The Internet of Things is turning maintenance on its head.
The Internet of Things is transforming many industries, and electrical power industry is no exception. Thanks to an explosion of sensors combined with advanced computer modeling, energy utilities are changing the way they maintains assets, moving away from preventive maintenance to predictive maintenance and embracing predictive analytics to make smarter asset management decisions.
For energy utilities that have made the investment, early benefits have already begun to emerge.
How does predictive analytics work?
Predictive analytics forecasts possible outcomes. It offers a sort of crystal ball that can tell you things like which generation and transmission assets will need maintenance when, and which parts or systems are in danger of failing.
In a nutshell, software uses accumulated data about machine performance to create a model of normal operating behavior. Real-time data comes from sensors attached to equipment and IoT-connected devices that monitor instruments and systems. Algorithms compare the real-time data with the model and look for unexpected changes. For example, a slight voltage variation may disclose a problem with a substation capacitor bank.
Finally, machine learning provides the decision analysis needed to identify the best response.
Guiding maintenance decisions
If you knew which generation and transmission assets were going to be most in need of maintenance, you could prioritize that maintenance and even postpone maintenance on assets that, despite being on the maintenance schedule, are humming along just fine. Predictive analytics supplies those insights.
Software dashboards show load patterns, indicate the probability of an asset failure and analyze the root causes of problems with control systems, turbines, energy management systems and other plant processes. With this information in hand, supervisors can adjust work planning, prioritization and scheduling.
In addition to informing better decisions about resource management, predictive analytics can lead to faster responses to critical maintenance issues.
Predicting asset failures
Regular preventive maintenance routines may not catch subtle signs of underperforming older equipment that can lead to a breakdown. But predictive analytics can identify these problems, as well as critical conditions that can cause an outage, well before an outage occurs.
Take a gas turbine, for example. Decision tools based on regression analysis find patterns in large data sets and pinpoint factors that influence the operation of the turbine. If the software detects changes from the predicted normal operation, it issues an early warning about possible failures. Sensors may indicate that the heat rate and output power have surpassed an operating threshold, for instance.
Since the software pinpoints the location of the probable fault,technicians can make better decisions about the parts needed for the repair.
Predicting storm-related outages
When a storm looms, it’s hard to predict which assets might be damaged. But here again, predictive analytics can help. It uses data from SCADA, weather services, outage management systems and distribution management systems to build models that show potential weaknesses in the system. Predictive analytics also forecasts the path of the storm, its strength and its impact on the weak points. You can use these models to project the number and location of outages, adjust for changing storm conditions and prepare restoration resources to respond to customer needs.
After the storm, predictive software overlays accumulated information about storm impacts on data that shows impassable roads and downed lines. Having this composite information helps crews to restore power in several ways. Algorithms calculate the best routes for crews to restore power for the most customers. And the software matches the skillsets of work crews with the type of outage.
Everyone knows what happens when you make assumptions, which often prove to be wrong. With predictive analytics, companies can ditch the assumptions and make decisions based on up-to-the-minute insights from “big data” that no human could arrive at on their own.
John Ross has written about industrial, automotive and consumer technologies for 17 years.
Image Credit: Factory_Easy / Shutterstock.com