板式蒸发器换热与压降关联式的改进BP网络模型辨识

为了获得精确的换热及压降关联式以提高仿真模型的精度,本文在建立板式蒸发器数学模型的基础上,提出了用换热及压降关联式修正系数以修正经验关联式计算结果,并采用改进BP网络模型(IBPNM)对修正系数进行了辨识研究；与已有的板式蒸发器的换热和压降关联式相比,本文得到的结果与试验数据吻合更好,优于现有的研究结果。

Improved BP Network Identification of Heat Transfer and Pressure Drop Correlations in Plate Heat Exchangers (PHE)

Presented a new method to predict the heat transfer coefficients and pressure drops in the simulation models.Based on the training data composed of the modification coefficients (MCs) of the current heat transfer and pressure drop correlations calculated by the PHE mode,the improved BP (back-propagation network) was trained to identify the relations between the MCs and the five parameters-refrigerant mass flow,inlet pressure,inlet enthalpy, inlet water temperature and water mass flow.The new correlations multiplied by the attained MCs can output heat transfer coefficients and pressure drops fitting well with the experiments results.