| 摘要: |
| 元素硫在高含硫气体中溶解度的研究是硫沉积机理研究、硫沉积预测和处理技术研究的前提和基础,也是元素硫沉积室内研究工作的核心课题。为了关联和预测硫在高含硫气体中的溶解度,提出误差逆向传播人工神经网络(BP ANN)模型,并设计了该模型的计算过程,讨论了该模型的参数设置。计算结果表明,该模型可作为模拟和内推硫在高含硫气体中溶解度的一种较好手段,但外推效果较差。与现有其他硫溶解度计算模型相比,该模型计算结果优于Chrastil缔合模型和经验公式,与状态方程法和六参数缔合模型的计算结果相当。 |
| 关键词: 硫沉积 BP神经网络 预测 元素硫 高含硫气体 溶解度 |
| DOI：10.3969/j.issn.1007-3426.2015.03.001 |
| 分类号:TE642 |
| 基金项目:国家自然科学基金“天然气管道跨越结构清管动力响应实验及理论研究”(51174172)；教育部博士点专项科研基金“高温高压复杂天然气集输工艺基础理论研究”(20125121110003)。 |
|
| Prediction of sulfur solubility in high sulfur gas based on BP neural network |
|
Chen Lei, Li Changjun
|
|
(School of Petroleum Engineering,Southwest Petroleum University, Chengdu 610500, China)
|
| Abstract: |
| Research on the elemental sulfur solubility in high sulfur gas is the premise and foundation of sulfur deposition mechanism, sulfur deposition prediction and treatment technology research, as well as the core subject of indoor sulfur deposition research work. To associate and predict the sulfur solubility in high sulfur gas, a Back Propagation Artificial Neural Network (abbreviated as BP ANN) model was proposed. Implementation procedure and parameters setting of this model were introduced in detail. The results showed that the model could simulate and interpolate the solubility of sulfur in high sulfur gas, while the extrapolative effect was poor. Compared with other existing model, the caculation results of BP ANN was model better than that of the Chrastil association model and the empirical formula, which was in accord with the calculation results of the equation of state method and the six parameters association model. |
| Key words: sulfur deposition BP neural network prediction elemental sulfur high sulfur gas solubility |
