Corrosion Influence Factors Analysis Using Neural Network for Injecting Pipeline

Two methods are put forward to sort for injecting pipeline corrosion influence factors and to determine main influence factors, namely gray relation analysis and two layers BP neural network in this paper.The field examples show that prediction results of two layers BP neural network are better than those of gray relation analysis.Therefore, two layers BP neural network method should be adopted to analyze injecting pipeline corrosion influence factors to determine main influence factor.In this experimental zones, main influence factor of corrosion are sorted as below: O2 (0.877))pH (0.856)>SRB (0.84) >temperature(0.811)> pressure (0.78) >CO2 (0.76)>flow velocity (0.736)>0.7.