基于神经网络预测技术在建筑主体及基坑变形监测中的运用

Application of Neural Network Prediction Technology in Deformation Monitoring of Building Main Body and Foundation Pit

随着工程测量快速发展,变形监测成为保障工程建设安全措施之一。预测关键部位的变形量可以实时把控施工的安全,文中以南昌市某大型建筑工程项目为例,将神经网络预测模型应用到基坑及主体的变形监测中,并将预测结果同实际测量数据进行对比分析。结果表明,神经网络预测值的平均MAE值为0.15 mm,平均RMSE值为0.17 mm,神经网络预测模型的预测值精度较高,满足变形监测预测的精度要求。因此,利用神经网络可有效地保证大型建筑物施工期间的安全,并在未来的建筑施工项目变形监测中有更加广泛的应用价值和前景。

With the rapid development of engineering surveying,deformatoin monitoring has become one of the measures to ensure the saftey of engineering construction.Predicting the deformation amount of key parts can real-time control the safety of construction.This article takes a large construction project in Nanchang City as an example,applies a neural network prediction model to deformation monitoring of foundation pits and main structures,and compares and analyzest he predicted results with actual measurement data.The result sshow that the average MAE value of the neural network prediction value is 0.15 mm,and the average RMSE value is 0.17 mm.Therefore,the prediction accuracy of the neural network prediction model is high,meeting the accuracy requirements of deformation monitoring prediction.Therefore,the use of neural networks can effectively ensure the safety o flarge buildings during construction,and has more extensive application value and prospects in deformation monitoring of future construction projects.