Freeze-drying modeling and monitoring using a new neuro-evolutive technique

This paper is focused on the design of a black-box model for the process of freeze-drying of pharmaceuticals. A new methodology based on a self-adaptive differential evolution scheme is combined with a back-propagation algorithm, as local search method, for the simultaneous structural and parametric optimization of the model represented by a neural network.Using the model of the freeze-drying process, both the temperature and the residual ice content in the product vs. time can be determined off-line, given the values of the operating conditions (the temperature of the heating shelf and the pressure in the drying chamber). This makes possible to understand if the maximum temperature allowed by the product is trespassed and when the sublimation drying is complete, thus providing a valuable tool for recipe design and optimization.Besides, the black box model can be applied to monitor the freeze-drying process: in this case, the measurement of product temperature is used as input variable of the neural network in order to provide in-line estimation of the state of the product (temperature and residual amount of ice).Various examples are presented and discussed, thus pointing out the strength of the tool.

► A new approach for monitoring/modeling the freeze-drying process is presented.
► The methodology combines a self-adaptive differential evolution with neural networks.
► A local search algorithm is added for improving the overall methodology.
► Optimal characteristics of the model are accurately and easily determined.
► Applications include recipe design and optimization of a freeze-drying process.