人工神经网络在中浓纸浆流体化仿真研究中的应用

Simulation Study of Medium-Consistency Pulp Suspension Fluidization Based on ANN

中浓纸浆悬浮液在剪切室内流动时,其所受剪切应力与湍流发生器的几何形状、转速及浆料浓度都有关系,通过改进算法的BP网络对实验数据的拟合,人工神经网络模拟仿真了各个因素对中浓纸浆所受剪切应力产生的影响,此外通过网络训练达到的稳定模型,可以预测中浓纸浆在更高浓度和湍流发生器转速下所受的剪切应力值。

The shear stress received by the medium-consistency pulp suspension when it flowed in a shearing chamber was related to the eometrical shape, rotation speed of the turbulence generator and the consistency of pulp suspension. Simulation study of the effect of various factors on the shear stress received by the medium consistency pulp suspension based on ANN was a good modeling method. BP (Error Back-Propagation) artificial neural networks that ameliorate the arithmetic fit experimental data very well, based on the networks stable model, the shear stress received by the medium-consistency pulp suspensions could be forecasted even in the higher consistency and higher rotation speed of the turbulence generator.