2023 ASEE Annual Conference & Exposition

This paper presents a new testing method for an induction machine’s torque-speed curve. The incumbent testing method for determining an induction machine’s torque-speed curve uses a DC generator coupled to the shaft of the induction machine,[1] delivering the energy to an external resistive load bank. Consistent and precise steady state characterization is readily obtained from this setup and then used to verify an appropriate model. This proposed new method uses a squirrel cage induction machine driven by either variable speed regenerative drive or a synchronous machine, replacing the coupled DC machine and load bank. This opens an insightful new way to operate induction machines in a teaching laboratory.

In this new configuration, the test machine is electrically connected at its specified frequency, typically 50 Hz or 60 Hz. The variable speed drive runs the load machine over a range of frequencies between the difference and sum of the specified frequency and the test machine’s rated slip. Within these limits, the torque-speed curves of the test machine and the load machine intersect safely at a two unique points of operating torque and speed for each frequency applied to the load machine. One of these points is always stable and within the test machine’s normal operating range. Torque may be measured by a transducer or calculated using data from a model of the load machine. In addition to torque and speed, other important operating information and test data about the test machine can be measured, such as current, power factor, and energy efficiency. For better noise performance, a synchronous machine may be used to supply the load machine at variable frequency.

The context for this new test method is a twin simulation of a wind turbine generator. The test machine is a Type III generator. The load machine is configured to mimic the wind turbine that is directly coupled to the generator. Together, they provide a hardware test platform for a wind turbine generator. The software twin of this system is a complete digital simulation of it. Ideas can be tried and improved in software before verifying them on the hardware. Of course, a good software model is a must for such a test bed concept, hence this paper on how to obtain important performance information without the expense of decoupling and modifying the test bed.

The paper will describe how the test bed performed with students in the lab for both teaching and research. This will be supported by students’ simulation and experimental results.

[1] “IEEE Standard Test Procedure for Polyphase Induction Motors and Generators,”IEEE Standard 112, 1991.