LAUSR

Accurately getting thermal limits (peak capacity and duration, maximum temperature, thermal dynamic, etc.) is very useful to avoid damage, verify reliability, reduce size, and determine overloading capability of motors and power devices. Conventional model-dependent thermal analysis often leads to complexity and requires empirical knowledge. If feasible, experimentally evaluating one thermal behavior through another test is not only simple but also fast and accurate. By exploring model basic properties not a precise model, this article provides a “superposition-based” simple equivalent temperature rise test method. Unlike conventional experiments with persistent loading until the stable high-temperature point, the proposed method is able to reconstruct high-temperature data from much lower temperature data by short-term loading, avoiding the risk of permanent damage and reducing the energy consumption. It works for various heat transfer relations, cooling conditions, and loss compositions, as validated by the experimental results. No cumulative error occurs. The nonlinearity influence and the corresponding linear approximation are analyzed in detail. Discussions on inherent nonlinearity influence reduction are provided. Two “superposition-based” accuracy improvements for the nonlinear heat transfer coefficient and variant power loss are also provided. It could be suitable for evaluating the overloading ability or thermal limits without damage risk and with reduced energy consumption.