摘要
为更好地对变形体未来的形变量进行预测,研究粒子群BP神经网络模型,并对该模型进行改进,建立改进粒子群BP神经网络模型,结合工程实例,进行对比分析。研究表明,粒子群BP神经网络模型的拟合和预测精度均高于传统BP神经网络模型;改进后的粒子群BP神经网络模型进一步提高了拟合和预测精度,且随着改进方法的不断深入,其预测精度也逐渐提高;改进后的粒子群BP神经网络模型可以更好地对变形进行预测,优势显著。
Deformation monitoring and prediction is very important in engineering construction,the deformation forecast model of the research is of great significance,in order to have a better forecast of deformation,the particle swarm BP neural network model is studied,and further improve the model,the improved particle swarm BP neural network model is established,combined with the engineering example,to carry on the comparison and analysis,draws the following conclusion: the particle swarm BP neural network model of fitting and prediction accuracy is higher than the traditional BP neural network model; the improved particle swarm BP neural network model further improves the fitting and prediction accuracy,and the prediction accuracy is gradually improved with the continuous improvement of the improved method; the improved particle swarm BP neural network model can predict the deformation better,and the advantages are significant.
作者
冯康
张亚超
王文涛
孙朝印
FENG Kang;ZHANG Ya-chao;WANG Wen-tao;SUN Chao-yin(Yellow River Engineering Consulting Co.Ltd.,Zhengzhou 450003,China)
出处
《水利科技与经济》
2018年第8期70-74,共5页
Water Conservancy Science and Technology and Economy
关键词
预测模型
BP神经网络
粒子群
改进
应用
prediction model
BP neural network
particle swarm algorithm
improve
application