Improving the ability of energy systems optimisation modelling to inform national energy policymaking
University College Cork
Energy Systems Optimisation Models (ESOMs) are extensively used to inform energy and environmental policymaking. They generate valuable insights into the possible pathways that reduce our reliance on fossil fuels and achieve ambitious clean energy transition goals. However, the academic literature identifies a number of priority areas for development with ESOMs to improve their ability to generate useful insights applicable to the energy transition. This thesis explores and delivers key developments in several of these dimensions: spatial resolution, energy-economy linkage, significance of model skill, and heterogeneity of consumers. From a policy perspective, this thesis seeks to improve the model-based analysis in the context of national-level energy sector decarbonisation and thus, mitigation policies are critically investigated. Moreover, the impacts of the mitigation actions on local air pollution levels and promoting energy security are also explored. Accordingly, the main contributions of this thesis are improvements to the state-of-the-art energy modelling methods and applications of the enhanced models to answer key policy questions with convincing evidence. The improvements are demonstrated via two well-established energy systems modelling tools in Ireland and Iran. The thesis concludes with several modelling and policy insights and suggestions on interesting areas for further investigation to strengthen the contribution of ESOMs to ensure improved climate mitigation and energy policies. The first weakness is the limited spatial and consumer granularity in ESOMs which constrains their ability to analyse region-specific energy transition pathways. This thesis develops a multi-regional representation of the transport sector within the TIMES-Ireland Model (TIM), an ESOM used to develop ambitious mitigation pathways for Ireland’s energy system. The multi-regional approach captures region-specific characteristics of transport technologies and infrastructures across 26 counties. It also incorporates the heterogeneity of the impact of air pollution in sub-national regions and estimates the ancillary pollution benefits of the mitigation targets in those regions. The spatially explicit modelling approach also reveals higher economic co-benefits than single region modelling. The single-region method masks the higher damage costs in medium and large cities, thus underestimating total benefits. This thesis also develops a multi-consumer approach, more accurately capturing consumer heterogeneity. Having homogeneous consumers in ESOMs tends to oversimplify purchase decisions, especially for capital-intensive technology adoption. TIM simulates vehicle purchase decisions using hurdle rates. This thesis disaggregates consumers into five groups, ranging from low- to high-income families, to incorporate a more realistic representation of their behaviour in vehicle purchasing decisions. The results demonstrates that the model with heterogenous consumers offers higher Electric Vehicle (EV) adoption than a single region model calibrated with average national data and identical consumers. Spatially explicit analysis presents valuable insights into regional EVs diffusion and their electricity consumption at a subnational level which are usually challenging to achieve through an aggregated national model. Secondly, ESOMs often ignore the effects of changes in energy costs on energy service demands, despite their key ability to balance supply and demand. The thesis addresses this by developing a comprehensive representation of the power sector within the MESSAGE model, an ESOM used to explore the impacts of different subsidy reform scenarios in Iran. The thesis develops a soft-linked framework combining MESSAGE with an economic model and analyses both supply and demand sides under harmonised assumptions. The novel soft-linking addresses the structural weakness of ESOMs in capturing the effects of energy price on demand. The hybrid model is used to investigate the impacts of subsidy removal on power demand and the required generation mix. The findings reveal that under an early and steady reform scenario, the system avoids lock-in effect, and thus the development of renewable energy technologies and energy efficiency plans become cost-competitive. By contrast, the late subsidy reform path even with radical removal fails to tackle the lock-in effect’s risk. On the other hand, the long-term energy system transition is deeply uncertain. The hybrid modelling framework in this research is also used to conduct an ex-post analysis exploring the extent to which electricity subsidy reform could have reduced Iran’s energy demand during the last three decades. To minimise the uncertainties, both energy and economic models are calibrated with three decades of historical data. The cost-optimal modelling results are then compared with the real-world transition, revealing a 50% lower cumulative cost in the subsidy removal scenario compared with the real-world transition. This deviation highlights what could have been achieved through the implementation of different policies in the absence of uncertainties, providing valuable insights for informing future policy initiatives. Finally, this hybrid framework is also used to show how synergies and efficiencies from Iran’s energy subsidy reforms and lifting its sanctions could enhance global energy security, with a focus on natural gas. It demonstrates that significant opportunities could be realised through a combination of national energy policy reforms and cross border cooperation in a favourable international environment.
Energy systems optimisation modelling , Natianl-scale policymaking , Spatial resolution , Consumer hererogeneity , Decarbonisation of power and transport , Energy-economy linkage
Aryanpur, V. 2023. Improving the ability of energy systems optimisation modelling to inform national energy policymaking. PhD Thesis, University College Cork.