In late August, the Department of Energy released its long-awaited report on the reliability of the U.S. energy grid. Not surprisingly, it found natural gas the main culprit for the retirement of coal and nuclear power plants. But what’s missing from the report is what will disrupt the entire energy industry next: predictive analytics.
Thanks to low-cost sensors and cheap cloud storage, we can now gather significant amounts of data across the energy chain — from generation to transmission to consumption. That data, combined with data science, can generate insights and predictions to stop problems before they start, making machines more productive and reliable.
When it comes to energy, all sources will benefit from predictive analytics but none more than intermittent renewable energy sources like wind.
Just a few days after predictive analytics software began monitoring wind turbines in Iowa, a data science model spotted an anomaly in the wind turbine’s gearbox — one of the most important parts of a turbine. In response, the owner performed a few hours of predictive maintenance that cost $5,000. But, in the end, that work saved the owner from having a more catastrophic, $250,000 problem and days of downtime.
This small instance highlights how more production and reliability from all energy sources will leave the U.S. electric grid with fewer baseload plants than we have today.
First, costs across the industry will go down. Knowing potential problems before they begin eliminates more costly problems. Maintenance becomes more efficient too by having the right people at the right place with the right information at the right time. Renewables like wind, whose costs have dropped 66 percent since 2009, according to the American Wind Energy Association, will benefit the most from this and become even more competitive.
Not only will more energy be produced, the reliability of that energy will increase. Knowing you can count on energy generation assets with greater certainty creates new opportunities and less need for the redundancy baseload plants provide. When a wind operator can tell a local utility with greater certainty the amount of energy it will be able to produce, the need to maintain as large of a baseload drops. Greater reliability also allows renewables to participate and sell more in the day-ahead energy markets.
Predictive analytics isn’t just changing the generation of energy, it will change the way and how much energy we consume too. Large buildings are rapidly incorporating smart technologies into their operations, reducing costs and energy use on lighting, elevators and heating and cooling systems. All this means less energy used even as more buildings are built. In 2016, the Energy Information Administration found that nationwide electricity sales dropped by 1.3 percent in 2016 — the sixth time in nine years. This is at the same time as the country continues constructing more buildings, according to the architecture nonprofit Architecture2030.
Predictive analytics drives efficiencies and reliability for thermal sources just like renewables. Oil and gas will need those efficiencies to stay competitive in the marketplace. In the end, more efficient thermal generation, along with more renewables, will decrease overall CO2 levels and move us to a sustainable and cleaner energy future.
The change the energy grid has experienced is just the beginning. When DOE issues its next report, don’t be surprised when predictive analytics is listed as the disruptor.
Sonny Garg is the global energy solutions lead for Uptake, a $2 billion predictive analytics company in Chicago whose products make industries more productive, efficient, safe and secure.
Morning Consult welcomes op-ed submissions on policy, politics and business strategy in our coverage areas. Updated submission guidelines can be found here.