Joel Höglund demonstrating smart house

ProjectNOBEL: Neighbourhood Oriented Brokerage Electricity Monitoring System

NOBEL: Neighbourhood Oriented Brokerage Electricity Monitoring System

Using wireless smart meter networks for power quality monitoring

The future of Smart Grids lies in standardisation and open protocols. By using the capabilities of modern smart meters, communicating over wireless radio, grid power quality monitoring can be improved while keeping costs low.

In NOBEL we have been developing middleware and services for the future smart electricity grids. One of the key aspects from a lower level networking aspect point of view is how to best make use of the more fine grained smart meter measurement data. We have studied how smart meter data can be used to complement and partially replace more expensive grid monitoring equipment, while still offering the same or better power quality monitoring. This is done using smart meters communicating over low power radio, making the smart meters form a wireless sensor network, WSN.

The NOBEL project is an EU FP-7 project. The partners are SICS from Sweden, UDE and SAP from Germany, ETRA and Alginet from Spain and CERTH from Greece.

Part of the work has been to explore and outline the business advantages that can be found in improved power quality monitoring. Improved monitoring gives electricity providers proof of power quality that can be delivered to residential users and be used to attract new customers. It would allow residential customers to monitor their own power quality. It offers better possibilities to do early problem identification and preventive maintenance. Finally, it could enable electricity providers to better optimize the voltage delivered, which may lead to reduction of energy up to 4% for Sweden which would correspond to 5.6 TWh in 2011.

The next generation WSN based tests have been performed at a testbed at SICS in Stockholm with the receiving enterprise services running at UDE in Duisburg. Larger scale business service user trials have been done with existing infrastructure in Alginet, a small town outside of Valencia, Spain. In the lab tests, the smart meters are modelled with relatively cheap, low power, hardware. As an important partial goal is to reduce the overall consumption, new infrastructure should not offset the savings by its own energy consumption.

By being based on the Contiki operating system, the smart meter middleware is using implementations of the wireless standards which have gone through interoperability testing across different platforms. Contiki is being supported by a growing number of hardware platforms, and has an active developer community.

Using the latest routing standard for wireless networks, RPL, the smart meter network becomes self-configuring. A newly installed device will automatically connect with the rest of the network, and acquire the needed network configuration data. This lowers installation costs, and makes adding new meters easier.

Based on local voltage deviation detections, smart meters can trigger events to alert the local grid operator. The events could either trigger a request to automatically increase the sampling rate in the affected area, or be forwarded to a grid operator for manual decision making.

As the radio packet sizes in WSNs are smaller than on the wired internet, fragmentation and address compression become important issues. This is handled by a 6LoWPAN layer which ensures the maximum packet proportions can be used for actual program data. Still, it is worth spending some design effort to keep messages – meter readings – short as to limit the delays of fragmentation which would increase network delays .

In our evaluations we have shown that the testbed could handle meters reporting data every 60 seconds. On top of that we made extra subscriptions when the system detected a voltage deviation, where a subset of the meters started reporting every 10 seconds to aid the tracing of quality problem origins. These results show that the suggested architecture is feasible with acceptable overhead. The power network will continue to need high cost network analysers to get instantaneous state information at critical points in the grid. Still there are benefits to be made by complementing the architecture with modern smart meters capable of delivering fine grained monitoring data.

The NOBEL project started in the spring 2010, and is in its evaluation stage in 2012. Security has not been addressed within the project, but for real life deployments the customer data needs to be protected, as unauthorized access to electrical usage data presents a privacy breach. Since the design is standard based, it is well prepared to be extended with IPSec, something we are currently investigating.

External homepage:
Number of items: 7.

Duquennoy, Simon and Wirström, Niklas and Dunkels, Adam (2011) Demo: Snap – Rapid Sensornet Deployment with a Sensornet Appstore. In: International Conference on Embedded Networked Sensor Systems (ACM SenSys 2011), 2011, Seattle, WA, USA.

Duquennoy, Simon and Wirström, Niklas and Tsiftes, Nicolas and Dunkels, Adam (2011) Leveraging IP for Sensor Network Deployment. In: Extending the Internet to Low power and Lossy Networks (IP+SN 2011), 11 Apr 2011, Chicago, IL, USA.

Dunkels, Adam and Eriksson, Joakim and Finne, Niclas and Tsiftes, Nicolas (2011) Powertrace: Network-level Power Profiling for Low-power Wireless Networks. [SICS Report]

Marqués, Antonio and Serrano, Manuel and Karnouskos, Stamatis and Marrón, Pedro José and Sauter, Robert and Bekiaris, Evangelos and Kesidou, Eleni and Höglund, Joel (2011) NOBEL - A Neighborhood Oriented Brokerage ELectricity and Monitoring System. Energy-Efficient Computing and Networking, 54 . pp. 187-196.

Höglund, Joel and Finne, Niclas and Eriksson, Joakim and Tsiftes, Nicolas and Dunkels, Adam and Chauvenet, Cedric and Pouillot, Mathieu and Goudet, Pierre-Emmanuel and Tourancheau, Bernard and Genon-Catalot, Denis (2010) Poster Abstract: Interconnecting Low-Power Wireless and Power-Line Communications using IPv6. In: 2nd ACM Workshop On Embedded Sensing Systems For Energy-Efficiency In Buildings (BuildSys), 2 Nov 2010, Zürich, Switzerland.

Vasseur, Jean-Philippe and Dunkels, Adam (2010) Interconnecting Smart Objects with IP - The Next Internet. Morgan Kaufmann. ISBN 978-0123751652

Rajaraman, Rajmohan and Moscibroda, Thomas and Dunkels, Adam and Scaglione , Anna, eds. (2010) Distributed Computing in Sensor Systems, 6th IEEE International Conference, DCOSS 2010, Santa Barbara, CA, USA, June 21-23, 2010. Lecture Notes in Computer Science, Computer Communication Networks and Telecommunications, 6131 . Springer Verlag. ISBN 3642136508

This list was generated on Fri Mar 23 04:04:32 2018 CET.