Practical Applications of AI for Today’s Power Companies

Energy experts around the world are talking more and more about artificial intelligence (AI). But it’s time to walk the walk. How can energy companies actually start putting AI theories into practice?

The energy sector is undergoing a fundamental shift. It’s facing daunting economic challenges, coupled with significant changes that are impacting traditional business models.

The grid is becoming increasingly decentralized, renewables and retail power are on the rise, consumers are disrupting the status quo by gaining more choice and control over their energy usage, and today’s industrial workforce and energy subject matter experts (plus the institutional knowledge they hold) are retiring at a rapid rate.

As if that weren’t enough, energy companies are also having to do more with less. Research shows that electricity generation O&M costs have soared by 74% over the past 20 years — an upward trend that simply isn’t sustainable.

Top three ways energy leaders are putting AI into practice.

For the above reasons, the energy industry has hailed AI as a source of hope. But theories can be a far cry from reality. What steps should energy companies be taking today with AI to optimize their businesses while preparing for the future?

1) Lower costs and improve the efficiency of operations.

It’s all too common for decision-makers to focus on technology first. However, this approach is problematic because it leads to debates and delays, as the path to driving business value is not clearly defined.

Instead, focus first on the outcomes you’re looking to achieve most immediately, such as:

  • Lower O&M costs while maintaining or reducing risk.
  • Increase overall plant reliability and meet power KPIs.
  • Gain cross-fleet visibility to make more informed business decisions.

Most power and energy companies have a wealth of data at their fingertips via historians, SCADA systems, CMMS, ERP and other systems — including spreadsheets kept on operation management’s desktops. With the proliferation and lower cost of machine sensors, the explosion of data can be daunting.

In other words, power companies are data rich, but insight poor.

Cleansing, correlating and discovering insights from this data is where AI can help. As an example, Uptake’s Work Order Cleansing AI Engine can take years of data, identify missing components, normalize and standardize the information, and present a detailed cost view of O&M activities down to the asset level. That can be a real eye-opener for operations executives.

Just this step alone can set in motion a change in maintenance strategies that can provide millions in annual savings for power leaders. Ameren is one such Uptake customer, as is Arizona Public Service (APS).

2) Free capital for more forward-moving initiatives.

Reducing O&M costs by improving maintenance strategies may be the first step, but power leaders can build on that success and integrate predictive analytics into their operations as well. The result of this action is that they gain deeper knowledge around the failure potential of assets under management.

Conversely, they gain a perspective on which assets are not prone to fail — despite OEM recommendations for refurbishment or replacement. Armed with data-driven insights, operations executives can work with their financial colleagues for more refined capital planning approaches.

Forestalling replacement of even a portion of a plant’s assets can result in freeing up capital for pressing initiatives such as grid improvements, expansion into new fuel sources or customer-facing applications — paving the way for new business models and strategies.

3) Set yourself up for the future and start using your data.

When data is collected and organized within one platform, it’s only the beginning. Suddenly you have access to mine and leverage this data for multiple purposes.

Imagine if you could precisely know, in real time, the energy usage across the grid? How about being able to accurately predict renewable contribution both from your wind assets as well as from consumer rooftop solar — hour by hour? Or what if you could more precisely bid power into the market, taking into account factors such as asset use (wear and tear), market pricing and customer demands?

This is the point at which business disruption occurs, and we’re not far from that point today.

At Uptake, we talk about how AI is re-coding the power industry. This isn’t just software, but fundamental business processes and models. Human interaction with systems — and with each other — will change. The ability to leapfrog institutionalized processes that have been around for decades becomes possible.

Taking the step to gather, mine and use your data today — and build with new information as you go — is the most practical step you can take to actually be the disruptor, not the disrupted.

How will you thrive in a new power market? It starts with a belief that AI is recoding energy as we know it.

AI was front and center at CERAWeek 2019, the energy industry’s top conference. Check out this video of Uptake’s president, Ganesh Bell, discussing AI with Carolyn Seto, Director of Upstream Technology and Innovation at IHS Markit.

Want to know how AI can help combat the rising O&M costs of energy? Dig into our research to learn about the financial value that’s at stake: