论文部分内容阅读
油气管道腐蚀作用机理复杂,诸多因素存在模糊性与相互关联性,对管道腐蚀速率的有效预测存在一定困难。运用具有处理线性和非线性因素能力的灰色线性回归组合模型对油气管道腐蚀进行建模,并引入BP人工神经网络模型对组合模型的残差进行修正,以进一步提高预测精度。将所建模型应用于某油气管道腐蚀速率的预测中,结果表明:所建灰色线性回归组合模型在对油气管道进行腐蚀预测时,既充分考虑到了原始数据的线性因素,又考虑了其非线性因素,预测效果较好,同时适用于解决其他复杂腐蚀体系的预测问题。
The mechanism of corrosion in oil and gas pipelines is complex, and many factors are ambiguous and interdependent. There are some difficulties in the effective prediction of pipe corrosion rate. The combination of gray linear regression model with linear and nonlinear factors was used to model the corrosion of oil and gas pipelines. BP artificial neural network model was introduced to correct the residuals of the combined model to further improve the prediction accuracy. The model was applied to the prediction of the corrosion rate of an oil and gas pipeline. The results showed that the proposed gray linear regression model not only fully considered the linearity of the original data but also considered its nonlinearity Factors, the prediction effect is better, at the same time suitable for solving other complex corrosion prediction problems.