By Corrie Goldman
Led by Professor Philippe Drobinksi, from École Polytechnique in France, an international research team is using experimental microgrids to develop management tools that make renewable energy a viable option for smart grids. With support from the Siebel Energy Institute, the interdisciplinary project created algorithms that manage uncertainty in all phases of the microgrid energy cycle - from energy storage to energy consumption.
Governments across the globe are calling for the use of more renewable energy sources. But renewable power, like solar, hydro, and wind can be unreliable, making it difficult for grid operators to integrate them into their systems.
With support from the Siebel Energy Institute, an international research team is using experimental microgrids to develop smart grid systems that can successfully run on variable renewable energy supplies.
Philippe Drobinksi, an associate professor of geophysics at École Polytechnique in France, is leading an interdisciplinary team that is running power use simulations on microgrids to help them develop management tools that will make renewable energy a more viable option for smart grids.
For the project, called “Trans-Disciplinary Approach for Renewable Energy Development (TREND),” Drobinski and his team developed algorithms that characterize, model, and forecast the uncertainties that arise in a microgrid that provides power where there are renewable energy sources and dramatic variations in power load and demand.
As Drobinksi’s research has shown, smaller is better when it comes to testing the renewable energy chain. The limited nature of a microgrid makes it exponentially more prone to disruption than a large grid. Even minor disturbances – like five homes using more power than anticipated, or three days of overcast weather – can cause the lights to go out.
From a research perspective, “it’s valuable to figure out how to integrate fluctuating and intermittent resources into a contained system,” said Drobinski.
Simulated electricity scenarios incorporated weather forecasts, geographically localized energy harvesting and storage systems, geophysical models, and electricity networks. The algorithms also took into account energy price analysis and socio-economic factors of the consumers. The integrated modeling tool allowed the team to analyze a range of “real” use cases and to run sensitivity experiments on various components of the renewable energy chain.
“The holistic end-to-end approach is the key for innovation in microgrids, and this is only possible with a true interdisciplinary research,” said Drobinski.
Importantly, Drobinski noted that the renewable energy chain is typically studied one component at a time (resource, energy conversion, storage, etc.), which leaves scientists without the full picture of how each phase impacts the other. By experimenting with all of the components in a single nanogrid system, Drobinksi said can he can “better address the uncertainty propagation in the chain and its impact on electrical grid management optimization.”
European utility giants EDF and ENGIE, and energy technology companies Itron and eLum, have been collaborating with the TREND team, providing guidance and support, given the exciting potential for both optimizing microgrids and scaling the technology for larger smart grid systems.
Foundation for Innovation
The project began with seed funding from the Siebel Energy Institute, which supported the construction of a desktop sized nanogrid capable of simulating how all components on a microgrid interact. The battery powered grid network includes one photovoltaic panel, emulated load data, and a secondary power source. On the nanogrid, researchers tested and developed algorithms that ensure a steady flow of electricity, even during extreme fluctuations in energy demands.
Using the nanogrid as a model, the researchers are building two functional microgrids – one on the École Polytechnique campus near Paris and the other in Tahiti at the University of French Polynesia, where Drobinski’s research partner, professor Pascal Ortega, will oversee experiments. The very different weather climates and consumption profiles of the two locations will allow researchers to test the microgrids under a wide range of conditions.
The École Polytechnique microgrid will build on the various components of a smart grid on campus that includes district heating and cooling, intelligent buildings, photovoltaic solar panels, smart metering, electric mobility, and hydrogen storage. The microgrid in Tahiti will be interconnected with the Tahitian power grid, which is mainly powered by five thermal power plants and hydroelectric stations. Such a grid, with few generation sources, is very sensitive to the integration of renewable energy sources that can induce frequency and voltage instabilities.
The modular nature of each microgrid will allow the researchers to test and then modify individual components. The experimental microgrids consist of a computer modeling algorithm for forecasting renewable resources based on both observation (now-casting) and numerical modeling (forecasting), innovative photovoltaic and wind electric conversion systems, a compressed air flywheel, and thermodynamic and battery storage systems. The systems also include optimization software for the management of the network, which accounts for the cost of electricity, consumption forecast, and the local electricity production.
Together, results from experiments will guide Drobinski’s team as they develop a range of flexible energy management tools that address everything from managing storage to managing energy consumption patterns with incentivized consumer programs.
These tools, said Drobinski, will be pivotal to “innovating technology for small smart grids, a.k.a. microgrids, and integrating renewable energy into smart grid networks.”
Drobinski outlined plans to build on the results of the TREND research in a project he named “Grid4Earth,” which aims to strengthen the systemic analysis of microgrids for improved management strategies by integrating bottom-up (vulnerability first) and top-down (resource first) approaches. A proposal for Grid4Earth has already made it to the second round of review for inclusion in the Université Paris-Saclay’s MISTIGRID consortium, part of a strategic industrial plan launched by the French government to develop new applications for smart and flexible management of smart grids. MISTIGRID project results will be made publicly available and will contribute to research and policy making in the context of French and EU policies and regulations on microgrids.
EDF, ENGIE, and eLum are among the numerous energy companies that have committed to being on Grid4Earth’s stakeholder board, should it be approved.
Because “the current market design is not adapted to microgrid research” ENGIE Global Energy Management COO Marc Pannier said that projects such as Drobinski’s are essential to creating “the appropriate technical and economic framework for the development of microgrids.”
Corrie Goldman directs communications and events for The Siebel Energy Institute, a nonprofit global consortium that funds cooperative research grants in data analytics and machine learning to accelerate advancements in energy science. The nine Siebel Energy Institute consortium member universities are: Carnegie Mellon University; École Polytechnique; Massachusetts Institute of Technology; Politecnico di Torino; Princeton University; Tsinghua University; University of California, Berkeley; University of Illinois at Urbana-Champaign; and The University of Tokyo .