Sensorless control of permanent magnet synchronous machines with multiple saliencies
Final Report Abstract
This project is carried out mainly to solve the sensorless control problem of the SPMSM with strong multiple saliencies (cwSPMSM). The two questions (1) how to decouple or combine stronger multiple saliencies, (2) initial position estimation in the presence of multiple saliencies, are answered. The emphasis of this project is on the saliency tracking based sensorless method due to the fact that the sensorless control problem in the low speed range arising from multiple saliencies of the cwSPMSM is more challenging. Firstly, the experimental results are obtained to build the special saliency model of the cwSPMSM. The explanation of causing strong multiple saliencies is also given. Then, three kinds of saliency tracking based sensorless methods are proposed from different perspectives. The first one is the Multi-Signal Injection method, which makes use of the dependency of the saliency signal on the frequency of injected HF signal. It combines the results of two HF signals and synthesizes a new signal that has the same behavior as the primary saliency signal. And then, the failed traditional demodulation methods become valid again. The MSI method is an effective combination method but it consumes more energy and brings more acoustic noise to the system. The second one is the Secondary Saliency Tracking method, which belongs to the decoupling group. This method utilizes the rotation difference of the primary saliency signal and secondary saliency signal to separate one from another. In this method, the secondary saliency signal is extracted out from the NSC by means of a specially designed BPF and an ANF. Moreover, an initial position estimation scheme is proposed accordingly as well to start the cwSPMSM smoothly without heavy load. However, the start at standstill with full load is still challenging for this method and the decoupling performance around zero speed is not satisfactory. The third one is the Repetitive Control method, whose principle is also decoupling. This method treats the secondary saliency signal as a disturbance and builds an RC model to eliminate it from the NSC. The new concept of an angle domain RC is proposed in the method and shows satisfactory tracking performance. Similar to the previous methods, it has a drawback i.e. the divergence possibility in case of a drastic dynamic change. The proposed three methods have advantages as well as disadvantages as explained. No single method can solve all the problems. In addition, the methods are designed for low speed operation and they all require the injection of HF signals. When the speed of the cwSPMSM goes higher, it is mandatory to change to a fundamental model based method to eliminate the energy consumption and acoustic noise caused by HF signals. All the above proposed methods are carried out on an experimental testbench, which is introduced. Three improvements are made to further enhance the performance of the existing controller. Firstly, an SVPWM algorithm is designed in the FPGA to increase the utilization of the DC link voltage and meanwhile release the calculation burden of the CPU. Secondly, a resolver demodulation circuit and relevant decodesoftware are developed as well to make the controller adaptive to different interfaces. Thirdly, the C code and the VHDL code generation through Matlab/Simulink are achieved, which enables the designers get rid of the programming burden and accelerate the development progress. They will become more and more popular in various applications in the future. In this project, a complete solution for the sensorless control of a cwSPMSM is provided and some key questions are discussed in corresponding publications. The comparative studies of saliency tracking based method, fundamental model based method and parameter identification are given throughout in the dissertation. It can be found that the same problem can always be solved by different methods created from different perspectives. Therefore, to balance the benefits and drawbacks is the key criteria for selecting the proper method, which is fit for other occasions as well.
Publications
- “Hybrid Sensorless Control for SPMSM With Multiple Saliencies”, 41st Annual Conference of the IEEE Industrial Electronics Society, 2015
Z. Chen, Z. Zhang, G. Luo and R. Kennel
(See online at https://doi.org/10.1109/IECON.2015.7392261) - "A Novel Nonlinear Modeling Method for Permanent-Magnet Synchronous Motors," in IEEE Transactions on Industrial Electronics, vol. 63, no. 10, pp. 6490-6498, Oct. 2016
G. Luo, R. Zhang, Z. Chen, W. Tu, S. Zhang and R. Kennel
(See online at https://doi.org/10.1109/TIE.2016.2578839) - "Enhanced Sensorless Control of SPMSM based on Stationary Reference Frame High-Frequency Pulsating Signal Injection," IEEE IPEMC 2016-ECCE Asia, Hefei, China, May 22-25, 2016
Z. Chen, F. Wang, X. Cai, and R. Kennel
(See online at https://doi.org/10.1109/IPEMC.2016.7512403) - “Decoupling of Secondary Saliencies in Sensorless PMSM Drives Using Repetitive Control in an Angle Domain,” J. Power Electron., Vol. 16, no. 4, pp.1375-1386, July 2016
C. Wu, Z. Chen, R. Qi and R. Kennel
(See online at https://doi.org/10.6113/JPE.2016.16.4.1375) - “Secondary Saliency Tracking- Based Sensorless Control for Concentrated Winding SPMSM,” IEEE Trans. Ind. Informatics, vol. 12, no. 1, pp. 201–210, Feb. 2016
Z. Chen, F. Wang, G. Luo, Z. Zhang and R. Kennel
(See online at https://doi.org/10.1109/TII.2015.2503639)
