谷歌的DeepMind AI项目可在涡轮开始旋转之前36小时预测发电量，从而让电厂能够提前一天确定向电网提供的输电量。

一年前，谷歌与DeepMind将机器学习算法应用于谷歌美国中部的一组风力发电厂。这些风力发电厂的风电容量达到了700兆瓦，能够提供一座中型城市所需的电力。

通过研究风力发电厂、天气预报以及涡轮的历史数据，经调试的DeepMind用于在实际发电之前预测风能发电量。此举也为风力发电厂的容量带来了更大的价值，因为发电厂可以提前对向电网的供电量进行规划。

DeepMind在一份声明中表示：“我们无法消除风力的波动性，但研究的早期结果显示，我们能够使用机器学习大大提升风力发电的可预测性和其价值。”

公司称，DeepMind的预测将谷歌风能的价值提升了约20%。（财富中文网）

译者：冯丰

审校：夏林

Google’s DeepMind AI program can predict wind power output 36 hours before the turbines start spinning, allowing one to make delivery commitments to the power grid up to a day in advance.

The news comes a year after Google and DeepMind applied machine learning algorithms to Google’s fleet of wind farms in central United States. With 700 megawatts of wind power capacity, these wind farms can generate as much power needed by a medium-sized city.

By studying the wind farms, weather forecasts, and historical turbine data, DeepMind was configured to be able to predict the wind power output ahead of actual generation. This brings greater value to wind farms’ capabilities, as power delivery to the grid can be scheduled in advance.

“We can’t eliminate the variability of the wind, but our early results suggest that we can use machine learning to make wind power sufficiently more predictable and valuable,” said DeepMind in a statement.

According to the company, DeepMind’s predictions have boosted the value of Google’s wind energy by roughly 20%.