Regression Model for Tree Trunk Temperature for Energy Harvesting

In recent studies, it has been proved that the temperature difference in the cross section of a tree trunk can be used to generate electrical energy by using semiconductor thermoelectric devices and, as an example of application, working as a battery to power sensors for environmental monitoring. To know the behavior of the temperature distribution within a tree trunk is very important to infer the amount of electrical energy that can be generated. In this paper, the authors present a linear regression model for the temperature distribution within a tree stem based on weather condition parameters, as well as on core and bark temperatures measured onsite in Brazil. This model contributes to a proof of concept for the feasibility of using trees as an ambient power source for wireless low-power devices, and it is purely based on experimental results. As weather parameters are readily available, and the temperature data collected are of typical tree types in the region (mango trees of similar age and physical parameters), our model should generalize easily to help gain understanding of the thermoelectric currents than can be generated in that region.