空气污染现场监控站建设工程造价智能化预测方法研究

Research on Intelligent Prediction Method of Construction Cost for Air Pollution Monitoring Station

针对传统监控站工程造价预测方法不能调整预测表达式的阈值,导致工程造价预测值误差较大的问题,提出一种空气污染现场监控站建设工程造价智能化预测方法。首先依据中国现行建设项目总投资的结构,确定监控站工程造价指标,整理并预处理工程造价指标费用数据,利用Z-score标准化法计算工程造价中偏离指标平均值过远的费用数据,处理为费用自变量的回归方程,利用最小二乘法对费用方程求解,拟合方程得出的解,得到预测工程造价表达式,然后利用BP神经网络检测得出预测表达式的误差,根据误差值的大小调整预测表达式的阈值,实现智能化预测。

The traditional monitoring station project cost prediction method can not adjust the threshold value of the prediction expression,which leads to a large error in the prediction value of the project cost.An intelligent prediction method for the construction project cost of the air pollution field monitoring station is proposed.First,according to the structure of the total investment of the current construction projects in China,the project cost index of the monitoring station is determined,and the cost data of the project cost index is sorted out and pre-processed.The cost data deviating from the average value of the index is calculated by Z-score standardization method,which is treated as the regression equation of the cost independent variable.The cost equation is solved by the least square method,and the solution is obtained by fitting the equation for project cost prediction.The BP neural network is used to detect the error of prediction expression.The threshold value of prediction expression is adjusted according to the size of error value,and realize intelligent prediction.