Monitoring of Foundation Pit Deformation based on Backpropagation Neural Network

The precise monitoring of foundation pit deformation is crucial for ensuring construction safety and structural stability. Foundation pits are affected by multiple geotechnical factors, which can lead to unexpected deformations and potential structural failures. Accurately predicting these deformations is essential for early risk mitigation. However, conventional monitoring techniques often struggle to model the nonlinear and dynamic nature of geotechnical parameters, limiting their effectiveness in real-time prediction. To address this challenge, this research presents a foundation pit deformation monitoring system based on a Backpropagation (BP) neural network. The model is trained on historical deformation data, soil characteristics, and environmental factors to improve predictive accuracy. By capturing complex nonlinear relationships, the proposed model enables precise forecasting of deformation trends, facilitating proactive safety measures. Experimental findings demonstrate the superiority of the proposed system, achieving a Mean Absolute Error (MAE) of 0.021 mm, Root Mean Square Error (RMSE) of 0.032 mm, and an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score of 0.96, highlighting its strong predictive performance across conditions. This AI-driven approach offers a resource-efficient and intelligent approach for geotechnical monitoring, enhancing safety management in deep excavation projects.