where using a static elasticity does not give a bias. -Authors
Technology costs have been chosen in accordance with the cost–benefit analysis study for offshore wind. Input data have been composed with utmost attention and care, but the true future costs remain highly dependent on external factors. This chapter presents results of an application of Markowitz Portfolio Theory (MPT) to the future portfolio of electricity generating technologies in the Netherlands in the year 2030. Projections of two base-case generating mixes and general scenario assumptions have been taken from two specific scenarios designed by the Dutch Central Planning Bureau (CPB). Risk estimates were derived following a predefined methodology, and projections of long-term cost and risk for generating options specifically and portfolios at large remain difficult, even under the most up-to-date approaches. Furthermore, fuel correlations and technology parameter correlations are indicative and based on expert judgments. This chapter focuses on the electricity cost–risk dimension of the Dutch portfolio of generating technologies and the potential for additional deployment of renewable generating technologies to enhance the efficiency of base-case generating mixes in year 2030. The major results of this study are, in both scenarios, that the base-case generating mix is not very efficient. Graphical analysis suggests that diversification may yield up to 20% risk reduction at no extra cost; promotion of renewable energy can greatly decrease the portfolio cost risk and defining mixes without renewables results in significantly riskier mixes in the absence of concomitant significant changes in portfolio costs; because of its relatively low cost risk and high potential, large-scale implementation of offshore wind can reduce cost risk of the Dutch generating portfolio.