Method of Displacement Parameters Prediction Based on Neural Network for Natural Gas Pipeline
Theoretical investigation is not enough for natural gas pipeline displacement technology of nitrogen isolation method without isolator, which makes displacement parameters hard to be controlled accurately. Therefore, the method of natural gas pipeline displacement parameters prediction based on neural network is proposed in this paper. From Fluent numerical simulation, some useful information is acquired, such as displacement parameters under the typical working conditions, the relationship between different working conditions and nitrogen injection capacity. Moreover, the samples gained by simulation are trained by neutral network which has association function. Thus the aim of forecasting process parameter for nitrogen displacement technology without isolator is attained, which provides an important reference for the production of nitrogen displacement of nature gas pipeline without isolator.
Yiyi Zhouzhou and Qi Luo
S. L. Liu et al., "Method of Displacement Parameters Prediction Based on Neural Network for Natural Gas Pipeline", Applied Mechanics and Materials, Vol. 79, pp. 232-236, 2011