Data to Asset Health – Improving Asset Performance and Predicting Better Outcomes

Running an asset for as long as possible without a catastrophic failure is the ideal situation most engineers and operations managers hope for but going beyond the traditional useful life of an asset can be a risky game. At that stage the equipment operating costs are being traded for reliability and production output. Few would appreciate this better than the utilities who own and operate a very large network of assets, spread across hundreds or thousands of kilometers, operating on 24/7 basis with expectations for some of the highest reliability mandates. Luckily, engineers are getting better, with the help of online monitoring and data analytics, at better understanding asset health indicators with the ability to intelligently predict equipment at a higher risk of failure, taking out some of the guessing game. This will be a topic of discussion at an industry forum I will be hosting with experts from Hydro Ottawa, Burns & McDonnell and Siemens at the 31st IEEE’s Canadian Conference on Electrical and Computer Engineering (CCECE 2018) on May 14th in Quebec City. This whitepaper will discuss what are some of these innovative developments for better asset health.

Much of the electrical grid that was built in the rapid economic growth period after the second world war in the 1940s through the 1980s is reaching its useful life. Aging equipment brings higher failure rates and increased outages, affecting the economy and society. As well, older assets and facilities result in higher inspection maintenance costs, further repair and added restoration costs. Determining the sweet spot that asset life can be pushed to before reliability scores start to rapidly deteriorate and operating cost spiral out of control is always been difficult to predict. The best guidance engineers have had in addition to personal experience has been the performance curves by equipment type to determine the residual life. These performance curves have been developed based on average observed life expectancy of various equipment types over the years. 

Fig-1: Determining Asset Residual Life. Source – US Environmental protection agency (EPA)

To make an effective use of these performance curves it is been tricky to determine the current health of an asset. Traditionally, the age of an asset has been the dominant factor in making that determination. However, there are several other factors that play an important role that should be considered, such as loading profile, type and frequency of maintenance and inspections done, fault history, test results, weather conditions, environmental factors, make and model and observations of physical condition. Collecting, recording and mining that information across thousands or even millions of assets has not been a possibility until recently. With the advancement of technology and lowering of costs, today utilities can install networked instruments on major equipment such as transformers, circuit breakers and regulators that remotely monitor various operating parameters such as power flow, tap position, temperature, loading ratio and oil or gas pressure. As well, with proliferation of mobile devices with barcode scanners technicians can record and tag the results of equipment maintenance and test result so their values can be tracked against acceptable norms to flag any abnormalities that may warrant further inspection or issuance of a maintenance work order. If this information can be monitored and tracked for all major equipment, then at any time there’s a good indication of enterprise wide assets health. With data at your finger tips it is easy to start looking for patterns such as assets nearing end of optimal useful life before reliability metrics start being impacted or risk of catastrophic failure start becoming unacceptable. 

Fig-2: Distribution of Asset Conditions. Source - US Environmental protection agency (EPA)

In fact, using data analytics with intelligent algorithms analyzing various operating conditions, equipment failure scenarios and likely impacts it is possible to start predicting with some level of confidence potential upcoming problems that may crop-up for example on a hot summer day, heavy rain or during a high wind or ice storm. All this data and predictive scenarios equip engineers and utility executives with a dashboard to help them play various asset management scenarios to fine tune their approach to better align their operating practices and investment program with their corporate business objectives. It makes it easier to determine the return on investment (ROI) of various operating scenarios and investments options, for example, to refurbish, modernize or replace with new assets.

Date can provide a dashboard comparing ROIs of various operating scenarios and investments options

These developments are quite timely because most utilities are under intense pressure to maintain or lower consumer bills at a time when assets are aging and rates of failure increasing while consumers and regulator expectations for reliability are getting more stringent. And our dependence on electricity continues to go up. Citizens can’t live without charging their mobile devices and industries increasing reliance on automation means little tolerance for outage disruptions. And when an outage happens consumers demand instant updates, that they are used to getting from the technology providers, about what caused the outage and when the power will be restored. Having better data about your assets and their status makes providing useful updates easier.

The industry is responding in kind pouring considerable amounts of research and development by the technology firms in partnerships with the equipment manufacturers that promise better data capturing ability and improved predictions, especially as more equipment failure data is captured creating better performance history. Already, there are intelligent asset management solutions that are much more sophisticated from the ones that were available even five years ago and much of that development has been due to advances in data analytics.

More data and better analytics ultimately means better awareness of your assets and better ability to predict behavior and asset reliability while lowering operating costs and extending asset life. It is time to start paying attention to how intelligent your organization is about managing their asset health.

To read more on trending topics like this one, visit AMPLIFIED PERSPECTIVES. AMPLIFIED PERSPECTIVES

About Author: Ahsan Upal is a regional manager with Burns & McDonnell responsible for Canadian business development and leading engineering, project management and regulatory teams for major electrical distribution and transmission projects across Canada and the United States.