摘要
盐水中CO_2的溶解度参数对CO_2地质封存至关重要,通过实验方法获取溶解度数据耗时费力,因此,需要建立理论模型来进行预测,而神经网络由于能够关联复杂变量之间的情况而广受关注。BP神经网络是1种应用最广泛的前馈神经网络,利用实验数据建立1个三层结构的BP神经网络模型用于预测盐水中CO_2的溶解度,并对网络的结构参数进行优化设计,得到1种盐水中CO_2溶解度预测的BP神经网络模型。同时,利用修正后的亨利定律计算不同条件下的溶解度,并将实验数据、BP模型预测结果与亨利定律做对比,为确定盐水中CO_2的溶解度提供了1种新方法。
The solubility of CO_2 in brine water is essential for CO_2 geological sequestration.However,the experimental method to obtain the solubility data is time-consuming.Therefore,a theoretical model should be established to predict the solubility.Neural networks have attracted much attention because they can be used to correlate the situation between complex variables.In this paper,BP neural network is one of the most widely used feedforward neural networks.A three-layer BP neural network model is established by using the experimental data to predict the solubility of CO_2 in brine,and the structural parameters of the network are optimized.A BP neural network model for predicting the solubility of CO_2 in brine is obtained,and the solubility under different conditions is calculated by using the modified Henry's law.The experimental data and BP model prediction results are compared with Henry's law.The BP model provides a new approach to determine the solubility of CO_2 in brine.
出处
《中国科技论文》
北大核心
2017年第24期2831-2834,共4页
China Sciencepaper
基金
国家自然科学基金资助项目(51404037)
关键词
油气田开发工程
BP人工神经网络
CO2封存
溶解度
预测
oil-gas field development engineering
BP artificial neural network
CO2 sequestration
solubility
prediction
