Calibrating a Motor Controller Through Test
In an electric powertrain the motor, inverter, and gear box need to work together to give the desired performance and experience for the end user. Ideally, these components should be designed to work together, but that is rarely the case, as the parts come from a variety of suppliers. All the components have different performance and efficiencies at the different voltages, temperatures, and loadings throughout the range of operation. Fortunately, engineers have a wide variety of variables in the control computer that they can alter to optimize the performance of the powertrain. While computer models are a powerful tool for this, the reality is complex, and engineers need to perform many tests for the best set of variables.
Motor Controller Background
In the electric powertrain, the inverter drives the motor to achieve the torque and speed targets that are demanded by the user. The inverter receives commands from a computer often referred to as the “motor controller.” The goal of this controller is to move from one speed safely and efficiently to another, while meeting the performance demands of the user. The controller achieves this safe and efficient transition by knowing exactly how the electric motor will apply torque based on an applied inverter voltage.
To implement a controller that will know how the motor torque reacts to an inverter, all parts of the powertrain need to be characterized and mathematically modeled. This model will then be implemented into hardware with sensor inputs for feedback on the current operation state (figure 1). The development of this model can be difficult because each component requires different operations at different temperatures, voltages, or fault conditions. To accurately build a model, engineers need to test the powertrain in all the conditions that the product will experience throughout its operation resulting in hundreds of tests and significant time required to optimize the control strategy.