| 引用本文: | 杨茂,孟庆虎,王勃.基于电量-功率的多模态映射的长预见期光伏集群功率预测[J].电力自动化设备,2024,44(11):60-65,87. |
| YANG Mao,MENG Qinghu,WANG Bo.Power forecast of photovoltaic cluster in long forecast period based on electricity-power multimodal mapping[J].Electric Power Automation Equipment,2024,44(11):60-65,87. |
|
| 摘要: |
| 现有光伏功率预测方法的时间尺度大多短于7 d，长预见期时间尺度下建立的时序预测模型的拟合能力难以满足需求。提出一种基于电量-功率多模态映射的长预见期(8~15 d)光伏集群功率预测方法。对天气进行分型并提取粗粒度下的光伏出力特性，在此基础上对细粒度电量进行预测；基于出力特性构建约束，将预测电量还原为功率，实现长预见期下光伏功率的有效预测。将所提方法应用于甘肃某光伏集群，预测精度提升了6.58个百分点，在实现长预见期预测的同时，提高了可靠性。 |
| 关键词: 长预见期光伏集群功率预测 电量预测 天气特征预测 日净空功率曲线 电量重构 |
| DOI：10.16081/j.epae.202408011 |
| 分类号:TM615 |
| 基金项目:国家重点研发计划项目(2022YFB2403000) |
|
| Power forecast of photovoltaic cluster in long forecast period based on electricity-power multimodal mapping |
|
YANG Mao1, MENG Qinghu1, WANG Bo2
|
|
1.Key Laboratory of Modern Power System Simulation and Control & Renewable Energy Technology, Northeast Electric Power University, Jilin 132012, China;2.National Key Laboratory on Operation and Control of Renewable Energy and Energy Storage, China Electric Power Research Institute, Beijing 100192, China
|
| Abstract: |
| The time scale of existing photovoltaic power forecast methods is mostly less than 7 d, and the fitting ability of time series forecast models built under the time scales of long forecast period is difficult to meet the needs. A power forecast method for photovoltaic cluster based on electricity-power multi-modal mapping under long forecast period(8~15 d) is proposed. The weather is classified and the photovoltaic output characteristic under coarse grain is extracted, on this basis, the fine-grained electricity is forecasted. The constraints are constructed based on the output characteristic, and the forecasted electricity is restored to power, which realizes the effective forecast of photovoltaic power under long forecast period. The proposed method is applied to a photovoltaic cluster in Gansu province, and the forecast accuracy is improved by 6.58 percentage points, and the reliability is improved while the long forecast period is realized. |
| Key words: power forecast of photovoltaic cluster in long forecast period electricity forecast forecast of weather characteristic daily clearance power curve electricity reconstruction |
