BP Neural Networks Based Soft Measurement of Rheological Properties of CWS

Rheological properties of the CWS are very important for pumping and combustion, but online measurement of them is still a problem for CWS’s Non-Newtonian fluid characteristics. Based on the special nonlinear mapping function of BP neural networks, a kind of soft measurement technology is put forward to predict the rheological characteristics of CWS. From the solids concentration and one or two apparent viscosities of the suspension at certain shear rate, the BP neural network can predicate the apparent viscosities at other shear rates with time dependence history. The result shows its effectiveness comparing with the Yield Power Low equation.