论文部分内容阅读
用过去或现在的管道投资费规划设计未来的管道建设，可能导致建设期投资费不足，需先预测投资费再作规划设计；不同直径、壁厚的油气管道，其投资费亦不相同，需用回归方法来拟合其离散数据。因此，采用灰色预测和二元线性回归方法相结合，建立油气管道投资费的预测数学模型。算例用１４年的价格指数组成时间序列，采用新陈代谢ＧＭ（１，１）动态预测方法预测价格指数；将管道基期投资费按四组小范围的管径规格进行划分，再分别建立二元线性回归模型，并与管径规格不分组的回归模型进行比较，用分组的回归模型计算的基期管道投资费精度高、误差小。计算机运行结果表明预测、回归结果令人满意。经计算、分析、综合后，获得四组不同管径范围的管道投资费预测数学模型，预测出第１５年的管道投资费。由此可预测和计算第１６、１７、１８年的管道投资费。同时，还强调预测与计算中需注意的三个问题。
The past or present pipeline investment cost planning and design of the future pipeline construction may result in the construction period of investment costs are not enough, need to predict the investment costs for planning and design; different diameter, wall thickness of the oil and gas pipelines, the investment costs are not the same, Regress the discrete data by regression. Therefore, the combination of gray prediction and binary linear regression method is used to establish the mathematical model of prediction for investment cost of oil and gas pipelines. The example uses 14-year price index to compose the time series, and uses the metabolic GM (1,1) dynamic forecasting method to forecast the price index. Divides the pipeline base-period investment cost into four groups of small-diameter pipe size specifications and establishes the binary linear Regression model and compared with the regression model without pipe size specification. The investment cost of pipelines calculated by grouping regression model has high precision and small error. Computer run results show that the forecast results are satisfactory. After calculation, analysis and synthesis, four sets of mathematic models of pipeline investment cost forecast with different diameters and diameters were obtained, and the 15th year pipeline investment cost was predicted. From this it is possible to predict and calculate the pipeline investment costs for the 16th, 17th and 18th years. At the same time, it also emphasizes three issues that need attention in forecasting and calculation.