基于强降水时序特征的TSVGG-Light暴雨灾情指数预测模型构建及风险灾情一致性分布验证

Construction of TSVGG-Light rainstorm disaster index prediction model based on the temporal characteristics of heavy precipitation and verification of risk disaster consistency distribution

文章针对强降水灾情预测问题,提出了一种基于强降水时序特征的TSVGG-Light暴雨灾情指数预测模型。该模型利用气象观测数据中的降水时序特征作为输入,对数据进行特征提取和建模,预测出未来一段时间内的强降水灾情指数。为了验证模型的有效性,文中还进行了风险灾情一致性分布验证。实验结果表明,TSVGG-Light模型在强降水灾情预测上具有较高的准确性和稳定性,并且模型预测结果与实际灾情分布具有较好的一致性。TSVGG-Light模型可以成为预测和评估强降水灾情的有效工具。

This paper proposes a TSVGG-Light rainstorm disaster index prediction model based on the temporal characteristics of heavy precipitation disaster prediction.This model uses the precipitation time series characteristics from meteorological observation data as input to extract and model features from the data,and predicts the index of severe precipitation disasters in the future.In order to verify the effectiveness of the model,the paper also verified the risk disaster consistency distribution.The experimental results show that TSVGG-Light model has high accuracy and stability in the prediction of heavy precipitation disaster,and and the model prediction results have good agreement with the actual disaster distribution.The TSV GG-Light model can be an effective tool for predicting and assessing heavy precipitation disasters.