Page 243 - AI for Good Innovate for Impact
P. 243

AI for Good Innovate for Impact



                   Use Case 15: AI Enables New Energy Meteorological Power

               Forecast                                                                                            Change  4.2-Climate









               Organization: China Huadian Corporation LTD. & Huawei Technologies Co., Ltd.

               Country: China

               Contact Person:

                    Primary: Fang Peng, fang -peng@ chd .com .cn
                    Secondary: Hao Wen, wenhao23@ huawei .com


               1      Use Case Summary Table


                Item                 Details
                Category             Climate Change/Natural Disaster

                Problem Addressed    Under the overarching goal of global sustainable development, the
                                     installed capacity of renewable energy has been experiencing rapid
                                     expansion worldwide. Both regulatory assessments and electricity
                                     market operations now require renewable energy operators to achieve
                                     significantly enhanced forecasting accuracy and operational efficiency.
                                     Nevertheless, the inherent characteristics of renewable sources, includ-
                                     ing the randomness, volatility, and intermittency of wind and solar power
                                     generation, pose substantial challenges to power grid dispatching
                                     systems. These challenges simultaneously create pressing demands for
                                     the advancement of traditional numerical weather prediction models to
                                     better accommodate the integration of renewable energy into modern
                                     power systems.
                                     Traditional numerical weather forecasting predicts changes in wind
                                     speed and irradiance. This model, based on atmospheric dynamics,
                                     is complex and has not seen new breakthroughs in the past decade. It
                                     takes 6 to 10 hours to complete a prediction, which is time-consuming
                                     and has poor timeliness. The limited improvement in mechanism models
                                     and poor computational timeliness both affect the final prediction accu-
                                     racy.

                Key Aspects of Solu- 1)  AI Large Model for meteorological forecasting.
                tion                 2)  Cloud-edge collaboration architecture (Group meteorological
                                        model, regional micro meteorological engine, site edge power
                                        prediction service).

                Technology Keywords AI Large Model, Cloud-Edge Collaboration, Meteorological Forecast-
                                     ing, Renewable Energy, Power Grid Optimization, Electricity Trading

                Data Availability    Private











                                                                                                    207
   238   239   240   241   242   243   244   245   246   247   248