Abstract:A transport container with a controlled temperature was developed, where the temperature was regulated using low-temperature phase-change materials. The cold energy was first stored in low-temperature phase-change materials and then released when the temperature in the container was out of target range under an intelligent control system. However, there were still some issues that need to be solved, such as the difficulties in controlling the release rate of cold energy, the prediction of remaining cold energy during the transportation work. The release rate of cold energy depended directly on the convective heat transfer coefficient between the surface of the cold storage plate and the ambient air. In this study, an experimental platform was developed to investigate the influence of different environments and parameters of cold storage plates on the convective heat transfer coefficient between the cold storage plate surface and the ambient air. A quadratic regression orthogonal experiment was adopted to clarify the coupling effects among the factors, including the air velocity and temperature at the entrance of the cool storage area, heat transfer area of the cold storage plate, and the space between them on the surface convective heat transfer coefficient. After that, the experimental data were analyzed. A second-order prediction model of surface convective heat transfer coefficient was built that the relationships between the influence factors and the surface convective heat transfer coefficient and the factors with significant effects were obtained, as well as the optimal values of such factors. Consequently, there was the most significant interaction between the entrance air temperature and the heat transfer area of the cold storage plate. The prediction model of surface convective heat transfer coefficient built by response surface method presented a higher accuracy, where the best combination of parameters was velocity=4 m/s, temperature=25 ℃, area=0.455 m2, spacing=0.04 m, and the determination coefficient value was 0.927 4 and the coefficient of variation was 5.78%. The calculated results of such regression model were in good agreement with the experimental, with the maximum error of 3.58% and an average relative error of 2.69%, indicating that such model can be used to quickly and accurately predict the convective heat transfer coefficient between the surface of cold storage plate and the ambient air under different conditions. The finding can provide accurate control on the release rate of cold energy in the temperature phase-change materials, and the prediction of remaining cooling energy for transport containers with controlled temperature.