WSN Update

(CIO.com) - At Edison's Nuclear Generating Station in San Onofre, Calif., each time one of the large, '70s-era motors fails, the plant's power output drops between 20 percent and 25 percent for the three days it takes to fix or replace the equipment. The company then must spend as much as $400,000 to buy replacement power at emergency rates to meet customers' energy demands. Until now, engineers checked the motors' temperature monthly, hoping to predict which motors were about to fail so that they could preemptively rebuild or replace them during scheduled maintenance periods. Monthly manual readings, however, don't provide enough trend information to be particularly useful. "We did monthly readings and we missed a lot of things," says Lloyd Pentecost, maintenance engineer at the San Onofre plant. After two motors failed, it became clear that monthly monitoring was insufficient. "We increased to weekly readings, and it made no difference," he says, since readings can vary based on the time of day or outside temperature. Then Pentecost installed Wi-Fi-based sensors on eight motors to collect real-time trend data. "Once you get real-time data, the problem jumps out at you," he says. But Wi-Fi is expensive and poses a potential security risk. So Pentecost decided to try out wireless mesh networked sensors.

More...