Today’s distribution grids -- the networks of power lines, transformers and other gear that carry lower-voltage electricity from substations to end users -- mostly lack the sensor and CIT infrastructure needed to “light them up” to grid operators, so to speak. Distribution automation (DA) technologies are aimed at filling in that dark space, with a combination of sensors, communications networks and software that can link it back to both existing and new utilities and grid operator control systems, to make it useful.
Sometimes that’s described as making the distribution grid more like the internet -- or perhaps, to take our ambitions back a notch, more like the modern telecommunications network that makes the internet possible. Tollgrade Communications, a Pittsburgh-area company with a 25-year history of managing both telecommunications and, more recently, smart grid networks, has taken an interesting hardware-plus-software approach to that challenge, and has utilities including Duke Energy and Toronto Hydro using it in North America.
On Wednesday, it announced its first European client, Western Power Distribution (WPD), one of the U.K.’s six Distribution Network Operators (DNOs), which manage big sections of the country’s grid. WPD, which serves about 7.7 million customers across the Midlands, southern Wales and Cornwall, is tapping Tollgrade’s hardware and LightHouse Sensor Management System (SMS) software for a grid sensoring and analysis project to help it plan for all the new distributed generation resources it’s going to have to manage as part of the U.K.’s carbon emission reduction goals. Financial terms of the deal were not disclosed.
Wednesday’s announcements come along with the launch of Tollgrade’s new 50-hertz version of its LightHouse Medium Voltage (MV) sensor, making it available for grid operators and utilities in Europe and much of Asia. Tollgrade already makes 60-hertz sensors for the North American grid.
GTM Research places Tollgrade amongst a long list of smaller DA technology providers such as GridSense, Utilidata, PowerSense, Survalent, Sentient Energy and others that have smaller, more targeted product offerings and far smaller revenues than grid giants like GE, ABB, Siemens, Alstom and Schneider Electric. The company, which had been publicly traded on the NASDAQ exchange for nearly two decades, was taken private in May 2011 by Golden Gate Capital, a San Francisco-based private equity firm, for a reported value of $137 million.
We’ve been highlighting companies in the DA space for some time, with an eye on differentiating just what each one does to help solve the overall smart grid puzzle, and how it relies on and integrates with other technologies to make the most of what utilities have to work with. Here’s our snapshot of Tollgrade, broken down by the company’s hardware and software.
Hardware: Low-Power Sensors and a TiVo for the Grid’s Waveforms
Tollgrade’s Lighthouse platform starts with its medium-voltage sensors, which both look like and differ from many of the other, similar devices on the market. In basic terms, Tollgrade’s sensors are inductively powered, meaning they use the energy of the line they’re snapped onto -- a trait shared by other grid sensor vendors like Sentient Energy and GridSense, to name a few.
But Erik Christian, Tollgrade’s vice president of smart grid, said that the company’s sensors have much lower power requirements than many competing devices -- down to 3 amps, as compared to a more standard 25 amps or so. That allows them to be placed farther down feeder lines where there’s less current, he said.
That’s important, because the farther away from the substation you get, the harder it tends to be to know what’s happening. (Of course, utilities that have deployed smart meters do have the ability to measure power at the very end of those lines, which helps fill in some gaps.)
From there, Tollgrade’s sensors measure load current, fault current, electric field strength, power factor, phase angle, sags, surges, wire temperature and harmonics -- a long list of important data points for utilities. Christian noted that some of these features aren’t necessarily standard on every grid sensor on the market, as they require a different set of technical requirements, both on board the sensor itself and in the data collection and analysis infrastructure that’s hooked up to them.
Tollgrade takes an unusual approach to this challenge, using technology it calls “waveform oscillography” -- essentially, taking digital snapshots and recordings of the alternating current (AC) waveform of the electricity on the power line to which it’s attached.
“The sensor itself acts like a TiVo, if you will,” Christian said. That’s a lot different than using a sensor to take various measurements of grid power quality, since that method doesn’t actually capture the original data, which is the sine wave of the current itself. We’ve seen other technologies that deal with real grid power in its true waveform state, ranging from sensors that capture it to better manage voltage regulation schemes, to power electronics that actually alter and control power digitally.
Capturing real-time waveforms is a pretty complex task, but Tollgrade uses software-defined sensors that can be customized to suit each utility’s communications requirements and constraints, Christian said. Some utilities may have high-speed comms that can handle lots of raw data, for example, while others may want each sensor to crunch most of the data themselves, and only send emergency alerts in real time, while saving up data for daily backhaul to the utility for longer-term planning needs, he said.
Software: Predictive Analytics for Situational Awareness
While all of this device functionality is useful, “What really makes the solution sing is marrying the hardware with the predictive analytics software,” Christian said. That’s Tollgrade’s LightHouse Sensor Management System (SMS), a platform built on the company’s telecommunications network management experience and fine-tuned for grid operations.
“We bring all the information from all sensors, not just three phases, but all the circuits and sensors at a substation,” via integration with other grid sensors out in the field, he said. All of that goes to Tollgrade’s SMS, which can perform analytics directly or deliver data to utility SCADA, DMS, OMS and other central software platforms.
But before it hands over that data, Tollgrade’s SMS subjects it to a long list of analytical functions to cleanse it of errors or false alarms, as well as make connections between disparate sensor data points and signals to predict what the actual, real-world problem is on the grid. That’s similar to the work Tollgrade has done in the telecommunications field, where its hardware informs network providers keeping track of their real-world communications assets, Christian said.
“Instead of just overwhelming the utility or telecommunications provider with every single piece of data, we categorize and filter, so the important data gets through,” is how he described it. For instance, one utility customer used Tollgrade’s software to discover that of the 87 faults that its existing fault detection system was counting, only 6 were real faults, while the rest were more minor current spikes, he said.
We’ve certainly seen this kind of approach applied to smart grid communications infrastructure, via network management system (NMS) providers like Cisco, Proximetry Networks and GridMaven. Combining telemetry with real-world grid status and control systems is a trickier task, but one being tackled by distribution management system (DMS) vendors like GE, Siemens, Schneider Electric, Alstom and ABB, a host of mid-tier players like Schweitzer Engineering Laboratories and S&C Electric Company, and IT giants like Oracle, SAP, IBM and Microsoft, to name a few.
Customers, Channels to Market
Tollgrade offers its hardware-software combo both as a turnkey solution for utilities, as well as a hosted services, Christian said. As for its own integration partners, the company works with Google Maps, which has been making moves into the smart grid field via partnerships with utilities and vendors like GE, for its geographic information system (GIS) interface, and its Linux-based software can also integrate with workforce and mobile applications, he said.
As for how it’s being used, the Electric Power Research Institute (EPRI) has published reports on Tollgrade’s work with big U.S. utility Duke Energy on its McAlpine Substation virtual power plant project (PDF), which includes lots of grid status data collection and management, as well as integration into Duke’s outage management system (PDF). Toronto Hydro is using Tollgrade for power line and transformer monitoring (PDF) as it struggles to integrate more and more wind and solar power into its grid, and CenterPoint has used its sensors as part of a distribution grid monitoring pilot (PDF) aimed specifically at solving the false positives/false negatives problems of previous sensor systems.
Tollgrade is also working in Latin America via a partnership with an undisclosed utility, Christian said. While most of its customers are big investor-owned utilities that own their own communications infrastructure, Tollgrade is also working on cellular solutions that “coupled with our hosted software, is pretty appealing to the smaller utilities, because there’s really no investment cost for them,” he added. Municipal and cooperative utilities are both growing and underserved markets, and are hungry for fault identification, transformer monitoring and self-healing grid systems to reduce their long-distance truck rolls and other maintenance costs.