Special section on intelligent control systems and signal processing

The IFAC International Conference on Intelligent Control Systems and Signal Processing (ICONS 2003) was organized under the auspices of the recently founded IFAC Technical Committee on Cognition and Control, and it was the first IFAC event specifically devoted to this theme. Recognizing the importance of soft-computing techniques for fields covered by other IFAC Technical Committees, ICONS 2003 was a multi-track Conference, co-sponsored by four additional Technical Committees: Computers for Control, Optimal Control, Control in Agriculture, and Modelling, Identification and Signal Processing. The Portuguese Society for Automatic Control (APCA) hosted ICONS 2003, which was held at the University of Algarve, Faro, Portugal.

Temperature, pH, and conductivity control of the nutrient solution of an hydroponic greenhouse

One of the aspects of modern agriculture is characterised by a culture without soil (hydroponic cultures). These culture techniques are identified by possessing automatic control systems to control the nutrient solution. In first hydroponic cultures this control was accomplished by “on- off” analog controllers that applied a single control law implemented in hardware. Therefore, the changes of the control law resulted in the change of all interface electronics. In digital control implemented by micro-controllers the alteration of such control law is easily performed by changing only a computer program, leaving untouched all the interface hardware. In this way, the use and substitution of the control strategy is improved, as well, the use of advanced control strategies.

Completely supervised training algorithms for B-spline neural networks and neuro-fuzzy systems

Complete supervised training algorithms for B-spline neural networks and fuzzy rule-based systems are discussed. By interducing the relationship between B-spline neural networks and certain types of fuzzy models, training algorithms developed initially for neural networks can be adapted by fuzzy systems.

Training multilayer perceptrons for control systems applications - a comparison of different approaches

Multilayer perceptrons (MLPs) (1) are the most common artificial neural networks employed in a large field of applications. In control and signal processing applications, MLPs are mainly used as nonlinear mapping approximators. The most common training algorithm used with MLPs is the error back-propagation (BP) alg. (1).

Learning control systems

Novel method of controller (PID) autotuning, involving neural networks and genetic algorithms: to employ neural networks to map the identification measures and controller parameters to objective functions, adapt these models on-line; to employ the genetic algorithm to perform on-line minimization.

Dietary biodiesel-derived crude glycerol in gilthead seabream juveniles (Sparus aurata): effects on growth performance and metabolism

In the European Union the turn towards renewable energy sources has increased the production of biodiesel from rapeseed oil, leaving glycerol (also known as glycerin) as a valuable by-product. For every litre of biodiesel produced, approximately 79 g of crude glycerol are generated. As the biodiesel production grows, the quantity of crude glycerol generated will be considerable and its utilization will become an urgent topic. One possibility is the use of crude glycerol on animal feeds. Glycerol has been evaluated as a dietary energy source for several farm animals, including fish. A study was undertaken to assess the effect of dietary biodiesel-derived glycerol (from rapeseed oil) on the overall growth performance, digestive capacity and metabolic nutrient utilization in juvenile gilthead seabream fed a low fishmeal level diet. Two practical diets were formulated to be isonitrogenous (crude protein, 45.4% DM), isolipidic (18.5% DM) and isoenergetic (gross energy, 21.3 kJ/g DM). The control diet (CTRL) was formulated with intermediate levels of marine-derived proteins (19%). In the same basal formulation, 5% glycerol (GLY) was incorporated at the expenses of wheat. Each dietary treatment was tested in triplicate tanks over 63 days, with 20 gilthead seabream (Sparus aurata), with a mean initial body weight (IBW) of 27.9  0.12 g. At the end of the trial, fish fed the CTRL diet reached a final body weight of 84.3  2.2 g (more than 3-fold increase of initial body weight). Fish fed the GLY diet showed a significantly higher (P<0.05) growth, expressed in terms of final body weight and specific growth rate. Voluntary feed intake was similar between the two treatments, but both feed efficiency and protein efficiency ratio were significantly improved (P<0.05) in fish fed the GLY diet. Dietary glycerol had no effect (P>0.05) on the apparent digestibility of protein. In comparison to the control treatment, dietary glycerol significantly improved (P<0.05) protein and fat retention. Activities of digestive enzymes were significantly affected by the various dietary treatments. Fish fed the GLY diet showed an enhanced activity of alkaline phosphatase (ALP) and pepsin, while activities of lipase and leucine-alanine peptidase (LAP) were little affected by dietary glycerol. Fish show the ability to use crude glycerol as a dietary energy substrate.