Simulation Of Pipeline Steel Pit Corrosion In An Oil And Gas Field

The corrosion of oil and gas field in western is very serious, according to the actual working condition, simulation the oil and gas pipe line corrosion in two aspects of macroscopic and mesoscopic by artificial neural network method and the cellular automata.The corrosion time and effects of different corrosion block porosity on the pitting process are investigated by modeling.Modeling for oilfield corrosion pipeline perforation by artificial neural network, and dimension reduction by the fuzzy curve-fuzzy surface method, getting the influence of various factors on the corrosion from big to small order of wall thickness, temperature,pressure, flow rate, inner diameter, CO2, corrosion inhibitor added, water content, amount of fluid and mineralization. modeling results show that the data dimension reduction after improved the accuracy of artificial neural network modeling. Dimension reduction rise accuracy of the artificial neural network modeling in the beginning, continue to dimension reduction will bring down the modeling accuracy. The best result of modeling is 9 factor after removing salinity factor. That illustrate the fuzzy curve fuzzy surface method and artificial neural network method based on multi factor input output system modeling of single factor is more effective.Based on predecessors' study on simulation of the pitting process by cellular automata method, building the cellular automata model of pitting corrosion based on diffusion - reaction process, as cell size is 27 nm, cellular automata run unit time is 1.2×10-3s, dissolution probability is 0.01, the simulation results are consistent with actual corrosion process,simulation current and current density are consistent with actual measured value in the process of pitting. Getting the corrosion diameter, block deposit, Fe2+ ion concentration and hydrolysis, electrochemical noise of pitting corrosion, and other information, simulation results are in conformity with actual pitting process. After analysis that illustrate on the pitting process, the closures porosity greater than 9.88%, corrosion block content porosity is large,the occluded cell effect can't form, the ion is easy to exchange in the pitting corrosion, pitting corrosion is difficult to keep the acid environment, the matestable pit corrosion will be passivation; the closures porosity less than 9.88%, corrosion block content porosity is small,the occluded cell effect may form, the ion is difficult to exchange in the pitting corrosion,pitting corrosion is easy to keep the acid environment, the matestable pit corrosion will become stable pit corrosion. Cellular automata simulation result is consistent with the actual process of pitting,as a kind of mesoscopic modeling method,providing a new route for the study of the pit corrosion.