22 Sep 2022
22 Sep 2022
Status: this preprint is currently under review for the journal ACP.

Assessment of the wind energy resource on the coast of China based on machine learning algorithms

Boming Liu1, Xin Ma1, Jianping Guo2, Hui Li1, Shikuan Jin1, Yingying Ma1, and Wei Gong1 Boming Liu et al.
  • 1State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing (LIESMARS), Wuhan University, Wuhan, China
  • 2State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, Beijing 100081, China

Abstract. Wind is one of the most essential clean and renewable energy sources in today’s world. To achieve the goal of carbon emission peak and carbon neutrality in China, it is necessary to evaluate the wind energy resources on the coast of China. Nevertheless, the traditional power law method (PLM) relies on the constant coefficient to estimate the high-altitude wind speed. The constant assumption may lead to significant uncertainties in wind energy assessment, given the large dependence on a variety of factors. To minimize the uncertainties, we here use three machine learning (ML) algorithms to estimate high-altitude wind from surface wind. The radar wind profiler and surface synoptic observations at eight coastal stations from May 2018 to August 2020 are used as key inputs to investigate the wind energy resource. Afterwards, three ML and the PLM are used to retrieve the wind speed at 120 m above ground level (WS120). The comparison results show the random forest (RF) is the most suitable model for the estimation of WS120. As such, the diurnal variation of WS120 and wind power density (WPD) are then evaluated based on the WS120 from RF model. For land stations, the hourly mean WPD is larger at daytime from 0900 to 1600 local solar time (LST) and reach a peak at 1400 LST. This is mainly due to the influence of the prevailing sea breeze. On the contrary, the hourly mean WPD of island stations is relatively large at nighttime during 1800 to 2300 LST. This indicates that the wind energy peaks differ by the land surface types. In terms of the spatial distribution of the seasonal mean WS120 and WPD along the coastal region of China, the WPDs at Qingdao, Dayang, and Dongtou are higher than 200 W/m2 in most seasons, and the WPDs at Dongying, Penglai, Qingdao, and Lianyungang are much greater than at Fuqing and Zhuhai. The result shows that the coastal regions of Bohai Sea and Yellow Sea have more abundant wind resources than those of East China Sea and the South China Sea. These findings obtained here provide insights into the development and utilization of wind energy industry on the coast of China in the future.

Boming Liu et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on acp-2022-634', Anonymous Referee #1, 07 Oct 2022
    • AC1: 'Reply on RC1', Boming Liu, 05 Dec 2022
  • RC2: 'Comment on acp-2022-634', Anonymous Referee #2, 11 Oct 2022
    • AC2: 'Reply on RC2', Boming Liu, 05 Dec 2022

Boming Liu et al.

Boming Liu et al.


Total article views: 391 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
286 93 12 391 1 4
  • HTML: 286
  • PDF: 93
  • XML: 12
  • Total: 391
  • BibTeX: 1
  • EndNote: 4
Views and downloads (calculated since 22 Sep 2022)
Cumulative views and downloads (calculated since 22 Sep 2022)

Viewed (geographical distribution)

Total article views: 458 (including HTML, PDF, and XML) Thereof 458 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 24 Dec 2022
Short summary
Wind energy is one of the most essential clean and renewable energy in today’s world. The traditional measurement of wind usually uses meteorological masts equipped with anemometers, wind vane and other devices. However, the anemometer is usually installed at 10 m, while the shoreline wind turbine is usually installed at 100–120 m. Here, the radar wind profiler and surface synoptic observations are used to retrieve the wind speed at 120 m and investigate the wind energy resource.