AUGUST 9, 2016

Silicon Valley Power kickstarts IoT-based predictive maintenance




Silicon Valley Power (SVP) is the municipal electric utility owned and operated by the city of Santa Clara, California, and provides power to over 53,000 residential and business customers. Their main in-town generating facility, the Donald Von Raesfeld Power Plant, is a 147 MW natural gas facility. Operating since 2005, it is a two-on-one combined cycle plant, employing two combustion turbines and one steam turbine.


Traditionally, SVP employed a monthly walk around program for vibration monitoring. This method, though widely used in the industry, has three main challenges. First, by collecting data only once a month, it limits the ability to catch problems that start to emerge between monthly rounds. Second, by not collecting vibration data on the exact same spot each time results in inconsistent data leading to potentially inaccurate inferences about machine health. Finally, the analyst reports were either lost in the pile of numerous emails or were too technical requiring specialized knowledge in vibration analysis to interpret them.


To overcome these challenges, SVP decided to conduct a trial with Petasense in early 2016. With the promise of wireless and machine learning based analytics, Petasense Motes were installed to monitor some of the critical water injection pumps and SPRINT pumps. The water injection pumps are used to lower combustion temperatures and control nitrogen oxide emissions in the turbine generators. The SPRINT pumps increase the gas turbine performance by cooling the combustion air.


The benefits of using Petasense were apparent in just a few days. The wireless Motes made it easy to collect continuous vibration data instead of once a month. The Motes were permanently mounted on the machines facilitating consistent data collection not prone to manual errors. Petasense’s web dashboard gave the staff a holistic view of the entire plant’s health. Finally, Petasense’s Mobile app allowed them to receive instant alerts when a machine’s health status changed.

According to Paul Manchester, Division Manager of Generation at SVP, “Petasense’s solution was easy to get started with. The trial was installed in just a few days and I was immediately able to view a dashboard of the entire plant’s machinery. It’s great to get instant alerts on my iPad app when machine health changes. The weekly reports are easy to understand and I no longer have to deal with squiggly spectrum lines. Petasense has made it extremely easy to stay on top of all my balance-of-plant equipment.”

Petasense’s feature rich solution resulted in a tangible benefit for SVP. Within a few weeks after installation, the plant staff received an instant alert in their email about the deteriorating health of one of their Water Feed Pumps. Real-time detection of this fault was possible because of Petasense’s advanced machine learning technology that detected unusually high vibrations on the motor drive end. Empowered with this data-driven insight, SVP was able to activate a backup pump immediately, and safely shutdown the pump which prevented further deterioration of its health.

According to Paul, “We couldn’t be more pleased with the technology and the value Petasense provides. Their Vibration Motes quickly identified a failing machine and helped us avoid any unplanned downtime. The data was reliable, and the mobile software made it easy to stay updated on the condition of our pumps. In power generation, even a 1-2% improvement in equipment uptime can dramatically increase our revenues. Petasense has proved to be an innovative IoT company that can help us meet our goals.”

With roots in Silicon Valley, SVP is always looking for innovative ways to improve their efficiency and deliver safe, reliable power to their customers. By using Petasense’s predictive maintenance technology, SVP is at the forefront of embracing the latest innovations in sensing, wireless and big data analytics. Following the success of the trial, SVP is now in the process of implementing Petasense across their entire plant.