Proving the Business Case for the Internet of Things

ABB and Verdigris machine learning predicts power peaks

Steve Rogerson
December 3, 2019



Swiss engineering firm ABB and Silicon Valley-based Verdigris Technologies have developed machine-learning algorithms to predict unplanned peaks in power consumption and identify strategies to prevent them.
 
ABB is deploying artificial intelligence (AI) to help commercial and industrial buildings improve their energy management and tackle rising electricity peak tariffs. The company has added two AI-powered applications to the ABB Ability EDCS electrical distribution control system. These cover energy forecasting and intelligent alerts.
 
ABB has developed the AI functions in partnership with AI specialist Verdigris Technologies as part of the company's Open Innovation programme. The energy-forecasting app will enable users to lower their electricity bills by reducing peak demand charges. The intelligent alerts app uses machine learning algorithms to help users better manage their assets, identifying underlying issues before they become problems.
 
"Our use of AI to help customers make better energy management decisions demonstrates ABB's commitment to innovation in our products and quality in our services,” said Andrea Temporiti, digital leader for ABB's electrification business. “With the new energy forecasting and smart alerts apps, AI drills down into the facility's power data to pinpoint actionable opportunities for productivity improvements and energy cost savings."
 
The Open Innovation programme engages incubators, accelerators, innovation centres and start-ups in the co-development and design of digital products and business models. The company is building an ecosystem of partners to work on digital energy management services for applications that range from smart buildings to e-mobility. The collaborations help start-ups develop services that can be marketed directly to ABB customers via its digital marketplace; the strategy also lets ABB customers benefit from digital technologies much sooner.
 
The energy-forecasting app uses AI to give facility managers accurate power consumption predictions. It helps them take timely action to reduce unplanned consumption spikes by re-scheduling or switching off non-critical loads, and taking full advantage of time-of-use tariffs.
 
The AI uses neural network methods to identify and learn patterns in a circuit or a building's energy consumption, while also factoring in weather data. Using weather forecasts and historical data, it can then predict power consumption for the next 24 hours, updating its forecast every 15 minutes.
 
"This innovative digital service makes it easy to take the necessary corrective actions to minimise any peak demand charges," said Temporiti. "The precision of the forecasting reduces hedging positions, narrows variability and produces meaningful energy cost savings for commercial and industrial buildings."
 
The intelligent alerts app uses machine learning to help users better manage critical assets. It learns how various factors affect the building and key assets so it can reduce the distraction of false alerts and information overload, allowing facility teams to focus their time more productively. It also identifies the relevant circuits and makes potential recommendations to ensure any response can be swift and decisive.
 
"Verdigris AI is ten times more effective than traditional energy management methods,” said Thomas Chung, head of product strategy at Verdigris. “Our partnership with ABB enables our AI capabilities to reach a significantly larger ecosystem of ABB users. These energy and asset management tools will cut through the noise to deliver actionable insights, identify real energy savings and make resource allocation more effective."
 
Verdigris Technologies is based in the historic Nasa research park in Silicon Valley, California, with offices in the USA and Taiwan. It is a privately held company backed by venture capitalists and has developed several products for commercial and industrial energy management.