论文部分内容阅读
超临界流体的性质常与其密度相关。因此,如何精确计算超临界流体在不同操作条件下的密度值,对于超临界流体过程的研究和设计均十分重要。本文尝试采用人工神经网络技术来预测计算超临界流体的密度。网络结构为3层BP网,经优化中间隐藏层单元数为6。通过训练和学习,在压力6MPa-8MPa、温度300K-320K范围内,神经网络预测的密度值,其相对误差<0.35％。比P-B状态方程计算的结果精确。
The nature of supercritical fluids is often related to their density. Therefore, how to accurately calculate the density of supercritical fluid under different operating conditions is very important for the research and design of the supercritical fluid process. This paper attempts to use artificial neural network technology to predict the density of the calculation of the supercritical fluid. The network structure is a 3-layer BP network, and the number of hidden layer units in the middle of optimization is 6. Through training and learning, the relative error <0.35% is predicted by neural network in the range of pressure 6MPa-8MPa and temperature 300K-320K. The result is more accurate than the P-B equation of state.