Drones and AI combine to create predictive wind models for improved renewable energy solutions.
by DRONELIFE Staff Writer Ian J. McNabb
While scientists have struggled to accurately predict wind conditions, a Japanese company is working on what might be the secret to understanding atmospheric patterns, and it uses drones. The US Patent and Trade Office recently received an application from Japanese industry titan Mitsubishi Electric Co. (serial #202418746347) for a new UAV-based wind detection system that takes advantage of UAV’s ability to maneuver easily through the windstream to gather location, geodesic and wind-speed data, which then can be fed into a specially designed AI used to create more accurate and predictive wind models.
The goal of the project is to create systems that allow for more optimally-positioned wind farms, which involves a multistage (and multi-altitude) surveying process that involves knowledge of both what’s on the ground and what will be considerably above it. A drone, which can carry the proper sensors for both jobs, makes it a lot easier to calculate where a turbine could be safely positioned for maximum power output, leading Mitsubishi to integrate UAVs into their broader wind-prediction solution.
The full text of the patent (available here) includes a much more technical exploration of how the model works, but basically, the drone will use an AI-model to position itself and collect wind data, that will then be fed back into the model, creating a self-learning wind prediction system powered by UAVs. While we’re probably a few years away from seeing this technology actually brought to life, maybe, with the help of drones, the (famously capricious) element of wind won’t be unpredictable anymore.
The full text of the patent abstract reads as follows: “A wind condition learning device according to the present disclosure comprises an input unit (32) that receives input of a training data set, and an arithmetic unit (34) with an AI that performs learning on the basis of the training data set. One side of the training data set is a wind condition altitude distribution model value that follows a power law on the inflow side, and the other side of the training data set includes a wind speed average value, a wind speed maximum value, a turbulence energy, or a turbulence intensity in the wind condition distribution of an environment space obtained by simulation.”
More information on the patent, including authors, is available here.
Read more:
Miriam McNabb is the Editor-in-Chief of DRONELIFE and CEO of JobForDrones, a professional drone services marketplace, and a fascinated observer of the emerging drone industry and the regulatory environment for drones. Miriam has penned over 3,000 articles focused on the commercial drone space and is an international speaker and recognized figure in the industry. Miriam has a degree from the University of Chicago and over 20 years of experience in high tech sales and marketing for new technologies.
For drone industry consulting or writing, Email Miriam.
TWITTER:@spaldingbarker
Subscribe to DroneLife here.
