MAIN PAGE : TITLE : Open source energy system models
draft page now at : http://en.wikipedia.org/wiki/Wikipedia_talk:Articles_for_creation/Open_source_energy_system_models
BACKUP MATERIAL :
Open source energy system models are energy system models that also classify as open source software. Energy system models are used to explore the operations and/or structural development of energy systems — and are often applied to questions of public policy. Model development is usually a team effort and typically constituted as an academic project or as a genuinely inclusive community initiative. The models themselves may be aimed at autonomous, municipal, national, and/or regional energy systems.
Open source energy system modeling is a relatively new activity. Indeed there are relatively few projects that pre-date 2010. One reason for their slow appearance was the fact that many models are written as mathematical programs and only recently have free languages, such as GLPK MathProg,[1] become available and popular.
Energy system models, in general, vary tremendously in terms of their type, design, programming, application, scope, and level of detail. This page does not attempt to classify the models listed here in any systematic way (that may come later). Suffice to say that most models simulate and/or optimize energy systems in order to investigate and improve their performance and/or reduce their impacts. Some models are specifically suited to volatile renewable technologies, others to municipal systems, and others to long-term national capacity expansion (or contraction). Some attempt to capture the demand-side with some realism, while others treat electricity, fuel, and heat demand as exogenous inputs. Models also vary in terms of their positioning on the engineering–economics spectrum and can variously take costs as exogenous, embed agent-based price discovery, or include a partial or general equilibrium economy. Models which span decades need to represent technical progress and may do so using calibrated learning-by-doing relationships.
|
General considerations
editAn energy system modeling project — particularly one aimed at supporting public interest analysis — is more than a coding project. There must be a broad consensus on the purpose of the model and its applications. And there must be access to suitable datasets. Moreover, good quality documentation will help address any public mistrust of computer modeling per se. Indeed, collaborative internet-based projects can and do "generate and maintain valued datasets, tools and educational resources".[2] The project repository — comprising the codebase, datasets, and documentation — may be hosted on institutional servers or on public code-hosting sites.
Projects vary markedly in their attitudes to membership. Academic projects have, historically at least, been limited to trusted individuals. Non-academic projects like OSeMOSYS have adopted the open software movement's ethos of inclusion. Open projects, in addition, normally offer mailing lists, forums, and wikis, as well as distributed source control and issues tracking features. The software and documentation licenses can also vary. The GNU GPL license is widely used for the source code and Creative Commons licenses for the documentation.
A number of programming languages have been deployed, including: Python, R, GAMS, MathProg, C++, Java, Matlab, Octave, and Mathematica. Proprietary languages (such as GAMS) tend to be used for academic projects, whereas their free equivalents (MathProg) are preferred for community projects.
Recent peer-reviewed surveys on closed and open source energy system modeling have focused on decentralized planning,[3] modeling methods,[4] renewables integration,[5] and the use of layered models to support climate protection policy.[6]
Current projects
edit| Project | Host | License | Membership | Coding | Documentation | Comments |
|---|---|---|---|---|---|---|
| Balmorel | Denmark | no explicit license | registration | GAMS | manual | energy markets |
| ETEM | CRP Henri Tudor, Luxembourg | Eclipse Public License v1.0 | registration | MathProg | manual | municipal |
| OSeMOSYS | OSeMOSYS community | GPLv3 | open | MathProg | website, forum | national planning |
| renpass | CSES, Germany | invitation | R, MySQL | manual | renewables pathways | |
| TEMOA | North Carolina State University | GPL | registration | Python | website, wiki | system planning |
| xeona | TU-Berlin, Germany | GPLv3 | not yet released | C++ | object-oriented |
Balmorel
editBalmorel is a market-based energy system model from Denmark. A GAMS license is required to run the model.
ETEM
editThe ETEM model offers a similar structure to OSeMOSYS but is aimed at urban planning. A manual is available with the software.[7] The model has been used to study climate protection in the Swiss housing sector.[8] Note too that GMPL, referred to in the documentation, is an alternative name for MathProg.
OSeMOSYS
editThe OSeMOSYS project is intended for national policy development and uses an intertemperal optimization framework. The model posits a single socially motivated operator/investor with perfect foresight.[9] A number of publications are available from the project website. Some of the studies have been conducted in sub-Saharan Africa. The role of household investment decisions was investigated in another study.[10] Uses will need to join the Commend community to gain access to the project.
renpass
editrenpass is an acronym for Renewable Energy Pathways Simulation System. The software is being developed by the Centre for Sustainable Energy Systems (CSES), University of Flensburg, Germany.[11] Participation is currently by invitation. renpass is written in R and links to an MySQL database. git is used for source control. There is a manual. A report on Baltic Sea region is available.[12] The software has also supported analysis for the renewables targets in Germany.[13]
TEMOA
editThe TEMOA project stands for Tools for Energy Model Optimization and Analysis. The model is programmed in Pyomo, an optimization components library written in Python. You need support for Pyomo to run TEMOA. The sourcecode is hosted on GitHub. The project also runs a website and wiki. TEMOA is "strongly influenced by the well-documented MARKAL/TIMES model generators".[14]
xeona
editxeona is an object-oriented energy systems model which spans several modeling paradigms: short-run optimization, scheduling, agent-based, and network economic. It combines microeconomic and technical processes at high resolution to evolve the system of interest over a representative year. Structural decisions necessarily remain exogenous and must therefore be handled using scenario analysis. xeona links to the GLPK mixed-integer solver. xeona has been written and tested but not yet released. It should become available in mid-2013.
Historical projects
editSome projects are no longer under active development but are still useful to list here.
| Project | Host | License | Membership | Coding | Documentation | Comments |
|---|---|---|---|---|---|---|
| deeco | TU-Berlin, Germany | GPLv2 | C++ | website, manual, publications | uses stranded programming libraries |
Programming components
editComponent models
editA number of technical models are now also open source. While these component models do not constitute systems models aimed at public policy development (the focus of this page), they nonetheless warrant a mention. Component models can be linked or otherwise adapted into these broader initiatives.
- Sandia photovoltaic array performance model[15]
A number of electricity auction models have been written in GAMS, AMPL, MathProg, and other languages.[16] These include:
- the EPOC nodal pricing model[17]
- vSPD nodal pricing model[18]
Open solvers
editMost projects rely on a mixed-integer solver to perform classical optimization, constraint satisfaction, or some mix of the two. While there are several open source solver projects, the most commonly deployed solver is GLPK. GLPK has been adopted by ETEM, OSeMOSYS, TEMOA, and xeona. Proprietary solvers outperform open source solvers by a considerable margin, so choosing an open solver may limit performance.[19]
Open data
editVarious national governments and the European Union are developing meta-data standards and putting key policy statistics and datasets online. This includes energy supply data and energy trading data. One key component is the SDMX Statistical Data and Metadata eXchange standard. Sponsors of SDMX include Eurostat and various UN agencies. The US Department of Energy publishes energy information for the United States. The availability of municipal energy data depends on data policies of the relevant city administration and utility providers.
Wikipedia itself contains a growing set of information about national energy systems, including descriptions of power plant fleets.
Closure
editOpen source development methods are now making inroads into the realm of public policy energy system models. That trend is likely to continue.
Notwithstanding, it remains to be seen whether established closed source public energy models will be transformed into open source projects. While some project members may favor this move, their sponsoring institutions tend to be nervous about revealing their models in detail and/or publishing their codebases for general scrutiny.
See also
editOpen source public policy models in other domains:
- GHGProof open source land-use model
Closed source energy models (for comparison):
- United States National Energy Modeling System
External links
edit- Expert system for an Intelligent Supply of Thermal Energy in Industry (EINSTEIN) project for single-facility analysis
- OpenEnergyMonitor open source energy use monitoring project
- SAM Solar Advisor Model project for evaluating photovoltaic installations
- TRNSYS Transient System Simulation Tool
References
edit- ^ GLPK MathProg is also referred to as GMPL or the GNU mathematical programming language.
- ^ Bazilian, Morgan; Rice, Andrew; Rotich, Juliana; Howells, Mark; DeCarolis, Joseph; Macmillan, Stuart; Brooks, Cameron; Bauer, Florian; Liebreich, Michael (2012). "Open source software and crowdsourcing for energy analysis". Energy Policy. 49: 49–153. doi:10.1016/j.enpol.2012.06.032.
- ^ Hiremath, RB; Shikha, S; Ravindranath, NH (2007). "Decentralized energy planning : modeling and application — a review". Renewable and Sustainable Energy Reviews. 11 (5): 729–752. doi:10.1016/j.rser.2005.07.005.
- ^ Jebaraj, S; Iniyan, S (2008). "A review of energy models" (PDF). Renewable and Sustainable Energy Reviews. 10 (4): 281–311. doi:10.1016/j.rser.2004.09.004. Retrieved 2013-03-02.
- ^ Connolly, David; Lund, Henrik; Mathiesen, Brian Vad; Leahy, Marti (2010). "A review of computer tools for analysing the integration of renewable energy into various energy systems". Applied Energy. 87 (4): 1059–1082. doi:10.1016/j.apenergy.2009.09.026.
- ^ Unger, Thomas; Springfeldt, Per Erik; Tennbakk, Berit; Ravn, Hans; Havskjold, Monica; Niemi, Janne; Koljonen, Tiina; Fritz, Peter; Koreneff, Göran; Rydén, Bo; Lehtilä, Antti; Sköldberg, Håkan; Jakobsson, Tobias; Honkatukia, Juha (2010). Coordinated use of energy system models in energy and climate policy analysis : lessons learned from the Nordic Energy Perspectives project (PDF). Stockholm, Sweden: Elforsk. ISBN 978-91-978585-9-5. Retrieved 2013-03-02.
- ^ Drouet, Laurent; Thénié, Julie (2009). ETEM : an energy-technology-environment model to assess urban sustainable development policies — Reference manual version 2.1. Chêne-Bougeries, Switzerland: ORDECSYS (Operations Research Decisions and Systems).
- ^ Drouet, Laurent; Haurie, Alain; Labriet, Maryse; Thalmann, Philippe; Vielle, Marc; Viguier, Laurent (2005). "A coupled bottom-up/top-down model for GHG abatement scenarios in the Swiss housing sector". Energy and Environment. pp. 27–61. CiteSeerX 10.1.1.111.8420. doi:10.1007/0-387-25352-1_2. ISBN 0-387-25351-3. Retrieved 2013-03-01.
- ^ Howells, Mark; Rogner, Holger; Strachan, Neil; Heaps, Charles; Huntington, Hillard; Kypreos, Socrates; Hughes, Alison; Silveira, Semida; DeCarolis, Joe (2011). "OSeMOSYS : the open source energy modeling system : an introduction to its ethos, structure and development". Energy Policy. 39 (10): 5850–5870. doi:10.1016/j.enpol.2011.06.033.
- ^ Strachan, Neil; Warren, Peter (2011). Incorporating behavioural complexity into energy-economy models. Retrieved 2013-03-06.
- ^ Zentrum für Nachhaltige Energiesysteme (ZNES), Universität Flensburg, Deutschland.
- ^ Bernhardi, Nicolas; Bökenkamp, Gesine; Bons, Marian; Borrmann, Rasmus; Christ, Marion; Grüterich, Lauren; Heidtmann, Emilie; Jahn, Martin; Janssen, Tomke; Lesch, Jonas; Müller, Ulf Philipp; Pelda, Johannes; Stein, Isabelle; Veddeler, Eike; Voß, David; Wienholt, Lukas; Wiese, Frauke; Wingenbach, Clemens (2012). Modeling sustainable electricity systems for the Baltic Sea region — Discussion paper 3 (PDF). Flensburg, Germany: Centre for Sustainable Energy Systems (CSES), University of Flensburg. Retrieved 2012-03-01.
- ^
German Advisory Council on the Environment (October 2011). Pathways towards a 100% renewable electricity system — Special Report (PDF). Berlin, Germany: German Advisory Council on the Environment / Sachverständigenrat für Umweltfragen. Retrieved 2013-03-06.
{{cite book}}: CS1 maint: date and year (link) - ^ DeCarolis, Joseph; Hunter, Kevin; Sreepathi, Sarat (2010). The TEMOA project : tools for energy model optimization and analysis (PDF). Raleigh, North Carolina, USA: Department of Civil, Construction, and Environmental Engineering, North Carolina State University. Retrieved 2013-03-05.
- ^ King, David L; Boyson, William E; Kratochvil, Jay A (2004). Photovoltaic array performance model — Sandia report SAND2004-3535 (PDF). USA: Sandia Corporation. Retrieved 2013-03-02.
- ^ MathProg is a subset of AMPL. It is sometimes possible to convert an AMPL model into MathProg without much effort.
- ^ Guan, Ziming; Philpott, Andy (2011). Modelling summary for the paper "Production inefficiency of electricity markets with hydro generation" (PDF). Auckland, New Zealand: Electric Power Optimization Centre (EPOC), University of Auckland. Retrieved 2013-03-01.
- ^ Naidoo, Ramu (2012). Vectorised schedule, pricing and dispatch (vSPD) v1.2 : a guide to the Excel-based interface. Wellington, New Zealand: Electricity Authority New Zealand. Retrieved 2013-03-01.
- ^ Koch, Thorsten; Achterberg, Tobias; Andersen, Erling; Bastert, Oliver; Berthold, Timo; Bixby, Robert E; Danna, Emilie; Gamrath, Gerald; Gleixner, Ambros M (2011). "MIPLIB 2010 : mixed integer programming library version 5". Mathematical Programming Computation. 3 (2): 103–163. doi:10.1007/s12532-011-0025-9. Retrieved 2013-03-05.