Optimized Transient Excitation of Induction Machines
Optimized Transient Excitation of Induction Machines
Disciplines
Electrical Engineering, Electronics, Information Engineering (100%)
Keywords
-
Sensorless Control,
Pulse With Modulation,
Noise Reduction,
Tansient Excitation,
Current Sensors
With the development of fast signal processors and fast power electronic devices ac drives have replaced dc machines in various industrial applications. With the demand to reduce the costs of these drives as well as to increase the reliability many attempts have been made to realize torque/speed control without the shaft sensor. Up to now it seems that the only way to determine the flux/rotor position in the whole operating range including zero flux frequency without shaft sensor is to establish a high frequency or transient excitation of the machine and to exploit the response of the machine. Besides sensorless control this transient response can also be used advantageous for condition monitoring of machines. Due to the inherent saliencies of induction machines like saturation or slotting, this transient response is influenced by the flux and rotor position of the machine which in turn can thus be estimated. The high frequency or transient excitation has to be superposed to the fundamental wave excitation of the machine and thus influences the current control, the torque production and above all the noise emission. The goal of the proposed project is to develop a measurement set up and an inverter control scheme which includes and finds the optima for both, the fundamental wave excitation necessary for the torque production as well as the transient excitation necessary to estimate the flux/rotor position and to realize effective condition monitoring. Thus an optimization will be made to reduce the side effects of the transient excitation like noise emission and currentorque ripple. In addition, the measurement setup to determine the transient response of the machine will also be optimized in order to obtain a high signal quality. To achieve this, special hardware will be developed to implement the optimized switching pattern, to perform all necessary measurements of the machines current response and to calculate the estimated flux/rotor position. The proposed project is based on results obtained from a current project financed by the FWF
Induction Machines have, due to their robustness, encountered a lot of application fields as electrical drives. However, to achieve high dynamic performance of an induction machine drive in all speed ranges, an encoder or speed sensor has to be attached to the drive system. In the past, approaches were started to replace these electro- mechanical sensors by purely electrical measurements at the machine terminals, reducing the size and costs and at the same time increasing the reliability. In this project, the optimization of a mechanical sensorless control method called INFORM (Indirect Flux Detection by Online Reactance Measurement) for induction machines is addressed. The method can be used to replace the encoder or the speed sensor especially in the low speed region where other methods are unstable. For that purpose, the method extracts the transient (high-frequency) electrical properties of the induction machine with the help of injection pulses generated by the power electronics. The research was mainly focused on the integration of the necessary pulse excitation into a standard pulse width modulation (PWM) in order to minimize the resulting disturbances. The injection of voltage pulses turned out to deliver a lot of important information of the induction machine at first , but as turned out it also has some adverse effects. The most disturbing one was the resulting audible noise due to the injection pulses. Additionally, the current controller was influenced by the injection sequence. Now the main goal of this research project was to develop a method for an integration of the test pulses into the standard PWM, thus reducing the audible noise as well as the problematic interaction of the injection scheme with the current control. Starting with some quite promising ideas, the research came up with the problem of limitations by the electrical sensors, yet not optimized for the extraction of transient phenomena in electric drives. In this field, the need for an improved sensor system was raised, which would cover the needs for a fast and robust measurement. For this purpose the improvement of specially redesigned current transducers was a big issue that had to be addressed for a proper integration of the injection pulses into the PWM. As turned out during the investigations, the considered exploitation of the machines transient electrical behavior also has additional fields of application outside mechanical sensorless control, for instance in the field of on-line condition monitoring of inverter-fed ac-machines. The improvements made during this project regard the sensitivity of the measurement chain especially in the field of derivative sensing of currents, as well as the injection pattern optimization. With respect to the mentioned disturbing effects the performance of the control method could be almost doubled, which means a big step towards a possible industrial application in the future.
- Technische Universität Wien - 100%
Research Output
- 55 Citations
- 6 Publications
-
2005
Title Using Optimized Transient Excitation Technique to Stabilize Model-Based Sensorless Control of Induction Machine DOI 10.1109/epe.2005.219479 Type Conference Proceeding Abstract Author Wolbank T -
2005
Title Influence of Inverter-nonlinearity and Measurement Setup on Zero Speed Sensorless Control of AC Machines Based on Voltage Pulse Injection DOI 10.1109/iecon.2005.1569138 Type Conference Proceeding Abstract Author Wolbank T Pages 1568-1573 -
2005
Title Sensorless Control of Induction Machines by Combining Fundamental Wave Models with Transient Excitation Technique DOI 10.1109/iemdc.2005.195902 Type Conference Proceeding Abstract Author Wolbank T Pages 1379-1384 -
2004
Title Closed-Loop Compensating Sensors Versus New Current Derivative Sensors for Shaft-Sensorless Control of Inverter Fed Induction Machines DOI 10.1109/tim.2004.830561 Type Journal Article Author Wolbank T Journal IEEE Transactions on Instrumentation and Measurement Pages 1311-1315 -
2004
Title Combination of Signal Injection and Neural Networks for Sensorless Control of Inverter Fed Induction Machines DOI 10.1109/pesc.2004.1355480 Type Conference Proceeding Abstract Author Wolbank T Pages 2300-2305 -
2003
Title Detection of Rotor Eccentricity Caused Air Gap Asymmetries in Inverter Fed AC Machines based on Current Step Response DOI 10.1109/peds.2003.1282884 Type Conference Proceeding Abstract Author Wolbank T Pages 468-473