作者: M. Dornier , M. Decloux , G. Trystram , A. Lebert
DOI: 10.1016/0376-7388(94)00195-5
关键词:
摘要: The neural network theory was used to dynamically model membrane fouling for a raw cane sugar syrup feed stream. use of networks enabled us integrate the effects hydrodynamic conditions on time evolution total hydraulic resistance under constant temperature and stream concentration. results obtained satisfactorily both variable transmembrane pressure crossflow velocity as filtration followed through time. hidden structure well scatter data quality modeling are discussed in this paper.