Developing an empirical model to predict the winding coil temperature for a compact wind turbine

Abstract

One of the problems with a compact wind generator is a temperature control. The aim of this research is to establish an empirical model for estimating the winding coil temperature through natural convection cooling. The modeling process began with physical experimentation at different operating conditions while measuring temperature at various positions on the compact wind generator. Later, simulation experiment is divided into two sections: Electromagnetic (EM) simulation and Computational Fluid Dynamic (CFD) simulation. The electromagnetic simulation was applied and tested successfully, with a Vpeak discrepancy of less than 4 percent. While the CFD software was then used to replicate the natural convection cooling effects at the compact generator. The CFD results were validated by experimental measurements with a difference in winding coil temperature of less than 9 percent. Then, using Design of Experiment (DOE), various operating conditions were simulated to evaluate the results and Analysis of Variance (ANOVA) was used to develop the empirical model. The empirical model for estimating the winding coil temperature with a difference of less than 10% from experimental data was successfully developed. As a result, the empirical model can be used to estimate the winding coil temperature at different operating conditions with a good accuracy.