Lauter Tun control by neural networks

This works aims to analyze and control one of the more complex elements in the brewing process: a vessel called Lauter Tun. The objective of this stage is to separate the spent grain from the wort. The use of neural networks is described to implement a control system for this important stage in brewing. This process presents a great interrelation of variables and it is defined by nonlinear and time variant differential equations. A neural network model was used to represent the real plant. This model was used to test the performance of the neural network controller developed. The controller was trained using the information obtained from correct function controlled by an expert operator and the response of a digital PID controller tuned for this application. The paper shows the results obtained for this process using neural networks to model identification and plant control