Distributed new energy output prediction method and related device
By adopting the Autoformer model based on a progressive decomposition architecture in the new energy output prediction, the problem of accuracy and inefficiency of new energy output prediction in the existing technology is solved, and more efficient and accurate prediction results are achieved.
CN119695873APending Publication Date: 2025-03-25ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID JIBEI ELECTRIC POWER CO LTD +2
0 Cites 0 Cited by
Patent Information
- Application Number
- CN202411790689.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-06
- Publication Date
- 2025-03-25
AI Technical Summary
Technical Problem
The existing technology has poor accuracy and low efficiency in the high proportion of distributed photovoltaic and fan access scenarios.
Method used
The Autoformer model based on the gradual decomposition architecture is used to predict distributed new energy output. Through the autocorrelation mechanism and gradual accumulation, the prediction results of the periodic part and the trend part are obtained and combined to finally obtain the predicted value.
Benefits of technology
It improves the accuracy and efficiency of new energy power generation forecasts, and can quickly and accurately obtain output forecast values, which helps to effectively connect to the grid and manage grid stability of distributed new energy.
✦ Generated by Eureka AI based on patent content.
Abstract
The invention belongs to the technical field of new energy output prediction, and discloses a distributed new energy output prediction method and a related device. The distributed new energy output prediction method comprises the following steps: acquiring generated power time sequence data of distributed new energy to be subjected to output prediction; carrying out output prediction by adopting an output prediction model based on the obtained power generation power time sequence data to obtain a distributed new energy output prediction value; the output prediction model adopts an Autoformer model based on a progressive decomposition architecture; in the output prediction model, hidden variables are decomposed gradually based on input generation power time sequence data, prediction results of a period part and a trend part are obtained through an autocorrelation mechanism and step-by-step accumulation and are combined, and finally a prediction value is obtained. According to the technical scheme, rapid and accurate prediction of distributed new energy power generation can be realized, and efficient operation and maximum energy utilization of a new energy power generation system can be ensured.
Need to check novelty before this filing date? Find Prior Art