The Study On Prediction Of Water Injection Pipeline Corrosion Rate Of Block Ain Changqing Oilfield

Most of our oilfield produced water will be treated and used as injection water, so that the injected water sources to solve the problem, but also reduce the pollution caused by the discharge of produced water. However, thisinjected water containing a large amount of corrosive substances, would cause water pipeline corrosion damage, increased oil production costs, economic decline. In order to effectively know the situation of pipeline corrosion, timely maintenance and reinforcement, this article conducted a study, combining the laboratory experiments and BP neural network algorithm, through a diversified experimental data to define the influence of various factors, and then according to Genetic BP Network algorithm programming, predicted corrosion rate under the influence of the different factors, comparative analysis to reduce errors at the last.Under the premise of the corrosion rate prediction, this paper investigates the status of corrosion of Block A water pipeline in Changqing Oilfield, analyzed the water quality of the injected water in Block A. Coupon weight loss method was adopted in laboratory experiments to evaluate the corrosivity of various factors of Block A water pipeline.On the basis of experiments, analyzed and sorted the impact of various factors with gray correlation method to evaluate the impact of various factors on the corrosion rate, then identified the main factors of the corrosion rate. Combined Genetic BP Network algorithm with laboratory experiments to predict the corrosion rate of the pipeline under the influence of different factors, and then drawn graphs of the comparison results. Finally, comparing the results showed that:prediction and experimental results fit a higher degree. A method of combination of experimental simulation and algorithms prediction,can be a good tool to predict water pipeline corrosion rate in Block A.