What are some features and limitations of the Flux Observer in the Motor Control Block set?
3 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
Amit Mathukiya
am 8 Jul. 2020
Beantwortet: Joel Van Sickel
am 1 Sep. 2020
Hi Team,
Working on a sensorless motor control application where full (max.) torque is required from zero speed and it is very dynamic in nature. I would like to know more about the features and limitations of the Flux Observer in the Motor Control Block set. It would be great if someone could answer following questions.
- Does it depends on the type of the motor? i.e. PMSM, SPM, IPM, BLDC, and ACIM or it works independent of the type of motor used?
- Since it is a flux observer, does it has any dependency of the saliency of the motor?
- How is the starting sequence is initiated? ( transition from open loop to close loop).
- What's the approx time it takes to converge when starting from zero speed?
- Is it possible to operate it in all four quadrants? (motoring and generating).
- Any information on transient response behaviour?
- Any information on typical/base line execution time of the observer? For example, it takes x us when executing on the 32-bit, 100 MHz, Single-Precision Floating-Point Unit target.
Best,
Amit
0 Kommentare
Akzeptierte Antwort
Joel Van Sickel
am 1 Sep. 2020
Hello Amit,
the particular algorithm in this observer utilizes known voltage, current, and resistance to estimate flux using standard PMSM equations. It will work for all quadrants of operation and given it's simple implementation can work for various types of motors and does not depend on saliency. It is an open loop observer and does not have transient or convergence behavior other than estimating traditional motor dynamics. it is best to start the observer as soon as voltage is applied to the motor to reduce potential error, but nothing needs to change if you switch between open and closed loop control. It has two resetable states that are tracking flux.
Because it directly calculates flux from voltage and current with assumed resistance, it is an efficient and light algorithm. I do not have exact execution time data for a specific piece of hardware, but there have been no issues using it on standard TI platforms.
As it is a prebuilt block and easy to use, I would recommend requesting a trial from your sales representative and seeing how it works under specific scenarios to see if it is actually accurate enough for your application.
For references that this observer was based off of, please see:
[1] O. Sandre-Hernandez, J. J. Rangel-Magdaleno and R. Morales-Caporal, "Simulink-HDL cosimulation of direct torque control of a PM synchronous machine based FPGA," 2014 11th International Conference on Electrical Engineering, Computing Science and Automatic Control (CCE), Campeche, 2014, pp. 1-6. (doi: 10.1109/ICEEE.2014.6978298)
[2] Y. Inoue, S. Morimoto and M. Sanada, "Control method suitable for direct torque control based motor drive system satisfying voltage and current limitations," The 2010 International Power Electronics Conference - ECCE ASIA -, Sapporo, 2010, pp. 3000-3006. (doi: 10.1109/IPEC.2010.5543698)
Regards,
Joel
0 Kommentare
Weitere Antworten (0)
Siehe auch
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!