Search

Copyright of Low-Carbon Chemistry & Chemical Engineering is the property of Low-Carbon Chemistry & Chemical Engineering Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Aiming at the problem of frequent occurrence of freezing blocking accidents in natural gas production pipelines during winter, the technology of BP (back propagation) artificial neural networks was utilized to predict the theoretical quantity of methanol injection as outputs, considering the wellhead pressure, the wellhead temperature, the daily gas production, the daily water production, and the relative density of natural gas as inputs. The Sobol sensitivity analysis Was used to find the key parameters that might significantly affect the theoretical quantity of methanol injection. Meanwhile, the distribution network relevant to methanol injection was also optimized. The prediction model of the 5-16-12-1 BP artificial neural networks evolves 2 764 epochs until the mean square errors of training samples, verification samples, and test samples are 0.005 5, 0.007 2, and 0. 08 5 respectively, which are less than the allowable convergence error limit of 0. 010 0. A good correlation is presented as the determination coefficients reaching 0.999 5, 0.998 6, and 0.996 4 respectively. The wellhead temperature, the relative density of natural gas, and the wellhead pressure have a greater impact on the theoretical quantity of methanol injection, while there may be significant interactions between the wellhead pressure, wellhead temperature, gas production per day, water production per day and other factors. On the conditions of the radial and circular pipe network mode, there is a shortage as the practical quantity of methanol injection is less than the theoretical quantity of methanol injection for Jing 99-66, Jing 99-65, Jing 99-66H2, and Jing 98-65 . However, the practical quantity of methanol injection is more than the practical quantity of methanol injection for will gas wells of Xinbei Station 5 under the dendritic pipe network mode to achieve the goal of preventing freezing blockage. The research results could provide an essential theoretical basis and data source for the consumption reduction and benefits increment of the methanol injection process of natural gas production pipelines. [ABSTRACT FROM AUTHOR]