Power-use monitoring smarter than ever
"We have got to stay relevant." Genesis Energy chief executive Marc England
An app that uses machine-learning to forecast how much power users are likely to spend on electricity in the week ahead has been released by Genesis Energy.
The software tool, Energy IQ, is initially being made available to more than 20,000 customers who signed up to trial new technology from the company.
Genesis Energy chief executive Marc England said it would add two additional features later this month.
One will let consumers see how their energy usage compares to similar households.
The other will attempt to break down people’s electricity usage and itemise how much they may be spending on activities such as hot-water heating and ‘‘always-on devices’’, based on their profile and their historical half-hour metering data.
Those forecasts are based on past consumption patterns, what customers have told Genesis about their household and electricity usage, and weather forecasts.
England said the forecasts would improve over time as the software applied machine-learning to understand why actual usage might have differed from its past predictions.
Power companies such as Genesis wanted to differentiate their electricity offerings, given the competitiveness of the retail market, he said.
England forecast in March that it was likely foreign technology and power companies would enter the New Zealand power market in ‘‘the medium term’’, citing the potential for competition from the likes Google, Apple and electric car and battery maker Tesla.
‘‘We have got to stay relevant’’, he said, and the app was just a first step.
In 2012, Genesis trialled plans that let customers pay different amounts for electricity based on the time of day, offering peak, offpeak and ‘‘shoulder rates’’.
Rival Flick Electric has taken that approach to its logical conclusion, charging its customers rates that change every 30 minutes, based on wholesale electricity prices at the time.
Genesis spokeswoman EmmaKate Greer said Genesis had found most customers preferred certainty. But she signalled it would again ‘‘explore and test various plan models with customers’’.
Lines company Vector has also turned to artificial intelligence and machine-learning to prepare for what it believes will be a new era in the electricity market. It sees homeowners conserving and trading electricity generated from their solar panels and stored in batteries and electric vehicles.
The company announced last week that it had expanded its partnership with Israeli technology company mPrest to develop and apply a machine-learning and artificial intelligence system that would better manage Auckland’s changing energy demands.
Chief executive Simon Mackenzie said the software could work out the best time to recharge electric vehicles.
It could also help minimise the impact of a local network outage by estimating what power might be stored and available in the area and where and when it could be discharged back into the grid.
England likened the appeal of solar home-generation to people growing their own vegetables, saying people valued considerations other than efficiency, such as ‘‘choice’’.