摘要
变形监测工程是一个复杂的综合系统,各种参数具有很大的不确定性。目前变形的预测分析多采用单一的预测方法,而各种方法都有各自的优缺点和应用范围,有时单一的预测方法对判定工程性质带来了困难。引入了组合预测的思想,在灰色GM(1,1)模型的基础上,构建了灰色+GA+BP神经网络组合模型,探索了时间序列的数据处理和预报问题,通过实例计算分析,证明该组合模型满足工程需要,具有一定的使用价值。
Because of various parameters have great uncertainty,deformation monitor project is a complicated integrated system.At present deformation prediction and analysis using a single forecasting methods,but each method has their own application scope.Sometimes a single forecasting method made it difficult to determine the nature of projects.This paper introduced the idea of combination forecasting,based on the gray GM(1,1) model,Construction of the gray + GA + BP neural networks combination models,explored the time series of data processing and prediction problems.Calculation and analysis by example,proved that the combined model met the engineering needs and had a certain value.
出处
《地理空间信息》
2011年第2期109-111,9,共3页
Geospatial Information
关键词
沉降监测
组合模型
灰色预测模型
遗传算法
BP网络
deformation monitor,combination model,gray prediction model,genetic algorithm,BP network
