Abstract
The control of fermentation is very important in order to ensure a sustained level of wine quality year after year. Usually wine fermentation is carried out at a constant temperature. This fact is due mainly to traditional reasons. Nevertheless the results year after year and from region to region may be very different due to large variations in must characteristics. A more complex control is needed to overcome this fluctuations. Recent works have introduced some innovation but still have some empirical characteristics and can not control the duration of fermentation. In order to predict the behaviour of the fermentation, a reasonably accurate mathematical model is required. Unfortunately mathematical models for wine fermentation are scarce, particularly for wine batch fermentation. In this work we develop a mathematical fermentation model and an industrial tank thermal model. A new control strategy is proposed: Relative density control instead of temperature control. A reference model is developed and we try to tune a PID controller to the process. The need of more complex controllers such as predictive controllers is addressed.

Abstract
In this article we present two methods for the real time automatic measurement of the relative density. The first one relies on the heat generated during the fermentation which requires no additional sensors in the tanks. It is based on the thermal model of the tank and the only additional requirements are the measurement of the temperature of the surrounding air. The other method is based on two low-cost pressure sensors with an algorithm for autocalibration and compensation of temperature perturbation. The autocalibration is done during fermenter feeding and with only one manual measurement of the relative density.

Abstract
The technique of genetic algorithms is proposed as a means of auto-tuning PID controllers. The technique revolves firstly using on-line data and the genetic algorithm to identify a model of the process. Then the identified model, the genetic algorithm and simulation methods, are used to off-line tune the PID controller, so as to minimize a time-domain based cost function. Finally, the genetically tuned controller is implemented on-line on the real process. The results of the genetic auto-tuner are illustrated by auto-tuning a PID controller on a laboratory heat exchanger, and comparing the genetic auto-tuning technique with the Astrom-relay auto-tuning technique.

Abstract

Abstract
This paper describes the implementation of a distributed data acquisition network based on the 80C592 microcontroller from Intel. Each Station is connected in a hierarchical way to form a tree topology. The lower level network stations, designated by Slaves, are dedicate to the data acquisition and the generation of control signals. The upper level, Masters, are responsible for the communications control. Both networks uses a CAN - Controller Area Network - Bus, for Data Transferring, and the global Network is also connected to a PC, via CAN. A device router, NetManager, was implemented to support total intrinsic requirements at the communication level. This type of connection allows total configuration from a personal computer, PC, in which runs a software application developed for Windows(TM) environments. The tests performed at the laboratory, with transmission rates varying from 40Kbits/s to 1Mbits/s, showed that the communications were performed without errors for cable lengths of 1100m to 40m, respectively. This system is now being installed in a set of environmental chambers and greenhouses located on UTAD, where it will be monitored and controlled the air temperatures and humidities, the CO2 and ammonia concentrations and the radiation level.

Abstract
For a greenhouse located at UTAD-University, the methods used to estimate (in real-time) the parameters of the inside air temperature model will be described. The structure and the parameters of the climate discrete-time dynamic model were previously identified using data acquired during two different periods of the year. Several experiments showed that the second-order models identified achieve a close agreement between simulated and experimental data. Later, it was found that parameters change with varying operational conditions. Thus, for an efficient use of these models in real-time control, a recursive identification technique was implemented for the estimation of the parameters. Copyright (C) 1997 Elsevier Science Ltd.