Model parameter adaptive GRU new energy short-term generation power prediction method

Abstract

The invention relates to a model parameter adaptive GRU new energy short-term generation power prediction method. The method comprises the steps of obtaining weather forecast data of an area where a target power station is located in a prediction time period and historical generated power corresponding to the target power station at at least one time point before the prediction time period; and inputting the weather forecast data and the historical power generation power into a pre-trained power generation power prediction model to obtain predicted power generation power of the target power station in the prediction time period. By adopting the method, the prediction data and the historical data are combined, factors influencing the generated power are also considered, and the pre-trained generated power prediction model adopts the hyper-parameter optimization algorithm to carry out hyper-parameter optimization, so that the model training is more perfect; by using the method, the predicted generated power of the target power station can be more accurate and more efficient.

Description

technical field [0001] This application relates to the field of artificial intelligence technology, in particular to a method for predicting short-term power generation of GRU new energy sources with adaptive model parameters. Background technique [0002] With the development of power generation technology, thermal power generation, nuclear power generation and new energy power generation technologies have emerged. The application of various power generation technologies will provide the country with electric energy protection. The stability of power generation will be the prerequisite for ensuring continuous power supply. Do a good job The prediction of power generation can provide a solution to the cause of the abnormality before the power generation is abnormal. [0003] In traditional technologies, statistical models are mostly used for prediction. However, the model parameters of this statistical model often require manual parameter adjustment. Since manual parameter a...

AI Technical Summary

Problems solved by technology

[0003] In traditional technologies, statistical models are mostly used for prediction. However, the model parameters of this statistical model often require manual parameter adjustment. Since manual parameter adjustment is more dependent on experience, users often cannot use the statistical model due to lack of experience in parameter adjustment. The model parameters are adjusted to the optimal model parameters, which leads to a large difference between the predicted results and the actual power generation, resulting in low prediction accuracy

Images