Train surface wind pressure prediction model training method, prediction method and related device

Abstract

The invention discloses a train surface wind pressure prediction model training method, a train surface wind pressure prediction method and a related device.The train surface wind pressure prediction model training method comprises the steps that a training set is adopted to train a preset hybrid model till a first loss function value of the preset hybrid model meets a training stop condition, and a target prediction model is obtained; wherein the preset hybrid model comprises a first preset model and a second preset model, the first preset model is used for predicting and obtaining first predicted wind pressure data corresponding to the failure pressure measuring point based on the training sample, and the second preset model is used for predicting and obtaining the failure pressure measuring point based on the graph structure information and the first predicted wind pressure data. And predicting to obtain second predicted wind pressure data corresponding to the failure pressure measuring point. Thus, information from different models can be integrated, the overall prediction performance can be improved, the second preset model can help predict or interpolate missing values through the graph structure by using graph structure information, and therefore, the target prediction model can be applied to any scene to predict wind pressure data, and the application universality is improved. 本申请公开了一种列车表面风压的预测模型训练方法、预测方法及相关装置，其中，该方法包括：采用训练集对预设混合模型进行训练，直至预设混合模型的第一损失函数值满足训练停止条件，得到目标预测模型。其中，预设混合模型包括第一预设模型和第二预设模型，第一预设模型用于基于训练样本预测得到失效压力测点对应的第一预测风压数据，第二预设模型用于基于图结构信息和第一预测风压数据，预测得到失效压力测点对应的第二预测风压数据。如此，可以整合来自不同模型的信息，提高整体预测性能，而第二预设模型利用图结构信息，可以通过图结构帮助预测或插补缺失值，因此，目标预测模型可以应用到任何场景下预测风压数据，提高了应用的广泛性。