A robust sliding mode flux and speed observer for speed sensorless control of an indirect field oriented induction motor drives
Introduction
Control of an induction motor without mechanical sensor has been widely used because of some advantages of the sensorless control such as reliability and low level of maintenance. Various methods to implement the sensorless control can be summarized as; using voltage and current signals available from the drive system, through the carrier frequency signal injection or creating saliency by changing the machine rotor structure. Last two methods mentioned above suffer from the disadvantage that they need either extra hardware or special rotor manufacturing and therefore, cannot be used for an off-the-shelf induction machine. The much-preferred choice for sensorless control for an off-the-shelf machine is through flux and speed estimation using terminal quantities. Therefore, the focus of this paper is flux and speed estimation using the voltage and current signals only.
Slip frequency control [1] and field orientation control [2] are the two major techniques for high performance sensorless control of induction machine. The slip frequency control has been documented to be more sensitive to the rotor resistance variation [3], [4], [5]. Many on-line identification schemes of the rotor time constant have been designed [6], [7], [8]. These methods have provided some improvement, but are quite complex because they either require more parameters or have hardware complications. Some fuzzy logic based techniques [9], [10], [11] have been proposed to overcome the detuning. However, these solutions are also very complex with respect to the software and require extensive calculation that put extra load on the processor.
The proposed observer in this paper estimates the machine speed as well as the rotor time constant and therefore, overcomes the problems, caused by rotor resistance variations, inherited by the slip frequency control. The sliding mode flux observers for induction machine have been investigated [12], [13], [14], [15], [16], [17]. However, most of the studied observer structures depend heavily on the machine parameters. In this paper a new sliding mode flux observer structure is proposed such that the convergence of the observed flux is guaranteed by the convergence of the observed currents. Once the convergence of the observed flux is guaranteed, then the rotor speed and the rotor time constant are found through the equivalent control. To avoid using sensors on the machine, terminal quantities of the machine are used to estimate the fluxes and speed of the machine. In this case, the success in achieving the field orientation depends heavily on how well the rotor flux position is estimated. To solve this problem, different algorithms are proposed. These proposed algorithms are categorized in two basic groups [18]. First one is ‘the closed-loop observers’ where the feedback correction is used along with the machine model itself to improve the estimation accuracy [19], [20]. Second one is ‘the open-loop observers’ in a sense an on-line model of the machine, which do not use the feedback correction [21], [22]. One of the main problems for both of those observer structures is the integration process inherited from the induction machine dynamics, and some work is based on cancellation strategies to avoid the integration effect. The other important problem is insufficient information about the machine parameters, which yield the estimation of some machine parameters along with the sensorless structure. In this study, a closed-loop sliding mode observer has been used and a low pass filter (LPF) has been used to solve the integration effect. When the motor frequency is lower than the cut-off frequency of the LPF, an estimation error will be produced. To estimate exactly stator flux in a wide speed range, the LPF should have a very low cutting frequency [22]. It is observed that when the filter cut-off frequency is higher than 15 Hz, the closed loop system performance degrades considerably, especially at low speed the closed loop system suffers from excessive noise and oscillations. From our experience, the cut-off frequency of the LPF has been chosen as 5 Hz in the implementation.
Section snippets
Indirect field oriented (IFO) sensorless control structure
The slip frequency proposed sensorless control scheme is shown in Fig. 1. In this drive system, the inner feedback loop performs the synchronous current regulation. The current command is produced by the outer speed control loop based on the command speed and the observed speed . This speed regulation is conventionally done by using a PI controller [23].
The induction machine model is defined by the stator currents and rotor fluxes as state variables in rotor flux oriented
Design of the sliding mode current and flux observers
The proposed speed and rotor time constant estimation structure is based on sliding mode current and flux observers. Ensuring the convergence of the current observer, the equivalent control is produced. Then, it is used in the flux observation to produce fluxes along the d and q axes. Once the flux values are found, then the rotor speed and rotor time constant are estimated by using observed fluxes. For clarity (1), (2), (3), (4) can be written as
Simulation results
The block diagram of the indirect field oriented induction machine drive system with observer structure is given in Fig. 1. Note that the machine parameters used in this study are 220 V, 14.8 A, 5 Hp; Lls = Llr = 1.9 mH, Lm = 41.2 mH, 1800 rpm; Rs = 0.6 Ω, Rr = 0.412 Ω, four poles.
The validity of the observer structure is verified by the simulations, which are given in this section. Note that in the simulations the observed speed is used as feedback in the closed loop and for the initial value of the rotor time
Experimental results
The laboratory setup consists of a 5 Hp cage rotor induction machine and a high performance advance controller for electric machines (ACE). For high-speed performance, 900 rpm triangular, 700 rpm trapezoidal and 300 rpm step reference inputs are applied to the closed loop system. In addition for the low speed performance a 50 rpm square-wave reference input is applied. In the following figures these results are presented. External disturbances or load have been applied in some cases as well.
To be
Concluding remarks
A new sliding mode flux, speed and current observers are proposed in this paper. The flux observer accuracy is guaranteed through the current observer. The error between the actual current and observed current converges to zero, which guarantees the accuracy of the flux observer. Under this condition the rotor speed and rotor time constant are estimated. The proposed observers do not require any knowledge of the machine speed or rotor time constant. The rotor time constant update algorithm will
Nihat Inanc received the B.Sc. and M.Sc. degrees from the Istanbul Technical University, Istanbul, Turkey in 1988 and 1992, respectively, and Ph.D. degree from the Kocaeli University, Kocaeli, Turkey in 1996, all in electrical engineering. From 2000 to 2001, he worked as a visiting scholar at the Ohio-State University in the USA. He is currently working as an associate professor in the Electrical Engineering Department at the Yuzuncu Yil University in Van, Turkey. His research interests are in
References (23)
A new sliding mode flux and current observer for direct field oriented induction motor drives
Electric Power Syst. Res.
(2002)- et al.
Vector control of induction motor without shaft encoder
- et al.
Speed sensorless field-orientation control of the induction machine
IEEE Trans. Indus. Appl.
(1993) - et al.
Implementation of robust flux observer based field orientation controller for induction machine
- et al.
The influence of motor parameter deviations in feed forward field orientation drive system
IEEE Trans. Indus. Appl.
(1985) - et al.
Study of parameter sensitivity in high-performance inverter-fed induction motor drive systems
IEEE Trans. Indus. Appl.
(1987) - et al.
Secondary resistance identification of an induction motor applied model reference adaptive system and its characteristics
IEEE Trans. Indus. Appl.
(1987) - et al.
A rotor parameter identification scheme for vector-controlled induction machine motor drives
IEEE Trans. Indus. Appl.
(1985) - et al.
An extensive Kalman filter approach to rotor time constant measurement in PWM induction machine drives
- et al.
On-line fuzzy tuning of indirect field oriented induction machine drives
Fuzzy logic enhanced speed control of an indirect field-oriented induction machine drive
IEEE Trans. Power Electron.
Cited by (0)
Nihat Inanc received the B.Sc. and M.Sc. degrees from the Istanbul Technical University, Istanbul, Turkey in 1988 and 1992, respectively, and Ph.D. degree from the Kocaeli University, Kocaeli, Turkey in 1996, all in electrical engineering. From 2000 to 2001, he worked as a visiting scholar at the Ohio-State University in the USA. He is currently working as an associate professor in the Electrical Engineering Department at the Yuzuncu Yil University in Van, Turkey. His research interests are in power electronics, motor drives and their control systems.