A key smart grid concept is the notion that the sharp traditional distinction between electrical transmission and distribution will blur as local feeder systems become much more flexible and controllable, and as power is fed back into the grid from distributed generation and storage assets. Making distribution systems two-way and flexible represents an enormous engineering challenge, of course. But as it is met and surmounted, consumers will see tangible benefits.
Writing in a recent issue of Power Electronics, John McDonald, director of technical strategy and policy development for GE Energy Management in Atlanta, provides two vivid examples of the new challenges in electricity distribution. In San Diego, where well-to-do homeowners have installed photovoltaic arrays in large numbers along the coast, power output tends to spike around noon, when the morning fog has burned off. Since air conditioning is generally not required in the breezy homes, service panels usually are limited to 100 amps, so that a single service transformer feeds as many as 20 houses. “All these factors contribute to spikes in voltage variability when PV output is high, load is low, voltage regulation is electromechanical and, possibly, there’s high impedance on the feeder,” a source with Sempra Energy told McDonald, a designated IEEE smart grid expert.
In another example involving Sempra’s San Diego Gas & Electric, the utility is required to accommodate the needs of a large avocado farm, inland, where large PV arrays have been installed to help offset the power needs of the irrigation system. There too, output rises in the middle of the day, after fog has burned off, but the irrigation system is run mainly in the early morning. So, with net metering, “the farm injects substantial amounts of power onto the feeder, creating a voltage differential that results in reverse power flows,” the Sempra Energy source told McDonald. “That, in turn, leads to the maximum [transformer] tap, then lockout, of an upstream voltage regulator and has even affected voltage on the primary distribution line.”
Part of the solution, for SDG&E and for other utilities, lies in use of dynamic VAR devices assisted by power electronics—solid state devices that can modify current characteristics. (McDonald notes that in Europe, where government policies encourage high PV penetration, inverters to provide reactive power compensation are widely required.) Greater voltage variability in distribution systems can sorely stress transformers, and so “major vendors are doing R&D on prototype LTCs [load taps changers] employing power electronics,” writes McDonald.
Generally, it is proving helpful to gather information from the full range of smart sensors in a distribution system—from protective relays and feeder controllers to smart meters at homes—to monitor in real time how voltage is being affected by distributed assets. In Arizona, GE has helped a utility design and install a data collection system parallel to the SCADA system to detect and record voltage variability.
It’s a business in which not just the big established companies like GE, Oracle, SAP and IBM are active, but also a host of hard-charging startups like Opower in San Francisco and Space-Time Insight, in San Mateo, California. Space-Time, having developed systems to collect and process data from a wide variety of sources, has obtained contracts to enhance network intelligence in the Sacramento Municipal Utility District (SMUD), Ontario’s Hydro One, and the Czech Republic’s CES. The general expectation, as Rebecca Smith put in in a recent Wall Street Journal roundup, is that the improved power system intelligence will yield not only greater reliability but more timely and efficient repairs.