Combining first principles modelling and artificial neural networks: a general framework

In this work a general hybrid model structure for stirred-tank bioreactors is proposed. The general structure combines first principles modelling with artificial neural networks: the bioreactor system is described by a set of mass balance equations, and the cell population system is represented by an adjustable mixture of neural network and mechanistic representations. The identification of unknown parameters from measurements is studied in detail. The sensitivities equations are derived for the general model enabling the analytical calculation of the Jacobian Matrix. The identification procedure consists of a least squares optimisation that employs a large-scale Sequential Quadratic Programming (SQP) algorithm. The methodology is outlined with simulation studies.