ISSN 1000-0526
CN 11-2282/P
CN 11-2282/P
Short Wave radiation Forecast Correction Based on Machine Learning in Henan region
- Article
- Figures
- Metrics
- Preview PDF
- Reference
- Related
- Cited by
- Materials
Abstract:
IUsing the total irradiance data from 23 radiation observation stations in Henan Province in 2022 and CMA-WSP2.0 model products,,characteristic variables were selected by Lasso regression, training data sets and test data sets were established, and machine learning methods (Random forest, XGBoost, LightGBM) were used to train the model using the training data set, and the total irradiance forecast by CMA-WSP2.0 model in Henan Province was revised. The revised results were tested by site and region, season and total irradiance classification, and the following conclusions were obtained: The three machine learning methods of random forest, XGBoost, and LightGBM have good correction effects. Compared with the CMA-WSP2.0 model prediction results, the average absolute error and root mean square error are significantly reduced, and the 24-hour accuracy and 24-hour qualification rate are significantly improved. The average absolute error decreases by 18.32~32.91 W·, the average error decreases by 38~56%, and the root mean square error decreases by 36~52%. The 24-hour average accuracy and 24-hour average qualification rate increased by 7.3% and 5.7%. The results of regional statistics are consistent with those of the stations. For the five regions, the correction effect of western Henan is the best. The corrected deviation range of the three machine learning methods is narrower than that of the CMA-WSP2.0 simulation set, and the probability of the deviation distribution near 0 is greater. Among the seasonal test results, the three methods have more significant correction effect in winter. For different total irradiance levels, the three machine learning methods can effectively improve the CMA-WSP2.0 model prediction, and the correction effect tends to gradually weaken with the increase of total irradiance levels. The results can provide useful reference for improving the ability of total irradiance forecast in Henan Province.
Keywords:
Project Supported:
Clc Number:
Mobile website
