Weight Adjustment Method and Power Prediction Method Based on Multiple Numerical Weather Forecast Sources

Abstract

The invention discloses a weight adjustment method and a power prediction method based on multiple numerical weather forecast sources: 1. Determine the type n of numerical weather forecast sources and the weight coefficient ε of each numerical weather forecast source i ;2. Set the number of days days, and initialize a=1; 3. Input n kinds of numerical weather forecast source data w on day a in sequence i ; 4. Calculate power prediction data f ai ; 5. Obtain the weight coefficient ε of the i-th numerical weather forecast source on the a-th day ai ;6. Adjust and correct the weight coefficient; 7. Change the corrected weight coefficient ε ai Save, if a≥days, go to step 8, otherwise add 1 to the value of a and go to step 3; 8. Obtain the averaged one-dimensional array array_avg_pre[]; 9. Perform normalization processing to obtain the processed array array_avg[] k1, k2,...kn], k1, k2,...kn is each element in the array, which corresponds to the corrected weight coefficient of each numerical weather forecast source. Improve power prediction accuracy.

Description

technical field [0001] The invention relates to a weight adjustment method and a power prediction method based on multiple numerical weather forecast sources. Background technique [0002] At present, more and more wind farms and photovoltaic power stations are being built, and the power prediction results of wind farms and photovoltaic power stations are an important basis for rationally arranging power grid power generation plans. Since the power prediction results involve the future 72 hours or even longer time period, the influence of numerical weather prediction on it is very important. [0003] At present, in the process of power prediction calculation, there are usually multiple numerical weather prediction sources as input. The common practice is to calculate the average of multiple numerical weather prediction sources, and then use it as the input of power prediction to calculate the prediction results. Although the algorithm is simple, However, for the averaging o...

AI Technical Summary

Problems solved by technology

[0004] In view of the above problems, the present invention provides a weight adjustment method and a power prediction method based on multiple numerical weather forecast sources, which overcomes the low power prediction accuracy rate caused by the large error of a certain numerical weather forecast source among various numerical weather forecast sources, and The error of the same numerical weather prediction source in different time periods is different, which makes the accuracy of power prediction results low

Images