European Conference on Complex Systems - Satellite Workshop : 5th September 2012

Sustainable Energy, Complexity Science and the Smart Grid

Introduction and timetable

Professor Peter M. Allen introduced the workshop, presenting the framework for discussion and the timetable for the day as laid out in this presentation

Abstract and presentations

Below are the abstracts and presentations given by workshop participants (where available)

Rupert Gammon et al : CASCADE, Developing a Multi-Agent Model of the UK Electricity System

Presentation from Gammon et al, presented by Rupert Gammon

Abstract: The Complex Adaptive Systems, Cognitive Agents and Distributed Energy (CASCADE) project is developing a Framework based on Agent Based Modeling (ABM), which can be used both to gain policy and industry relevant insights into the smart grid concept itself and as a platform to design and test distributed ICT solutions for smart grid-based business entities. ABM is used to capture the behaviours – including those involving learning processes - of different social, economic and technical actors, which may be defined at various levels of abstraction, and their interactions. For example, CASCADE models 'prosumer' agents (i.e. producers and/or consumers of energy) and 'aggregator' agents (e.g. traders of energy in both wholesale and retail markets) at various scales, from large generators and Energy Service Companies down to individual people and devices. The Framework has a structure with three main subdivisions to contain linked models of electricity supply and demand, the electricity market and power flow. It can also model the variability of renewable energy sources caused by the weather, which is an important issue for grid balancing and the profitability of energy suppliers. The development of CASCADE has already yielded some interesting early findings, demonstrating that it is possible for a mediating agent (aggregator) to achieve stable demand flattening across groups of domestic households fitted with smart energy control and communication devices, where direct wholesale price signals had previously been found to produce characteristic complex system instability. In another example, it has demonstrated how large changes in supply mix can be caused even by small changes in demand profile. Ongoing and planned refinements to the Framework will support investigation of demand response at various scales, the integration of the power sector with transport and heat sectors, novel technology adoption and diffusion work, evolution of new smart grid business models, and complex power grid engineering and market interactions.

Enrique Kremers, José María González de Durana, Oscar Barambones and Andreas Koch : Towards complex system design and management in the engineering domain - The smart grid challenge

Presentation from Kremers, de Durana, Barambones and Koch, presented by Enrique Kremers

Abstract: One of the results of moving towards sustainability is a paradigm shift in the energy sector, characterized by the general trend of introducing new technologies and resources. This shift towards the smart grid requires the use of new tools, especially in the modeling and simulation areas. It has been shown [1] that complexity science could help to deal with the new challenges arising, which are mainly related to a more distributed system with large number of dynamic resources which will be more interconnected and communicating (ICT). Complexity science in the biological and sociological fields has mainly an explorative aim, which serves to analyze and better understand the behavior of Complex Systems (CS). However, man-made systems, have been engineered and designed to pursuit a certain aim. In the case of energy systems, the final aim is a reliable and efficient supply with energy for the final users. In order to fulfill this goal, not only exploration and analysis of the system are important. [2] introduces the term of controllability which aims to find the ways of influencing the behavior of complex networks. This is a step towards being able to design and manage CS in an efficient way. As the degree of complexity of a smart grid is increasing due to more and more frequent interactions and communications, this seems a promising approach. Techniques and methods gained in other disciplines related to CS can inspire new approaches in the engineering domain. Agent-based models as one of the tools to represent CS allow an integrative and systemic representation. Through some examples of modeling and simulation, we show how emergent phenomena such as oscillations in local demand side management systems arise. With these models it is possible to deal with the smart grid virtually, and test different control and management strategies.

[1] Viejo, P., et al. Towards an interdisciplinary approach for the simulation of future smart grid architectures from a complex systems science point of view. in European Conference on Complex Systems (ECCS). 2011. Vienna, Austria.
[2] Liu, Y.-Y., J.-J. Slotine, and A.-L. Barabasi, Controllability of complex networks. Nature, 2011. 473(7346): p. 167-173.

Marti Rosas-Casals, Lingen Luo and Ettore Bompard : Major events, intentional attacks and extended topological measures in transmission power networks

Presentation from Rosas-Casals, Luo and Bompard, presented by Marti Rosas-Casals

Abstract: Power grids have been considered complex networks for quite a long time now. Their structure and dynamics have been thoroughly studied and many topological measures have been used in order to classify them, evaluate their behavior in terms of robustness or model their dynamic response to malfunctions. In this sense, one way of classifying complex networks in general (and power grids in particular) is based on selectively attacking highly connected nodes and check global connectivity after every attack. Highly connected nodes, though, are not necessarily those more loaded or stressed in a grid. In order to overcome this simplification, several extended topological measures have been proposed to better account for the electrical characteristics of the network. On the other hand, results have been mainly theoretical and sound correlations between real grid’s dynamical behavior (i.e., malfunctions and major events) and any of the aforementioned measures have not yet been found. In this communication, electrical betweenness and entropy degree are used as extended topological measures in order to guide an intentional attack on four major European power grids: France, Germany, Italy and Spain. Similar behaviors are found between France and Spain, and Germany and Italy. We present also a first attempt to correlate these behaviors with real malfunction data in terms of both, aggregated and probability distributions of major events, identifying statistically similar dynamical responses among topologically similar grids. This would raise hopes in finding a more meaningful and significant linkage among power grids structural measures and real dynamical output (i.e., blackouts).

Peter M. Allen, Liz Varga and Mark Strathern : Models of Energy Supply and Demand Replacing Fossil Fuels

Presentation from Allen, Varga and Strathern, presented by Peter Allen

Abstract: A simple systems model of the UK energy system is presented which allows the exploration of possible futures under varied assumptions concerning the costs and impacts of different generation technologies. The model is distributed spatially over the UK and shows how different pathways exist that can take the UK to a 20% reduction in CO2 emissions in 2050, with a tripling of generation to 180GWs. This large increase in production results from assuming that both transport and heating will be electrified and that electricity generation will be done largely by renewables. This enables the exploration of the impact of different values concerning the desirability nuclear, wind or marine energy. It basically shows how excluding any of these leads to considerable difficulty in having enough low emissions generation capacity in 2050. It can help planning and shows how policy must try to make the powerful system agents adopt solutions that benefit the UK as a whole.

Peter Boait, Mark Rylatt and Richard Snape : Developing complexity in energy supply and energy efficiency - using simulation as a policy tool

Presentation from Boait, Rylatt and Snape, presented by Peter Boait and Richard Snape

Abstract: The provision of energy efficiency driven services rather than the supply of energy is in principle a preferable basis for business models that would be altruistic in their effects, especially in relation to climate change obligations. Through the effective decoupling of revenue from volume of supply the incentive to deliver policy goals becomes integral to the models and the need for complex, market-sensitive - and often ineffective - regulatory interventions would largely disappear. Transaction cost and the need for long term contracts to recover investment have been identified as serious barriers to the development and adoption of such models. However, in this paper it is argued that a unifying conceptual framework together with a platform of enabling technologies and standardised protocols are more fundamental requirements. In the context of the smart grid, it is shown how these could amount to the realisation of an artificial ecosystem as the most efficient way to exploit the solar energy gradient and how adoption of this coherent and unbounded exergy–seeking behaviour across the whole energy supply chain – engaging communities at different scales - should create the conditions in which novel and useful economic structures will emerge. It is explained how these will increase the dissipation of the gradient in accordance with the principles of non-equilibrium thermodynamics and how competition between community-focussed energy service companies will provide multiple “attractors” that satisfy the mathematical requirements for emergent complexity.

Samuel Thiriot, Enrique Kremers and Wolfgang Hauser : Generation of realistic consumer populations for electrical demand simulation in the context of the smart grid

Presentation from Thiriot, Kremers and Hauser, presented by Enrique Kremers

Abstract: Designing sustainable energy systems [1] involves the adaptation of the existing infrastructure: design of communication networks for gathering information about energy use and consumption, or adaptation of the old hierarchical electric networks to more horizontal and resilient ones. Agent-based simulations for energy systems require as an input both the structure of the system (a network or graph) and the attributes of the agents [2]; moreover, these agents have to be positioned on the structure according to their attributes. For instance, consumer types are also not distributed randomly over the distribution network. The topology of an electrical system has to be taken into account for its operation, therefore, we have to generate plausible networked populations. However, there is no trivial solution for positioning agents over a network according to their characteristics. Thus existing disaggregated models of residential energy consumption either do not describe the network (like [3]) or position the consumers randomly over the network (e.g. [4]). We investigate the impact of the positioning of residential consumers over the distribution network on the dynamics of the system. The attributes of individual consumers are generated according to survey data, the method used to position them over the distribution network [5] enables the exploration of several hypothesis: random distribution (base case) or heterogeneously generated population. The generated individuals are intended to populate an agent-based energy system model which has been applied in different case studies before [3, 6] and is able to represent the residential load curve based on a bottom up model in which appliances are modeled individually and managed by a sociological agent. In these simulations, the distribution grid allows to aggregate the electrical power generated by the consumption model and evaluate the impact of the consumption locally, for example at different branches of the network.

References: [1] Lopes, J.; Hatziargyriou, N.; Mutale, J.; Djapic, P. & Jenkins, N. Integrating distributed generation into electric power systems: A review of drivers, challenges and opportunities Electric Power Systems Research, Elsevier, 2007, 77, 1189-1203
[2] Harland, K.; Heppenstall, A.; Smith, D. & Birkin, M. Creating realistic synthetic populations at varying spatial scales: a comparative critique of population synthesis techniques Journal of Artificial Societies and Social Simulation, 2012, 15, 1
[3] Hauser, W., J. Evora, and E. Kremers. Modelling Lifestyle Aspects Influencing the Residential Load-curve. in 26th European Conference on Modelling and Simulation (ECMS). 2012. Koblenz, Germany.
[4] Paatero, J. and Lund, P. (2006). A model for generating household electricity load profiles. International Journal of Energy Research, 30(5):273–290
[5] Thiriot, S. & Kant, J.-D. Generate country-scale networks of interaction from scattered statistics The Fifth Conference of the European Social Simulation Association, Brescia, Italy, 2008
[6] Evora, J.; Kremers, E.; Morales-Cueva, S.; Hernandez, M.;Hernandez, J.J.; Viejo, P. Agent-based modelling of electrical load at household level. in ECAL 2011: CoSMoS - Proceedings of the 2011 Workshop on Complex Systems Modelling and Simulation. 2011. Paris.

Mark Strathern, Peter M. Allen and Liz Varga : Local Storage: Reducing the Need for Capacity Increase

Presentation from Strathern, Allen and Varga, presented by Peter Allen

Abstract: Cutting peak demand with local storage. This short paper shows how the existence of electricity local storage can cut peak demand considerably and constitutes in itself a ‘Smartness’ that is of considerable utility. Through quite reasonable investments at local levels, battery storage could be used to smooth local demand on the grid. This would mean that although demand may be expected to triple between now and 2050, this increase could be greatly reduced by local smoothing.

eccs_presentations.txt · Last modified: 03/10/2012 10:33 by rsnape     Back to top

Recent changes RSS feed Creative Commons License Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki