Softwood Pallet Stringer Temperature Estimation

Abstract

Monitoring temperature inside a pallet specimen during a test could be challenging. In this study, two methods were used to estimate temperature in a softwood pallet stringer at the time of testing based on the initial temperature of when it was removed from a temperature chamber and the duration of when it was removed from a chamber until the time it was tested. Five cooling down and three warming up temperature profiles were collected using thermocouples. In the first method, an artificial neural network was developed based on the collected data. In the second method, a mathematical model was suggested based on heat transfer principles. Collected data was used to validate the model. Both methods yield satisfactory results. The heat transfer model allows temperature estimation for specimens with different thickness and species, while the neural network is more precise but limited to the specimen used. Both methods allow other researchers to estimate the temperature without having to collect temperature data.