Thermal performance prediction of radial-rotating oscillating heat pipe by a novel fusion model: A case study of application in grinding

High temperatures in rotating machinery and machining processes necessitate effective thermal management for improving efficiency. Radial rotating oscillating heat pipes (RR-OHPs) have shown potential in enhancing heat transfer during rotating machinery operation. To facilitate the application of RR-OHPs in rotating machinery, an accurate prediction of their thermal performance is essential. This paper proposes a so-called “fusion prediction model” that combines an algorithm called GA-LightGBM and grey prediction techniques to accurately predict the thermal performance of RR-OHPs under different parameters. The GA-LightGBM optimizes hyperparameters globally though population iteration, while grey prediction explores systematic patterns using limited information. The combination of these algorithms results in a high-precision prediction with a relative error of 10%. The dataset for the model is obtained by an experiment under varying heat inputs and rotating speeds. To illustrate the application of the fusion model, we study the design of an OHP grinding wheel for enhanced heat transfer in grinding processes. The results confirm the high thermal performance of the OHP grinding wheel maintaining the maximum grinding temperature below 300 °C. Overall, this proposed prediction model is expected to expand the application of RR-OHPs and provide valuable guidance for their implementation in engineering.