Abstract
A control trade-off is made for the speed calculation using incremental signals for asynchronous motor control since the speed information has a dual requirement: from standstill the speed needs to be accurately calculated but at continuous operation an update rate of the speed information of once a revolution would be sufficient. This results in a challenging dimensioning of the sensor-bearing output: either resolution (the number of pulses per revolution) or accuracy of the pulses is preferred. To accomplish this task, SKF has developed a methodology to dimension the sensor such that efficiency and torque quality of the asynchronous motor can be optimized.
To determine the best suitable sensor-bearing configuration for which the motor performance is optimized, a DoE is proposed. It uses simulated populations of SKF sensor-bearings with two quadrature incremental output signals with different specifications. These sensor populations are used to generate speed errors using known speed calculation algorithms with two variants: one for high speed (counting the number of edges in a certain time window) and one for low speed (calculating the time between two pulses). These speed errors are included in the simulation model to map the torque ripple (a measure of the torque quality) and in the hardware-in-the-loop set-up where dSPACE includes the generated sensor errors in the feedback loop. In this way, the trade-off can be made between favouring resolution or pulse accuracy on the sensor signals.
The results show that when measurements with the worst sensors (highest resulting peak-to-peak speed error) of the sensor-bearing populations are performed, the loss of accuracy for higher resolution sensor-bearings is not compromising the system output accuracy; on the contrary, the higher resolution makes that for the same torque ripple, the sensor accuracy can be lower. Therefore the higher resolution sensor output is preferred (for the tested type of asynchronous motor and control) and this effect is quantified.
KEYWORDS – induction (asynchronous) motor, modeling, hardware-in-the-loop, component, control system