Av Alexey Pavlov

The local output regulation problem: convergence region estimates

In this paper, the problem of local output regulation is considered. The presented results answer the question: given a controller solving the local output regulation problem, how to estimate the set of admissible initial conditions for which this controller makes the regulated output converge to zero. Theoretical estimation results and an estimation procedure explaining the application of these results in practice are presented. An example of the disturbance rejection problem for a mechanical system (TORA system) is given as an illustration.

Real-Time Rotor and Stator Resistances Estimation of an Induction Motor

Abstract In this paper, four schemes for the real-time identification of induction motor parameters are presented. The first and second schemes are for separate identification of the rotor and stator resistances while the third scheme performs simultaneous identification of these two parameters. The signals used in the identification procedure are the measured stator current, voltage command and rotor speed. The fourth scheme for the rotor resistance identification require neither measuring nor estimating of the rotor speed. All schemes are based on the MRAS identification approach and require persistent excitation conditions to be satisfied.

Adaptive observers for sensorless control of an induction motor

Novel adaptive speed and flux observers are designed as a part of the speed sensorless control scheme of an induction motor (IM). The design is based on a special form of IM equations with filtered current and voltage command signals as inputs. The speed tuning algorithm incorporates the stator current observer with gradient or sign speed adjustment schemes. The proposed flux observer is based on an algebraic relation involving measured and filtered signals and it shows better performance than the conventional flux observers in the low speed region and during transients. Various simulation and experimental results are presented to demonstrate the potential and practicality of the approach.

Convergent systems and the output regulation problem

The notion of convergent systems is revisited from a control perspective. Sufficient conditions for a system to be convergent in a set are given. It is shown that a nonlinear system with an asymptotically stable linearization at the origin is convergent in some neighborhood of the origin. These results are applied to the output regulation problem. Based on the convergence property, a procedure for estimating the set of admissible initial conditions for a solution to the local output regulation problem is presented. An application of the procedure is illustrated by an example.

Direct torque and flux regulation in sensorless control of an induction motor

A novel scheme for direct torque and flux control of induction motors (IM) is presented. The scheme does not use a transformation to the rotor flux related frame and it guarantees the asymptotic stability of the torque and flux tracking. To perform the controller design a torque-flux dynamic model with stator voltage as an input is obtained and separation of the torque and flux regulation is achieved. The controller is elaborated as a part of the sensorless control scheme of IM and it shows robustness with respect to rotor time constant changes. For slow reference commands a control scheme based on the steady state solution of the torque-flux model is also analyzed. It is significantly simpler than the dynamic scheme and has better robustness properties with respect to the current noise. The controllers can also be applied if the rotor speed signal is available. The analysis of the stator current and voltage constraints determines the allowable set of the torque and flux reference commands. Simulation results are presented to verify the proposed approach.

Real-time identification of an induction motor using sinusoidal PWM voltage signals

In this paper, a novel technique is presented for real-time identification of induction motor (IM) parameters. The method is based on the IM model in the synchronously rotating frame and utilizes sinusoidal PWM voltage signals for parameter identification tests. The proposed algorithm provides accurate estimation of the stator and rotor parameters based on stator current measurement and voltage command. As compared to the commonly used techniques for the tuning of the slip gain the IM torque measurement is not required. Validation simulation and experimental results are presented.

The uniform global output regulation problem

In this paper, we present a new approach to the global output regulation problem. In this approach, a controller should be designed in such a way that the closed-loop system is uniformly convergent. This requirement allows to extend the theory developed for the local output regulation problem to what we call the uniform global output regulation problem. Such extension is made using global invariant manifold theorems, which serve as global counterparts of the center manifold theorems. It is shown that within the proposed approach solvability of the (extended) regulator equations is a basic necessary condition for the solvability of the uniform global output regulation problem. As an illustration, we present a solution to the uniform global output regulation problem for a class of nonlinear systems.