Development of model based sensors for the supervision of a solar dryer

Solar dryers are increasingly used in developing countries as an alternative to drying in open air, however the inherent variability of the drying conditions during day and along year drive the need for achieving low cost sensors that would enable to characterize the drying process and to react accordingly. This paper provides three different and complementary approaches for model based sensors that make use of the psychrometric properties of the air inside the drying chamber and the temperature oscillations of the wood along day. The simplest smart sensor, Smart-1, using only two Sensirion sensors, allows to estimate the accumulated water extracted from wood along a complete drying cycle with a correlation coefficient of 0.97. Smart-2 is a model based sensor that relays on the diffusion kinetics by means of assesing temperature and relative humidity of the air inside the kiln. Smart-2 model allows to determine the diffusivity, being the average value of D for the drying cycle studied equal to 5.14 × 10−10 m2 s−1 and equal to 5.12 ×  10−10 m2 s−1 for two experiments respectively. The multidistributed supervision of the dryer shows up the lack of uniformity in drying conditions supported by the wood planks located in the inner or center of the drying chamber where constant drying rate kinetics predominate. Finally, Smart-3 indicates a decreasing efficiency along the drying process from 0.9 to 0.2

► The inherent variability of the drying conditions drives the need for achieving low cost sensors.
► Three model based sensors are proposed in this work.
► Three model based sensors are proposed in this work.
► They use the psychrometric properties of the air and the temperature oscillations of the wood.
► They allow estimating complex parameters as drying rate and wood moisture in real time.
► Their implementation on embedded electronics and wireless sensor networks will turn into smart sensors.