论文部分内容阅读
随着建设工程施工的信息化与安全化,基坑的变形预测是基坑设计和施工的重要补充手段。基于BP人工神经网络模型以及时序分析ARX自回归各态历经模型,对基坑的沉降变形进行了预测,数据结果表明两种模型均能较好地对未来值进行较真实的预测;从BP模型与ARX模型的预测结果均方误差值大小的角度而言,BP模型的预测对于未来趋势的判断比ARX模型要更强一些。试验结果说明两种预测模型应用于实际工程的监测预测具有实际意义。
With the informatization and safety of construction projects, the deformation prediction of foundation pit is an important supplement to the design and construction of foundation pit. Based on the model of BP artificial neural network and ARX autoregressive models, the deformation and deformation of the foundation pit are predicted. The results show that both models can predict the future value more accurately. From the BP model Compared with the ARX model, the prediction of the BP model is more powerful than the ARX model in judging future trends from the perspective of the mean square error of the prediction results of the ARX model. The test results show that it is of practical significance to apply the two prediction models to the actual project monitoring and prediction.