基于粗糙集—RBF神经网络的单层RC厂房震害预测

Seismic Damage Prediction of Single-story Industrial Building Based on Rough Set and RBF Neural Network

RBF神经网络具有较好的仿真预测功能,粗糙集理论可以通过属性约简、重要度排序等对样本数据进行有效筛选。将粗糙集与RBF神经网络有机结合,建立单层RC厂房的震害预测模型。结合实际震例进行仿真训练,得到的单层RC厂房震害预测值与实际值基本吻合。表明该模型可对单层工业厂房进行较为有效的震害预测,且对震害预防也具有一定的指导意义。

Radial Basis Function neural network has good function of simulation prediction. Rough set theory can filter samples effectively through attribution reduction and attribute importance ranking. Rough set theory and artificial neural network are integrated into a model of seismic damage prediction for single-story reinforced concrete industrial building. The prediction results agree with actual seismic damage of single-story reinforced concrete industrial building. It shows that this model can be well used in the field of seismic damage prediction for single-story reinforced concrete industrial building and has guiding significance in the field of seismic damage prevention.