基于人工智能方法的近临界区CO_(2)热物性模化与预测

Modeling and Prediction of Thermophysical Properties of CO_(2)in the Region around the Critical Point using Artificial Intelligence Models

为更准确预测CO_(2)在临界点附近区域的热物性,分别建立了基于BPNN,SVR和GPR算法的智能模型来预测近临界区CO_(2)的密度、粘度和导热系数,并将3种模型进行比较。结果表明:基于BPNN的密度(R^(2)=0.9465)和粘度(R^(2)=0.9702)预测模型相较于其他智能模型精度更高,而基于SVR的导热系数的模型预测精度更高(R^(2)=0.9997);所提出的智能模型相较于传统模型中SW密度方程(R^(2)=0.5966)、Laesecke的粘度方程(R^(2)=0.8445)和J&H的导热系数方程(R^(2)=0.0218)的R^(2)提高了14.88%~4444.5%。

In order to more accurately predict the thermophysical properties of CO_(2)in the region around the critical point,the intelligent models of BPNN,SVR and GPR were developed to predict and compared the density,viscosity and thermal conductivity of CO_(2)in the near-critical region,respectively.The results show that the density(R^(2)=0.9465)and viscosity(R^(2)=0.9702)prediction models based on BPNN are more accurate than other intelligent models,and the thermal conductivity model based on SVR is more accurate(R^(2)=0.9997).Compared with the SW density equation(R^(2)=0.5966),Laesecke′s viscosity equation(R^(2)=0.8445)and J&H′s thermal conductivity equation(R^(2)=0.0218)in the traditional model,the R^(2)of the proposed intelligent model is improved by 14.88%to 4444.5%.