Electric power utilities have a difficult task. Despite aging infrastructure, the rise of renewable energy, and a shortage of skilled workers, they must continue to get as much out of their existing assets as possible.
In response, many have turned to data and analytics to improve operations, reduce maintenance costs, and extend the life of assets. Rapid advances in computing, networking, and storage technologies mean that utilities now have an abundance of data collected from a variety of different sources.
But as the volume of data has grown, so too has the complexity of the models required to bring everything together.
Today, the issue is not so much in collecting the data as it is in putting it to use. Without a clear business problem, utilities are left struggling to form a comprehensive, unified view of their entire system.
How Data Analytics Have Evolved At Utilities
Utilities understand the need for data. A recent survey found that more than 90 percent of utility leaders see data as crucial to their future success, and they have continued to embrace data across all levels of the organization.
Initially, utilities focused on the protection and control (P&C) systems needed to keep critical infrastructure running. Various sensors enabled the P&C teams to monitor the health and performance of assets and take immediate action to mitigate the risk of failure.
Over time, data and analytics were adopted into the distribution network. Smart grid technologies, especially smart meters, gave utilities better visibility, allowing them to coordinate distribution assets, operate them more effectively, and respond more rapidly to outages or sudden changes in demand.
From there, data was brought into the corporate side of the organization, with the insights being used to improve asset management, investment planning, and maintenance.
Most recently, the explosion of data has been driven by a push to implement artificial intelligence and machine learning (AI/ML) technologies, with IBM finding that 75 percent of energy & utility companies are embracing AI in their operations.
More Data - But Not Better Data
The push for more and more data shows no signs of slowing down. A recent market assessment projected that the global energy and utilities analytics market will grow to 4.3 billion USD by 2026.
But rather than improve the quality of the analysis, utilities are finding the volume of data to be overwhelming. Models become more complex, business cases become less defined, and the benefits of the investment become more difficult to achieve.
There are a few reasons for this. First, data has typically been siloed across the organization. As each department developed its own capabilities, it became difficult to integrate them together into a unified model.
Further, data from the P&C systems is transmitted on the Operational Technology network. By necessity, these networks are largely isolated from the rest of the organization, making it even more difficult to use the data outside the immediate protection of assets.
Finally, utilities are now competing with other industries for a limited pool of data analysts at the same time as many of their most experienced employees are retiring. As data analytics capabilities have become more critical, utilities have struggled to hire, train, and develop the right mix of skills among their teams.
Having been sold on the idea that data is the answer, many have simply found it too difficult to bring everything together into a comprehensive model.
Start With the Business Problem
Instead of trying to fix everything at once, utilities should take a step back and ask, “What is the business problem that we are trying to solve with data?”
Having a clear understanding of how the data will be used and what the implications are for the business makes it far easier to bring in the right data and develop a purpose-built model that answers that specific question.
For example, a utility may determine that it needs to enhance its asset management capabilities, focusing on high-value transformers spread across multiple remote locations.
Using this as a starting point, the utility can identify the most relevant information. The model might include performance and maintenance data, forecasted demand, the age of the equipment, asset utilization rates, thermal monitoring data, and repair records.
The data points chosen for other assets or business processes may be different - but that’s the point. By focusing on a specific problem, utilities can take a sustainable approach to analytics that allows operators to use the data available to them more effectively.
Managing the Health and Performance of High-Value Assets
Utilities act as the stewards of an incredibly complex system. They are responsible for building and maintaining a robust, reliable, and secure power system while simultaneously managing costs and keeping rates down for customers.
Data and analytics present an opportunity for utilities to make stronger asset management decisions, improve investment planning, and get the most out of their existing assets. But without a clearly defined business problem, utilities risk trying to solve everything at once.
By starting with how the data will be used to achieve a desired outcome, utilities can design purpose-built models that take the right data from the right sources. Instead of being overwhelming, data can play an integral role in achieving better results.