KDnuggets Home » News » 2019 » Jan » Opinions » How AI and Data Science is Changing the Utilities Industry ( 19:n04 )

How AI and Data Science is Changing the Utilities Industry


Together, artificial intelligence (AI) and data science are causing positive developments for the utilities providers that choose to investigate these things. Here are some examples of technology at work.



By Kayla Matthews, Productivity Bytes

I've got the power

The utilities industry is a sector that relies on providing consistent access to things like electricity and water.

Otherwise, households and enterprise-level customers could discover that they can't access the resources that should help them live and do business.

Together, artificial intelligence (AI) and data science are causing positive developments for the utilities providers that choose to investigate these things. Here are some examples of technology at work.

 

1. Easing the Transition to Renewable Energy

 
People increasingly feel excited about the potential for renewable energy. In addition to sustainability benefits, some individuals want to take advantage of federal or state tax credits for products such as solar power equipment.

However, a successful transition from fossil fuels to renewable energy means analyzing crucial data points first. Utility providers must assess the output of current or planned renewables projects and couple that anticipated total with the likely demand brought about by all the customers using renewables.

Beyond using big data platforms to examine those two variables and others, it's possible to depend on AI to take a predictive approach to the metrics. Such a system may determine that renewable energy sources get used the most during certain hours of the day or that there are particular areas of a community associated with exceptionally high usage.

A global poll of 26,000 respondents found that the majority of them believe it's important to move toward a world powered completely by renewable energy. Another finding from the report highlighted how people want to do away with so much dependence on coal.

Clearly, consumer desire exists, but if utility companies don't peruse the data first, they could deal with unintended and preventable consequences. Worse still, if these problems make enterprises conclude that renewable projects failed, they won't be encouraged to invest in future opportunities to use renewables.

 

2. Preventing Power Outages

 
Power outages are prohibitively disruptive for customers and costly for utility companies to fix, but projects are underway to use AI — and specifically machine learning — to predict the conditions that cause power outages.

A 2017 project at a Department of Energy laboratory sought to identify weak points in the electric grid and proactively repair them before outages occurred. Ten utility partners participated in the multi-year effort, and the goal was to create an autonomous system that handles ordinary fluctuations in power, plus responds to major events like storms.

Similarly, a research team at Texas A&M University applied big data to improve tree maintenance and prevent overgrown branches from triggering lapses in electricity service. The traditional approach utility companies take is to schedule regular and rotating tree-trimming outings, but that method is both time- and cost-intensive.

The new data-driven option tells utility companies which trees are most likely to cause issues. Then, those entities can prioritize certain areas over others. The created model also accepts information from several sources. For example, a utility brand can populate the system with operational data or known statistics about vegetation and growth patterns.

 

3. Learning More About Customers and Their Needs

 
Consumers have access to a wide variety of high-tech thermostats that help them cut utility costs. Statistics show that 23 percent of households using the Ecobee3 smart thermostat save nearly a quarter of their energy bills with that model. Moreover, options exist that allow people to control their thermostats by sending text messages via their phones or set them to power-saving modes when the house is empty.

Also, provided that utility companies get customers' permission, they could access the data from those thermostats and use it to learn more about needs and usage patterns. Having access to the data could facilitate making meaningful tweaks to marketing messages, including mentioning how a new offering caters to identified characteristics associated with a customer.

Exelon, one of the largest utility companies in the United States, is appealing to customer needs with a chatbot. The company built a prototype in under two weeks, and the bot answers questions about power outages and billing concerns. Theoretically, a company could also apply big data to a chatbot's interaction history to make business decisions.

For example, if a higher-than-average percentage of customers had queries about their bills after those documents got redesigned, a company might revert to the old design or make changes that clear up the confusion.

Or, if an exceptional number of customers in a subdivision or apartment complex asked the chatbot when they'd get power restored after an outage caused by a storm, the company might decide to send its teams there first before other places get tackled.

 

4. Reducing Waste

 
If utility companies can't find effective ways to deal with waste, they'll likely have difficulties operating profitably. Data analytics platforms can illuminate areas of inefficiency, allowing providers to take prompt action and keep their costs down.

Similarly, those providers could offer branded apps that empower customers to cut down on waste by tracking how much water and electricity they use.

However, it's not easy to spot problems if they relate to hidden leaks. An initiative from the University of Waterloo applies AI to find them in water pipes by using sound-processing capabilities to monitor for sounds that could indicate dribbles of water. This system is especially helpful because it can even locate slow leaks equivalent to 17 liters — less than 4.5 gallons — per minute in lab tests.

Currently, the crews that deal with leaks get dispatched once substantial flooding becomes apparent, or they check pipes that could fail soon due to old age. But the University of Waterloo's AI-based method could lead to more proactive efforts that curb waste.

 

Enticing Reasons for Utility Providers to Implement Big Data and AI

 
This list profiles some of the most compelling use cases for big data, AI or the two together to achieve a singular goal.

However, as the technologies progress, there will be even more reasons for utility companies to invest in them to conquer known problems and improve operations.

 
Bio: Kayla Matthews discusses technology and big data on publications like The Week, The Data Center Journal and VentureBeat, and has been writing for more than five years. To read more posts from Kayla, subscribe to her blog Productivity Bytes.

Original. Reposted with permission.

Related:


Sign Up