José Antonio Torres

An Auto-Tuning PI Control System for an Open-Circuit Low-Speed Wind Tunnel Designed for Greenhouse Technology

Wind tunnels are a key experimental tool for the analysis of airflow parameters in many fields of application. Despite their great potential impact on agricultural research, few contributions have dealt with the development of automatic control systems for wind tunnels in the field of greenhouse technology. The objective of this paper is to present an automatic control system that provides precision and speed of measurement, as well as efficient data processing in low-speed wind tunnel experiments for greenhouse engineering applications. The system is based on an algorithm that identifies the system model and calculates the optimum PI controller. The validation of the system was performed on a cellulose evaporative cooling pad and on insect-proof screens to assess its response to perturbations. The control system provided an accuracy of <0.06 m·s−1 for airflow speed and <0.50 Pa for pressure drop, thus permitting the reproducibility and standardization of the tests. The proposed control system also incorporates a fully-integrated software unit that manages the tests in terms of airflow speed and pressure drop set points.

Combination of image processing and artificial neural networks as a novel approach for the identification of Bemisia tabaci and Frankliniella occidentalis on sticky traps in greenhouse agriculture

Display Omitted Image processing coupled with artificial neural network were used in a IPM system.Fed-forward neural networks were used to identify two pest species in greenhouses.Proposed whitefly and thrip identification method had a high precision, 0.96 and 0.92.Common sticky traps were used in a semi-automatic early pest detection system. Integrated Pest Management (IPM) lies at the core of the current efforts to reduce the use of deleterious chemicals in greenhouse agriculture. IPM strategies rely on the early detection and continuous monitoring of pest populations, a critical task that is not only time-consuming but also highly dependent on human judgement and therefore prone to error. In this study, we propose a novel approach for the detection and monitoring of adult-stage whitefly (Bemisia tabaci) and thrip (Frankliniella occidentalis) in greenhouses based on the combination of an image-processing algorithm and artificial neural networks. Digital images of sticky traps were obtained via an image-acquisition system. Detection of the objects in the images, segmentation, and morphological and color property estimation was performed by an image-processing algorithm for each of the detected objects. Finally, classification was achieved by means of a feed-forward multi-layer artificial neural network. The proposed whitefly identification algorithm achieved high precision (0.96), recall (0.95) and F-measure (0.95) values, whereas the thrip identification algorithm obtained similar precision (0.92), recall (0.96) and F-measure (0.94) values.

Ecological Sectorization Process Improvement through Neural Networks: Synthesis of Vegetation Data from Satellite Images Using RBFs

This paper presents an application of neural networks that uses radial basis function net architecture as a tool for simplifying and reducing the cost of ecological mapping. The process speeds up and replaces the classic means of obtaining ecological variables through field studies. The radial basis function networks were applied to estimate field data remotely, using data captured by the Landsat satellite and correlating it with ecological variables in order to substitute for them in the mapping process. The trial was undertaken for an area in south-eastern Spain, whereby, in 43 out of the 45 cases, the ecological variables could be obtained using satellite data. This approach substantially reduces the time and cost of ecological mapping, limiting field studies and automating the generation of the ecological variables.

Image analysis to qualify soil erodibility into a wind tunnel

A wind erosion research was carried out in a wind tunnel where sediment samples acquired were studied by an artificial vision camera. These images could be enlarged for further analysis. Image analyses were mainly colorimetry, number of particles present and their size. Soil wind erodibility was analyzed with the image analyses supported by other laboratory results. Anthrosols were the most erodible soils, whereas Calcisols showed the highest resistance to the erosive action of wind. Sediment characteristics show the influence of trap height with decreasing particle size, number and darkness as transport height increases. A two-factor ANOVA for main effect height showed that there were significant differences in particle number and size for sediments trapped 0-15 cm and 40-70 cm high. Soils could be grouped by differences in particle number and size at different heights into highly erodible Anthrosols and Leptosols, non-erodible Calcisols and Arenosols, in which fine particles were already depleted by natural wind erosion. Aggregation showed a similar pattern with decreasing values from Calcisols and Leptosols to Anthrosols and finally Arenosols, where only single sand grains were observed in adhesive traps.