Published on March 13th, 2014 | by Joshua S Hill3
Teaching An Old Wind Turbine New Tricks
March 13th, 2014 by Joshua S Hill
Specialists working at Siemens Corporate Technology working with technicians from Technische Universität Berlin and IdaLab GmbH in the ALICE project (Autonomous Learning in Complex Environments) have developed self-optimisation software for wind turbines which will enable turbines to produce one percent more electricity annually under moderate wind conditions.
The software will “teach” turbines how to automatically optimise their operation in response to weather conditions, using sensor data to make changes to their settings based on wind speed and other factors.
Not only will this reduce wear and tear on the turbines, but it will help exploit the existing weather conditions to produce electricity.
The researchers behind the software will be demonstrating their work at the CeBIT trade show (March 10–14) in Hanover.
According to Siemens, the researchers have a demonstration wind turbine that is able to use its own operating data to gradually increase its electricity output. “The scientists’ approach combines reinforcement learning techniques with special neural networks,” Siemens said in a press release, a technology Siemens has been developing for several years in order to model and predict the behaviour of highly complex systems. Not only is this focused on wind turbines, but also for gas turbines, factories, and even stock markets.
Being able to instill such learning abilities in wind turbines will not only drive the technology forward, but it will go all the further towards making wind energy one of the most cost effective and cost efficient. Being able to draw more power from less wind is a goal many manufacturers are aiming for, and Siemens looks to have landed another step in that direction.
Keep up to date with all the hottest cleantech news by subscribing to our (free) cleantech newsletter, or keep an eye on sector-specific news by getting our (also free) solar energy newsletter, electric vehicle newsletter, or wind energy newsletter.