Pavel Brandstetter

Estimation Techniques for Sensorless Speed Control of Induction Motor Drive

Rotor position and speed sensors are required for vector control of induction motor. These sensors are sources of trouble, mainly in hostile environments, and their application reduces the drive robustness. The cost of the sensors is not also negligible. All the reasons lead to development of different sensorless methods for rotor position and mechanical speed estimation in electrical drives. The paper deals with the speed estimators for applications in sensorless induction motor drive with vector control, which are based on application of Kalman filter and artificial neural network. The development and DSP implementation of the speed estimators for applications in sensorless drives with induction motor are described in the paper

Rotor time constant adaptation using Radial Basis Function network

Our intention here was to highlight a replacement of adaptation algorithm in MRAS by the help of alternative artificial neural network (ANN) which has received great attention in recent years. The main objective was to find and design some alternative neural network within the electric drive control. After a short discussion of hardware components, an overview of radial basis function (RBF) neural networks will be given. Digital signal processors TMS320F2812 are used for these electric drives control applications. The hardware accessories used within the electric control drive included: interface board for the signal processor kit-developed in our department, and an 8 bit data-transfer microprocessor for data acquisition. The interface of the DSP is a general-purpose control system for power converters in the electric drives. The next section briefly outlines the estimation of the rotor time constant, which is necessary for the so-called current model. The current model is used in the vector control of the induction motor and is utilized to determine the quantities for the transformation from the stationary reference frame into the reference frame, which is oriented on the rotor flux space vector. The estimation of the rotor time constant for the adaptive model of MRAS is created with the support of a PI-controller which is then replaced with the radial basis function network. The final section presents simulations results, which have been performed in the Matlab-Simulink software.

AC drive with permanent magnet synchronous motor and sensorless control

In the paper there are presented results of the sensorless control of AC drive with the permanent magnet synchronous motor. The method is based on the rotor magnetic flux components estimation. Practical realization was made by the microcomputer control system with the digital signal processor TMS320F2812. The graphical visualization of the electrical controlled drive was made by the software Lab View. The measured results of proposed laboratory stand with applied vector control and sensorless vector control were shown at the conclusion.

Rotor time constant adaptation with ANN application

In the paper there are presented results of method for rotor time constant adaptation with application of artificial neural network. The method employs the model reference adaptive system (MRAS). The adaptation algorithm is designed according to the Popovs criterion for hyperstability. The method is based on application of current model and voltage model of rotor flux in MRAS. The estimation of rotor time constant for adaptive model of MRAS is realized by the help of PI-controller. In next part of the paper there is described a replacement of adaptation algorithm in MRAS by the help of artificial neural network. The estimated rotor time constant is necessary for so-called current model. The current model is used in the vector control of the induction motor and serves for the determination of the quantities for the transformation from stationary reference frame into reference frame, which is oriented on the rotor flux space vector. Simulations have been performed in the Matlab-Simulink. The control algorithms are implemented using TMS320C2812 DSP. At the end of the paper some simulation and experimental results are provided to demonstrate the effectiveness of proposed method.

Sensorless induction motor drive with neural network

The development and DSP implementation of AI-based speed estimator for applications in sensorless AC electrical drives are described in the paper. Rotor position and speed sensors are mostly required for vector control of induction motor. But there exist some applications in industry, where these sensors cannot be used. In the situations sensorless control technique should be used. In this paper rotor speed estimator are considered, which is based on a feedforward artificial neural network. It is demonstrated that such estimator, in contrast to conventional model-based approaches, does not depend on a knowledge of machine parameters. It is further shown that accurate estimates can be obtained in situation of varying load without the need for explicit load monitoring.

Non-model based sensorless vector control of permanent magnet synchronous motor

For the control of a permanent magnet synchronous motor (PMSM) such as the vector control and the direct torque control, the rotor position is required to perform commutations between phases and to control speed and torque too. The sensorless control is popular for several reasons: cost saving, system reliability, etc. The paper presents a non-model based method for the sensorless vector control of the permanent magnet synchronous motor which employs an injection of high frequency voltages for rotor position estimation. This method works just only if the permanent magnet synchronous motor contain magnetic saliency. The saliency will be discussed too. The simulation of these methods, with a view to saturation saliency, is very difficult task. However, for testing of these methods and an initial tuning of the overall drive will showed some simulation results only with a small geometrical saliency.

Procedure to Choose Alternative ANN in the Control of Electrical Drive

Intention of this paper is to introduce the way how new types of artificial neural networks can be chosen in the control of electrical drives. The procedure is demonstrated through the use of rotor time constant adaptation method in the control of electrical drive. The procedure used to choose consists of determination of the problem, selection of proper artificial neural network, choosing of the new neural networks and finally new neural networks verification. The estimation of the rotor time constant for adaptive model of MRAS is created by the help of PI-controller and then is replaced by the radial basis function network. The estimated rotor time constant is then used in the vector control of electrical drive. The different architectures of RBF network in the field of adaptation of rotor time constant parameter is discussed here. Simulations have been performed in the Matlab-Simulink.

An Implementation on MATLAB Software for Stability Analysis of Proportional Controllers in Linear Time Invariant Control Systems

Stability range of proportional (P) controllers can be obtained using Routh-Hurwitz criterion for continuous linear time invariant (LTI) control systems or Bistritz criterion, Jury criterion for discrete LTI systems. Conditions from these criterions bring out inequalities. In case of high-order plants, these inequalities are very difficult to solve directly. In the paper, an algorithm is developed on MATLAB software to solve polynomial inequalites. With the support of this algorithm, stability criterions are implemented to find stability range of P controllers.

MRAS Observers for Speed Estimation of Induction Motor with Direct Torque and Flux Control

The paper describes Model Reference Adaptive System (MRAS) observers for the speed estimation of an induction motor with direct torque and flux control. The first estimator is a reference frame MRAS (RF-MRAS) and the second estimator is a current based MRAS (CB-MRAS). At first, direct torque controlled induction motor drive with two estimators are implemented on Matlab-Simulink environment. Then, comparison of two observers is done by evaluation of the rotor speed difference. The simulation results confirm that both MRAS estimators are simple to simulate and experiment. By comparison of both observers, the CB-MRAS observer gives higher accuracy of the rotor speed estimation.

Sliding Mode Observer for Induction Motor Control

The control of induction motor drives constitutes a vast subject, and the technology has further advanced in recent years. In control algorithms, continuous rotor position is mandatory. But the presence of encoder increases cost, reduces reliability. Therefore, elimination of this sensor is desirable. A sensorless of the vector controlled induction motor means the vector control without using any speed sensor. In the paper, a sliding mode observer and its applications in the sensorless control of the induction motor drive are proposed. The mathematical equations of induction motor, sliding mode observer and vector control are described in the paper. The stability of observer is proved base on Lyapunov theory. Simulation results are also presented in the paper.