The mission of this group is to bring together utility professionals in the power industry who are in the thick of the digital utility transformation. 


AI System Helps Utility Companies Predict Energy Asset Failures

image credit: Photo by Alexander Popov on Unsplash
Jenna Tsui's picture
Energy and Sustainability Journalist The Byte Beat

Hi, my name is Jenna Tsui and I'm an energy, sustainability, and technology journalist located in Texas. I co-own a technology blog called The Byte Beat and I primarily write about technology...

  • Member since 2019
  • 11 items added with 30,710 views
  • May 28, 2020

Proactivity is one of modern technology’s greatest gifts. This is especially true in power generation and delivery. Monitoring infrastructure and assets in real-time to detect signs of faults and failures allows timely interventions to keep essential systems running.

Predicting equipment malfunctions and failures lets our lifestyles and productivity continue without interruption. That’s where artificial intelligence comes in.

AI unlocks a predictive maintenance model that’s especially useful in the utilities sector, where brief interruptions to service mean lost profits and potentially even lost lives.

Why Utilities Need a Better Approach to Maintenance

According to original research from E Source, the annual loss for U.S. businesses due to power outages now stands at $27 billion.

Batch manufacturing posts the highest losses per facility, with around $150,000 per year for each site. Due to the sheer number of locations, offices post total losses of more than $7.5 billion each year.

In order, the most widely impacted industries include batch and continuous manufacturers, financial and digital services, health care facilities, grocery stores, offices and retail locations.

Given the high costs and inconvenience to people and businesses, utilities need a reliable way to monitor their infrastructure for signs of trouble long before they become equipment failures and power outages.

iPredict is one such solution for utilities. It’s a “first-of-its-kind” technology platform that could serve as a blueprint as utility companies learn to prioritize ongoing maintenance, alongside new disruptions like unprecedented wildfires, climate change and extreme weather.

​​​​​​​How iPredict Uses Artificial Intelligence to Predict Failures

PA Consulting, the creator of iPredict, specializes in digital transformations, corporate strategy, and the application of new engineering and technologies for business success.

The company bills its new platform as a “world first” in using AI to prevent power outages by predicting asset failure in the energy sector. In some cases, the prediction algorithm helps engineers spot problems up to two weeks in advance.

There are many advantages to having this type of foresight, including:

  • Dispatching repair crews to sites proactively
  • Keeping customers’ lights on during emergencies, including hospitals, care facilities, schools and other vulnerable locations
  • Eliminating the “run to failure” model and prolonging the life of physical assets
  • Keeping maintenance costs reasonable over time instead of dumping sums of money into large, intermittent capital improvement projects
  • Providing more accurate estimates for outages and repair wait times

Every organization has an obligation to its workforce and its customers to have a well-rounded disaster recovery plan.

California provides a case study. There, unprecedented wildfires and mismanagement of power line maintenance made the state’s energy infrastructure especially vulnerable to disruption and widespread outages. Now, California utility PG&E is facing bankruptcy, and it's looking at a $2 billion hit from spiraling economic damage.

Not every disruption can be prevented or gracefully responded to. Unfortunately for California’s energy customers, PG&E already had a history of letting equipment maintenance lapse, long before the recent Camp Fire devastated the state and its infrastructure.

Clearly, prevention is worth a pound of cure, especially in the energy sector. The good news is, AI platforms like iPredict are taking emergency preparedness and energy grid maintenance to impressive new places.

According to its creators, iPredict boasts an 80% accuracy rate in predicting which pieces of equipment will fail within the next two weeks. When the window is shortened to three days, the accuracy rises to 90%. 

In practical terms, it works like this: Physically traveling to and proactively inspecting the condition of a dispersed system like an electrical grid is a costly and time-consuming affair. Instead, iPredict gathers “all relevant” data from grid components to do what human repair crews cannot — monitor the whole system at once. It delivers maintenance or repairs before components fail and take down the power grid with them.

For homeowners, power outages are a mild to severe inconvenience. For a city’s traffic lights, fire stations, hospitals, rescue services and other vital infrastructure, uninterrupted access to power — or even minimal power outages — is literally a matter of life and death.

​​​​​​​Artificial Intelligence and the New Energy Industry

From lower maintenance costs to improved public safety and higher uptime for essential services, AI is a major breakthrough for the utilities sector. In the near future, AI and other technologies will make smart electric grids and severe weather resiliency the new normal across the energy industry.

Jenna Tsui's picture
Thank Jenna for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member
Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Jim Horstman's picture
Jim Horstman on Jun 2, 2020

Just to quibble slightly, you are tarring all of California with a PG&E brush. In any case unless iPredict takes into account the environmental factors that typically play into wildfires that it really would be of help.

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »