ABSTRACT
Batch Rice Drying ................................................................................... 232 12.3 Sequential Approach Example: Optimal Distillation
of Young Spirits....................................................................................... 242 12.4 Large Scale Example with the Simultaneous Approach:
Calibration of a Wine Fermentation Metabolic Model ........................... 246 12.5 Summary .................................................................................................. 250 References ............................................................................................................. 251
Modern process engineering increasingly relies on model-based methods. Hence, mathematical models are widely used to design and establish optimum operating conditions of processing units or whole processing plants. Most of the advances in this area come from applications to chemical processing plants that operate in continuous (steady state)mode. Consequently, mathematicalmodels are defined by large sets of nonlinear algebraic equations, and optimal design can be formulated as a nonlinear programming (NLP) problem. Powerful tools are available to optimize the design and operation of such plants. By contrast, the large majority of the operations in food processing are batch or semicontinuous, meaning that they are intrinsically dynamic. In these cases, finding optimal designs and operating strategies are much more difficult problems. These problems are called dynamic optimization (DO) or open loop optimal control. In food process engineering, DO can handle a variety of problems, such as optimal design and operation of batch processes, nonlinear model predictive control, parameter estimation of dynamic models, and dynamic data reconciliation.
