用人工神经网络预测天然气管道内腐蚀速度

Predict the Internal Corrosion Rate of Gas Pipeline with Artificial Neural Network

摘要: 介绍了人工神经网络技术及其BP算法的网络模型。引入人工神经网络模型, 对输气管道内腐蚀速度进行了预测, 预测结果表明, 人工神经网络预测的结果与管道内壁实际腐蚀速度接近, 预测精度较高, 尤其是在处理非线性数据方面, 人工神经网络更优于目前普遍采用的逐步回归法。

Abstract: This paper adopts the artificial neural network to predict internal corrosive rate of gas pipeline. The predictable results show that the results obtained by the method is closed to the actual internal corrosion rate with a high accuracy. The writers consider that the artificial neural network is better than the stepwise regression analysis method, and suggest that this method can be used as a new method to predict internal corrosion rate of gas pipeline.