EPE Association: EPE 1991 - 15 - Session 2.4: PREDICTIVE AND ADAPTIVE CONTROL

EPE 1991 - 15 - Session 2.4: PREDICTIVE AND ADAPTIVE CONTROL
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 1991 - Conference > EPE 1991 - 15 - Session 2.4: PREDICTIVE AND ADAPTIVE CONTROL

   [return to parent folder]

   PREDICTIVE SPEED AND CURRENT CONTROL FOR D.C. DRIVES
By D. Schröder; H. Wanner [View]
[Download]

Abstract: Current- and speed-control of a d.c.-motor fed by a line commutated converter is a well known strategy with conventional control. But with these conventional control techniques the characteristics of the system are used only to a very restricted extent. With a special predictive control algorithm this nonlinear controlled system can be utilized to its physical limits. The predictive control algorithm has been realized on a multi-microprocessor system with the actual precalculation performed by a signal processor. A modification of the prediction parameter equations yields a greatly improved stationary accuracy. For identification of the process parameters a special method has been adopted that uses comparison of precalculated and real armature currents. The behaviour of the whole current control system is discussed. Especially the consequences for a superimposed speed controller are derived and a suitable speed control method is shown. Experimental results for the above topics are presented.

   A PREDICTIVE CURRENT CONTROLLER FOR SYNCHRONOUS SERVO DRIVES
By Hoang Le-Huy; Karim Slimani; Philippe Viarouge [View]
[Download]

Abstract: A real-time predictive current controller for PWM inverter feeding permanent-magnet synchronous motor is presented. Predictive current control scheme is applyied and the performance is studied and evaluated by computer simulation. The implementation of a predictive current controller using EPROM-based approach is described, in which calculations are executed by lookup tables. The experimental results obtained with a prototype 1-kW servo drive are presented and discussed.

   PREDICTIVE CASCADE CONTROL OF MACHINE TOOLS MOTOR DRIVES
By P. Boucher; D. Dumur; S. Daumuller [View]
[Download]

Abstract: We propose in this paper a real time algorithm developed for machine tools drives. The implementation of this cascade structure based on predictive control laws is possible due to the simplicity of numerical algorithms. A first part will present a polynomial form of the Generalized Predictive Control method, and give a structure of the equivalent linear controller. This approach can provide a real time loop quicker than with a "classical" algorithmic solution. Moreover, this polynomial form allows to study, in frequency domains, the influence of criteria parrameters, on the stability and robustness of the equivalent controller. All this will be shown, reformulating the expression of the control law for a cascade structure. Then a second part will develop and compare the results obtained on two kinds of motors, used as machine-tools drives, with very small sampling period: a brushless motor and a field oriented control induction motor. The advantages of the polynomial predictive cascade control law will be stressed in terms of rapidity, errors cancellations, disturbances rejection and robustness.

   ADAPTIVE CONTROL OF PM SYNCHRONOUS MOTOR
By L. Salvatore; S. Stasi; A. Dell'Aquila [View]
[Download]

Abstract: Permanent magnet synchronous motor drives are widely used in industrial applications because of their high performance. The use of field orientation control strategy gives external characteristics similar to those of well known dc drives without the need for slip rings and a field supply. In this paper we propose a new scheme for adaptive position control of a PM synchronous motor drive. This scheme employs an estimation method based on the Kalman filter theory. The algorithm can estimate the load torque and inertia moment quickly. The estimated torque is supplied to the PM synchronous motor to realize load compensation so to obtain no steady-state error and little transient-state position error. The speed feedback P-controller gain is adjusted by emploing the estimated inertia to avoid underdamped transient response. Numerical results are also shown.

   ADAPTIVE DISCRETE ESTIMATOR FOR INDUCTION MOTOR CONTROL
By B. De Fornel; J. C. R. De Oliveira [View]
[Download]

Abstract: This paper proposes an ESTIMATOR of electromechanical variables of an induction motor, through the use of an adaptive and easily implemented real-time algorithm. We simulate the algorithm, both in its continuous and in time discrete versions, studying its sensitivity to parameter variation. In consequence, we propose a cancellation of the DC components on its integrators, obtaining an Adaptive Estimator with good precision and dynamics, not only at start-up, but also at rotor speed inversion. To study the Estimator performance in the control of induction motors, we also simulate it in a scalar control scheme.

   PARAMETER ADAPTATION FOR A FIELD ORIENTED INDUCTION MACHINE FED BY A PWM-INVERTER AND DETERMINATION OF THE FUNDAMENTAL CURRENTS IN THE RANGE OF OVERMODULATION
By J. Fetz [View]
[Download]

Abstract: This paper presents a new method for on-line adaptation of the parameters of an induction machine fed by a voltage source inverter. A field oriented feed forward control with optional closed loop error control transfers amplitude and angular velocity of the reference voltage vector to a subordinated PWM-modulator, which is working in synchronized sinusoidal and optimized modulation mode. Algorithms for on-line parameter adaptation in steady state machine operation are derived for both field oriented control modes. For the range of optimized modulation a new method is presented to determine the fundamental current waveform by subtracting precalculated current harmonics from sampled phase currents. The entire control which includes PWM-modulator, parameter adaptation and fundamental current calculation is performed by a single 16 bit microcomputer. Simulation results and measurements on a 30 kW test drive document the obtained results.

EPE 1991 - 15 - Session 2.4: PREDICTIVE AND ADAPTIVE CONTROL
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 1991 - Conference > EPE 1991 - 15 - Session 2.4: PREDICTIVE AND ADAPTIVE CONTROL
	[return to parent folder]

	PREDICTIVE SPEED AND CURRENT CONTROL FOR D.C. DRIVES By D. Schröder; H. Wanner	[View] [Download]
Abstract: Current- and speed-control of a d.c.-motor fed by a line commutated converter is a well known strategy with conventional control. But with these conventional control techniques the characteristics of the system are used only to a very restricted extent. With a special predictive control algorithm this nonlinear controlled system can be utilized to its physical limits. The predictive control algorithm has been realized on a multi-microprocessor system with the actual precalculation performed by a signal processor. A modification of the prediction parameter equations yields a greatly improved stationary accuracy. For identification of the process parameters a special method has been adopted that uses comparison of precalculated and real armature currents. The behaviour of the whole current control system is discussed. Especially the consequences for a superimposed speed controller are derived and a suitable speed control method is shown. Experimental results for the above topics are presented.

	A PREDICTIVE CURRENT CONTROLLER FOR SYNCHRONOUS SERVO DRIVES By Hoang Le-Huy; Karim Slimani; Philippe Viarouge	[View] [Download]
Abstract: A real-time predictive current controller for PWM inverter feeding permanent-magnet synchronous motor is presented. Predictive current control scheme is applyied and the performance is studied and evaluated by computer simulation. The implementation of a predictive current controller using EPROM-based approach is described, in which calculations are executed by lookup tables. The experimental results obtained with a prototype 1-kW servo drive are presented and discussed.

	PREDICTIVE CASCADE CONTROL OF MACHINE TOOLS MOTOR DRIVES By P. Boucher; D. Dumur; S. Daumuller	[View] [Download]
Abstract: We propose in this paper a real time algorithm developed for machine tools drives. The implementation of this cascade structure based on predictive control laws is possible due to the simplicity of numerical algorithms. A first part will present a polynomial form of the Generalized Predictive Control method, and give a structure of the equivalent linear controller. This approach can provide a real time loop quicker than with a "classical" algorithmic solution. Moreover, this polynomial form allows to study, in frequency domains, the influence of criteria parrameters, on the stability and robustness of the equivalent controller. All this will be shown, reformulating the expression of the control law for a cascade structure. Then a second part will develop and compare the results obtained on two kinds of motors, used as machine-tools drives, with very small sampling period: a brushless motor and a field oriented control induction motor. The advantages of the polynomial predictive cascade control law will be stressed in terms of rapidity, errors cancellations, disturbances rejection and robustness.

	ADAPTIVE CONTROL OF PM SYNCHRONOUS MOTOR By L. Salvatore; S. Stasi; A. Dell'Aquila	[View] [Download]
Abstract: Permanent magnet synchronous motor drives are widely used in industrial applications because of their high performance. The use of field orientation control strategy gives external characteristics similar to those of well known dc drives without the need for slip rings and a field supply. In this paper we propose a new scheme for adaptive position control of a PM synchronous motor drive. This scheme employs an estimation method based on the Kalman filter theory. The algorithm can estimate the load torque and inertia moment quickly. The estimated torque is supplied to the PM synchronous motor to realize load compensation so to obtain no steady-state error and little transient-state position error. The speed feedback P-controller gain is adjusted by emploing the estimated inertia to avoid underdamped transient response. Numerical results are also shown.

	ADAPTIVE DISCRETE ESTIMATOR FOR INDUCTION MOTOR CONTROL By B. De Fornel; J. C. R. De Oliveira	[View] [Download]
Abstract: This paper proposes an ESTIMATOR of electromechanical variables of an induction motor, through the use of an adaptive and easily implemented real-time algorithm. We simulate the algorithm, both in its continuous and in time discrete versions, studying its sensitivity to parameter variation. In consequence, we propose a cancellation of the DC components on its integrators, obtaining an Adaptive Estimator with good precision and dynamics, not only at start-up, but also at rotor speed inversion. To study the Estimator performance in the control of induction motors, we also simulate it in a scalar control scheme.

	PARAMETER ADAPTATION FOR A FIELD ORIENTED INDUCTION MACHINE FED BY A PWM-INVERTER AND DETERMINATION OF THE FUNDAMENTAL CURRENTS IN THE RANGE OF OVERMODULATION By J. Fetz	[View] [Download]
Abstract: This paper presents a new method for on-line adaptation of the parameters of an induction machine fed by a voltage source inverter. A field oriented feed forward control with optional closed loop error control transfers amplitude and angular velocity of the reference voltage vector to a subordinated PWM-modulator, which is working in synchronized sinusoidal and optimized modulation mode. Algorithms for on-line parameter adaptation in steady state machine operation are derived for both field oriented control modes. For the range of optimized modulation a new method is presented to determine the fundamental current waveform by subtracting precalculated current harmonics from sampled phase currents. The entire control which includes PWM-modulator, parameter adaptation and fundamental current calculation is performed by a single 16 bit microcomputer. Simulation results and measurements on a 30 kW test drive document the obtained results.