Testing Wireless Sensors, Networks for the Smart Grid
Big changes are taking place in energy production, distribution and consumption.
By Jonas N. Olsen
Big changes are taking place in energy production, distribution and consumption. Fuel sources are changing toward renewables, and grid load needs to be more carefully managed and compensated with storage devices. It is becoming increasingly important for this new energy structure to be properly connected and intelligently managed. Key components for achieving this connected grid are wireless systems to transmit data from different end-points in the grid and the wireless modules and sensors those systems require. Testing integration and deployment models for these systems is the first step toward implementing a smarter grid.
Wide Area Wireless Networking
Utilities worldwide are currently evaluating their communications infrastructure to assess requirements for meeting future grid demands and government regulations. A large part of the focus thus far has been on smart metering, especially the electric meter. Creating a smarter grid, however, is about much more than hourly meter reads and downloads of consumer rate tables. Ultimately, a truly smart grid will be one that is connected and integrated from generation to consumption-where renewable energy is efficiently and seamlessly integrated and where intelligent networking throughout the grid enables early warning before failures, load balancing and more efficient grid planning.
A utility's capital assets and telecommunications are closely linked, from high speed broadband to sensor networks. Utilities are deploying a number of different communications topologies to meet demands. Some capital equipment requires high-speed data connections for applications such as SCADA and other low-latency links. Other applications can use cellular infrastructure. Due to cost and coverage issues, these topologies can connect only a small number of the overall utility endpoint devices. A different solution is needed to meet costs and coverage expectations of grid sensing applications and metering systems.
The answer lies in a wireless communication system that is designed for less data rate and latency sensitive applications in the grid, which include most grid end-points. Examples are electrical, gas and water meters, distribution grid sensors and substation and transformer monitoring sensors. Other examples are streetlight monitoring sensors and intrusion detection devices.
Most wireless systems addressing these applications employ some form of mesh technology. Mesh systems allow each node in the network to hop through other nodes to reach its destination. In this case the gateway or access point connects the wireless system to the backend application-such as a meter data management system. As such, the mesh technique is deployed primarily to overcome limitations in the range of the underlying radio technology. The primary complications faced by mesh systems, aside from range limitation, are the network coordination requirements, which create a drain on network capacity. As more time is spent coordinating the network routing between individual nodes, less capacity is available to transmit application payload-per occupied bandwidth unit.
A smart grid system should provide an extended range and be able to operate in a star topology, simplifying network coordination and leaving more capacity for application data throughput. This wireless system should reliably reach endpoints at four to six miles in suburban environments, limiting the network infrastructure required to cover vast areas. This lowers the overall network cost, not only for deployments, but also for operations and maintenance.
A system with an extended range must expand capacity too, as more endpoints will fall under the reach of each network infrastructure point. Because many relevant applications, such as transmission of an electric meter read four times a day, are highly duty cycled, utilities need to be able to configure tens of thousands of endpoints for each deployment. To provide wide area coverage, multiple infrastructure points can be deployed within the deployment area and work in a common network. These networks can provide coverage for industrial campuses, cities, counties or entire states.
On-Ramp, SEL Test Bed
To facilitate and speed up the integration process for its customers, On-Ramp Wireless is providing a test bed network in and around its offices north of San Diego, Calif. The network combines a number of access points installed on mountaintops around the On-Ramp headquarters and covers approximately 1,000 square miles. Once the endpoint wireless module has been integrated with the sensor or device, in this case a smart meter, the system can be deployed, provisioned and tested within the test bed network ahead of initial pilot deployments.
On-Ramp first tested this integration and deployment model in partnership with Schweitzer Engineering Labs (SEL), one of the first companies to integrate On-Ramp's Ultra-Link Processing (ULP) technology with its sensor application. SEL integrated the ULP eNode with its distribution grid sensor called the WSO-11 (wireless sensor for overhead lines). This sensor is designed to be deployed in the distribution grid and monitors current, voltage and ambient temperature. These measures are reported on a regular schedule, giving the utility a unique set of data for analysis of the distribution grid's current and historical condition. In addition, the sensor provides automated fault indication in the event of a grid failure, enabling faster fault location detection and quicker troubleshooting. Initially tested on the test bed network, this sensor is now in production and deployments are underway.
On the backend of the wireless network is a Web-based application that enables an easy platform for system verification and data visualization. This application is offered as an option in the development process. Other vendors have chosen to go directly to a third part application integration, such as a Modbus gateway for process automation applications.
Other applications have gone through the same integration process and are now destined for U.S. and global deployments. The test bed network has proven to be an invaluable tool in the business process as it speeds up the development and integration effort, and moves On-Ramp and its integration partners more quickly down the path of end-user acceptance.
About the author: Jonas N. Olsen is vice president of strategic marketing at On-Ramp Wireless. He has extensive experience in technology consulting and business development. Most recently, Olsen was vice president of business development at Blue Sky Network, a pioneer in global, satellite-based asset tracking.
CIRCLE 105 ON READER SERVICE CARD