Systematic mapping of power system models: Expert survey

The power system is one of the main subsystems of larger energy systems. It is a complex system in itself, consisting of an ever-changing infrastructure used by a large number of actors of very different sizes. The boundaries of the power system are characterised by ever-evolving interfaces with equally complex subsystems such as gas transport and distribution, heating and cooling, and, increasingly, transport. The situation is further complicated by the fact that electricity is only a carrier, able to fulfil demand for such things as lighting, heat or mobility. One specific and fundamental feature of the electricity system is that demand and generation must match at any time, while satisfying technical and economic constraints. In most of the world’s power systems, only relatively small quantities of electricity can be stored, and only for limited periods of time. A detailed analysis of supply and demand is thus needed for short time intervals. Mathematical models facilitate power system planning, operation, transmission and distribution, demonstrating problems that need to be solved over different timescales and horizons. The use of modelling to understand these processes is not only vital for the system’s direct actors, i.e. the companies involved in the generation, trade, transmission, distribution and use of electricity, but also for policy-makers and regulators. Power system models can provide evidence to support policy-making at European Union, Member State and Regional level. As a consequence of the growth in computing power, mathematical models for power systems have become more accessible. The number of models available worldwide, and the degree of detail they provide, is growing fast. A proper mapping of power system models is therefore essential in order to: - provide an overview of power system models and their applications available in, or used by, European organisations; - analyse their modelling features; - identify modelling gaps. Few reviews have been conducted to date of the power system modelling landscape. The mission of the Knowledge for the Energy Union Unit of the Joint Research Centre (JRC) is to support policies related to the Energy Union by anticipating, mapping, collating, analysing, quality checking and communicating all relevant data/knowledge, including knowledge gaps, in a systematic and digestible way. This report therefore constitutes: - From the energy modelling perspective, a useful mapping exercise that could help promote knowledge-sharing and thus increase efficiency and transparency in the modelling community. It could trigger new, unexplored avenues of research. It also represents an ideal starting point for systematic review activities in the context of the power system. - From the knowledge management perspective, a useful blueprint to be adopted for similar mapping exercises in other thematic areas. Finally, this report is aligned with the objectives of the European Commission's Competence Centre on Modelling, (1) launched on 26 October 2017 and hosted by the JRC, which aims to promote a responsible, coherent and transparent use of modelling to support the evidence base for European Union policies. In order to meet the objectives of this report, an online survey was used to collect detailed and relevant information about power system models. The participants’ answers were processed to categorise and describe the modelling tools identified. The survey, conducted by the Knowledge for the Energy Union Unit of the JRC, comprised a set of questions for each model to ascertain its basic information, its users, software characteristics, modelling properties, mathematical description, policy-making applications, selected references, and more. The survey campaign was organised in two rounds between April and July 2017. 228 surveys were sent to power system experts and organisations, and 82 questionnaires were completed. The answers were processed to map the knowledge objectively. (2) The main results of the survey can be summarised as follows: - Software-related features: about two thirds of the models require third-party software such as commercial optimisation solvers or off-the-shelf software. Only 14% of the models are open source, while 11% are free to download. - Modelling-related features: models are mostly defined as optimisation problems (78%) rather than simulation (33%) or equilibrium problems (13%). 71% of the models solve a deterministic problem while 41% solve probabilistic or stochastic problems. - Modelled power system problems: the economic dispatch problem is the most commonly modelled problem with a share of approximately 70%, followed by generation expansion planning, unit commitment, and transmission expansion planning, with around 40‒43% each. Most of the models (57%) have non-public input data while 31% of models use open input data. - Modelled technologies: hydro, wind, thermal, storage and nuclear technologies are widely taken into account, featuring in around 83‒94% of models. However, HVDC, wave tidal, PSTs, and FACTS (3) are not often found unless the analysis is specifically performed for those technologies. - Applicability in the context of European energy policy: more than half of the mapped models (56%) were used to answer a specific policy question. Of the five Energy Union strategic dimensions, integration of the European Union internal energy market was addressed the most often (27%), followed by climate action (23%), research, innovation and competitiveness (21%), and energy efficiency (15%). This report includes JRC recommendations based on the results of the survey, on future research avenues for power system modelling and its applicability within the Energy Union strategic dimensions. More attention should be paid, for example, to model uncertainty features, and collaboration among researchers and practitioners should be promoted to intensify research into specific power system problems such as AC (4) optimal power flow. The report includes factsheets for each model analysed, summarising relevant characteristics based on the participants’ answers. While this report represents a scientific result per se, one of the expected (and welcomed) outcomes of this mapping exercise is to raise awareness of power system modelling activities among European policy makers.JRC.C.7-Knowledge for the Energy Unio