This paper shows an inferential sensor that has been developed to be used in the olive oil industry. This sensor has been designed to measure two variables that appear in the elaboration of olive oil in a mill which are very difficult to be measured on line by a physical sensor. The knowledge of these variables on line is crucial for the optimal operation of the process, since they provide the state of the plant, allowing the development of a control strategy that can improve the quality and yield of the product. This sensor measures variables that in other case should come form laboratory analysis with large processing delays or from very expensive and difficult to use on line analysers. The sensor has been devised based upon artificial Neural Networks (NN) and has been implemented as a routine running on a Programmable Logic Controller (PLC) and successfully tested on a real plant.
[1]
Timothy A. Haley,et al.
A survey of automation practices in the food industry
,
2001
.
[2]
C. Bordons,et al.
Modelling and predictive control of an olive oil mill
,
2001,
2001 European Control Conference (ECC).
[3]
Carlos Bordons,et al.
PREDICTIVE CONTROL OF AN OLIVE OIL MILL WITH MULTI-OBJECTIVE PRIORITIZATION
,
2002
.
[4]
Li Zhang,et al.
An evaluation of back-propagation neural networks for the optimal design of structural systems: Part I. Training procedures
,
2002
.
[5]
David M. Haaland,et al.
New augmented classical least squares methods for improved quantitative spectral analyses
,
2002
.