State Grid Henan Electric Power: Improve the Accuracy of New Energy Power Forecasting
After the winter, the northern Henan region, which is rich in wind and solar resources, has entered the stage of new energy high-power operation. On the morning of November 21, Li Wenguang, the dispatching squad leader of the power dispatching and control center of Anyang County Power Supply Company in Henan Province, used the new energy power forecast results released by the new energy meteorological monitoring and power prediction platform, combined with the actual operation of local wind power and photovoltaic power stations, to accurately carry out the power limit analysis of the power flow section of the power grid and the balance of new energy power.
Since June this year, State Grid Henan Electric Power has relied on its self-developed new energy meteorological monitoring and power forecasting platform and related technologies to innovate the application of the regional new energy total power prediction model and carry out new energy power forecasting in 18 cities across the province on a regular basis. Compared with the same period last year, the short-term and ultra-short-term prediction accuracy of new energy in the province has increased by 1.6 and 1.2 percentage points on average, which has effectively guaranteed the scientific dispatch of new energy and the safety and reliability of the power grid in the province.
At present, the cumulative installed capacity of Henan new energy has exceeded 50 million kilowatts, becoming the second largest power source in the province. The traditional new energy power prediction model cannot reflect the impact of weather change on the whole region, and there are problems such as low prediction accuracy, short prediction period and weak prediction autonomy to varying degrees. In February this year, the State Grid Henan Electric Power Organization scientific research team developed a new energy meteorological monitoring and power prediction platform and related technologies, the platform integrates D5000, the Central Meteorological Observatory, Henan Provincial Meteorological Observatory and other rich data resources, containing 6 modules, including meteorological monitoring, meteorological forecasting, power forecasting and so on, supporting the platform to use longer time series of meteorological forecasting to achieve longer period of power forecasting. Among them, the daily ultra-short-term, short-term and long-term forecast cycles have been extended to 10 hours, 14 days and 45 days respectively, which further increases the time margin of regulation and control and provides necessary data for the formulation of the province's power balance work plan.
At the same time, the company divides the province's 18 prefectures and cities into 25 regions according to the similarity of micrometeorology and micro-terrain, and based on various meteorological prediction data and algorithms such as long short-term memory neural networks in the region, it takes into account the meteorological evolution process such as strong winds and cloud shifts in the region, and builds a regional new energy total power prediction model to achieve accurate prediction of four types of space new energy power in the station, prefecture, region and province, and upgrades the prediction model for complex and changeable weather, accurately predicts the impact of extreme weather and issues new energy power generation warnings in advance.
In the next step, the company will continue to improve the regional new energy total power prediction model, do a good job in distributed photovoltaic refined power forecasting and improve the accuracy of power forecasting during peak load hours, overcome difficult problems such as power forecasting in extreme weather, governance and application of measured meteorological data at stations, support the scientific and efficient development of power dispatching, and ensure the safe and reliable supply of electricity and the efficient consumption of new energy. (Huo Xin)
