基于BP神经网络的建筑物沉降预测模型研究

Prediction model of building settlement based on BP neural network

人工神经网络(ANN)是一个拥有高度非线性映射能力的计算模型,有较强的动态处理能力。在对其进行研究的基础上,利用MATLAB建立BP神经网络的建筑物沉降预测模型,指导建筑物的沉降预警工作。通过将建筑物沉降的实测数据和模型的预测数据进行对比分析,发现两者间的误差相对较小,预测模型能很好地反映建筑物沉降的发展趋势,对于建筑物沉降预警工作有着极其重要的意义。同时,研究结果也证明了BP神经网络预测模型具有较高的精确性和稳定性,可以在类似工程中加以应用。

Artificial Neural Network （ANN） is a computing model that has highly nonlinear mapping ability and strong dynamic processing capabilities. On the basis of deep research, MATLAB is used to build BP neural network of building settlement prediction model, to guide building settlement warning. Compared with building settlement of the measured data and model forecast data, the error between the two is relatively small and the prediction model will reflecte the development trend of building settlement accurately. So it has an important significance for building settlement early warning. At the same time, the results also proves the high accuracy and stability of BP neural network prediction model and it can be applied to the similar projects.