The methodology of the article was composed by four stages: (i) a wireless sensor network (WSN)-based monitoring system was deployed in the drying installation to capture input variables, (ii) an artificial neural network was developed to estimate output variables using input variables, (iii) different experiments were performed to analyse the proposed methods, and (iv) experimental results were analysed to evaluate the proposed method.2.?Tobacco Drying2.1. MaterialsA tobacco dryer from the Agrotex Company located in Rosalejo (C��ceres, Spain), was employed in this research. This tobacco dryer has a hot water valve to heat the air, a fan to move the air, and two air hatchways to remove the humidity.

In this dryer, three measurement motes were deployed in its drying chamber to acquire the temperature and relative humidity in different points of the dryer.

Figure 1(a) shows the schema of the cross section of this tobacco dryer, where the aforementioned equipment is represented. Figure 1(b) shows the spatial distribution of the three motes deployed inside the drying chamber: the first one was placed next to the supervision window (green), the second one was placed next to the roof (purple), and the third one was placed into the tobacco mass (red). The sensors of these measurement motes have been numbered as Sensors 1, 2, and 3, respectively, for easy reference.Figure 1.(a) Cross section of the tobacco dryer, where an air fan, a water valve, and two air hatchways are represented in their real locations.

(b) Spatial distribution of the measurement motes inside the drying chamber: next to the supervision window (green), .

..A WSN with MEP510 motes from the Crossbow Company was employed to take the measurements. This mote has a SHT11-Digital Humidity Sensor Batimastat from the Sensirion Company to acquire both temperature and relative humidity. LabVIEW and Matlab language AV-951 programming tools were used to develop applications to acquire data from the WSN, to store it in a structured database, and to develop an ANN to estimate and predict these data.2.2. Description of the Tobacco Drying ProcessThe specific tobacco drying process monitored in this research was based on increasing the tobacco dryer chamber temperature in order to extract the plants’ moisture. To this end, a time-varying target temperature as that shown in Figure 2, which is a typical target temperature evolution in this type of dryer [5], was chosen.Figure 2.Target temperature evolution in the analysed drying processes.