Clik here to view.
Predicting power demand used to be a simple science: People use more power during certain times — like the morning, when they cook breakfast and turn on their lights — and less during others, like when they hit the sack. Relying on predictable sources of electricity — like gas- and coal-fired power plants — utilities were able balance supply and demand with some fairly straightforward math based on historical records and other data.
But the steady rise of renewable energy made the power landscape infinitely more complicated. On the supply side, changes in wind or cloud cover can sharply shift the amount of power available. Demand has also become harder to nail down as more consumers manage their power use with smart thermostats and appliances like connected ACs.
At the same time, market forces demand better power forecasts. Power plants and fuel are expensive, and they don’t want to operate or buy more equipment than they may need. “In some countries, regulators are asking power generators to guarantee the quality of their forecasts,” says Olivier Cognet, CEO of Swiss-based startup Predictive Layer. “It’s no longer possible to say ‘We’ll sell you 20 turbines and see what they produce.’ It’s ‘We’ll produce x amount of energy by noon, y amount of energy in two hours and z energy in one month.”
Cognet’s company, a 2016 graduate of GE Digital’s Paris-based European Foundry startup accelerator program, is using the power of big data, machine learning and artificial intelligence to make better predictions about how much power will be generated and when. Predictive Layer starts with the basics, using a variety of consumption and production models to estimate supply and demand. But its software is designed to teach itself and fix its mistakes.
Clik here to view.
Energy forecasting may seem simple — people need more power during their waking hours and less overnight. But the rise in inherently variable power sources like wind has complicated matters. New machine-learning tools that can crunch mountains of data and project future energy output and demand can lead to more efficient and profitable outcomes for utilities. Image credit: GE Ecomagination
For example, the software incorporates data about how much power an area consumes and produces, integrating information from power plants — both fossil fuel and renewable — as well as from a host of other sources, including holiday calendars, sports schedules, weather forecasts and historical weather data. By correlating the data, the software can predict spikes and drops in usage and tie them to holidays or major sports games. The goal is to come up with the most accurate energy consumption predictions. In France, its energy usage predictions are already 0.79 percent more accurate than next-best predictions on the market.
That sounds like a miniscule fraction. But in the case of France’s energy consumption, it translates into 300 megawatts that doesn’t need to be produced. “That’s the equivalent of 300 wind turbines or a gas-fired power plant,” Cognet says.
Processing this volume of data would take an experienced mathematician a month, Cognet says; by the time they were done, the resulting energy forecast would be horrendously out of date. But with the help of Predix, GE’s software platform for the Industrial Internet, Predictive Layer can perform these calculations automatically, producing constant, self-correcting updates to its forecasts.
This isn’t to say that Predictive Layer is foolproof. While it learns from the past, it cannot predict something that has never happened before. So, when a disaster like a record-breaking storm hits, the program’s predictions suffer. But it’s able to constantly learn. As unforeseen “hundred-year storms” may become more common, this could become vital.
Ultimately, Cognet hopes, Predictive Layer will enable the industry — and all of its customers — to work with a much greater level of profitability and efficiency. “We’re not looking to disrupt the power industry,” he says. “We’re helping it to transform so it can do its job so much better.”