Shulou(Shulou.com)11/24 Report--

Thanks to CTOnews.com netizen grass Luo Yuzi for the clue delivery! CTOnews.com, August 28, according to National Energy Voice official account, Longyuan Electric Power Engineering and Technology Company recently launched China's first wind turbine power curve image recognition AI model. Officials said, "it took the lead in realizing the automation and intelligence of wind turbine power curve analysis and screening, filling the gap in the industry."

▲ Source National Energy Voice official account CTOnews.com learned after inquiry that wind turbine power curve is an important index to assess wind turbine performance and unit generation capacity. Abnormal power curve will not only lead to power loss, but also reduce equipment power generation efficiency and shorten the operation cycle of components. The conventional methods of power curve screening rely on the personal experience of professionals, and the efficiency is not high and the accuracy is uneven.

It is reported that Longyuan Electric Power Engineering Technology Company has collected and labeled tens of thousands of abnormal pictures of typical power curves, independently trained the AI model based on mainstream image recognition models, and continuously optimized model algorithms and adjusted first-order and second-order optimization functions, realizing the identification of eight typical problems in power curves of different types of wind turbines.

▲ Map Source National Energy Voice official account in addition, the company has also applied the model to other wind turbine control maps to establish correlation models and continue to iterate. After the experimental verification of the 3-month operation data of the station, the model detected 1860 abnormal units among nearly 14000 wind turbines, with an accuracy of more than 80% and a work efficiency increase of more than 3 times.

The official said that this technology has realized the application of image recognition technology in the field of fan power curve screening and analysis for the first time, completed the automatic display of analysis results, effectively improved the location speed and fault analysis efficiency of problem units, and provided strong support for improving the power generation performance of equipment and promoting intelligent operation and maintenance of field stations.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.