S. Moreau

Robust Speed Control of PMSM using Generalized Predictive and Direct Torque Control Techniques

This paper introduces a generalized predictive controller (GPC) for the speed control of a PMSM. The performance of the proposed controller is evaluated in simulation and compared with a IP controller. The stator flux linkage and electromagnetic torque are controlled by (DTC) strategy. For the speed regulation, the generalized predictive control strategy is used. Simulation results indicate that the proposed (GPDTC) is reliable and effective for the speed control of the PMSM over a wide range of operations of the PMSM drive.

Parameters Estimation of the Actuator used in Haptic Interfaces: Comparison of two Identification Methods

In advanced motor control systems, an accurate knowledge of motor parameters is essential in order to achieve good performances. Some of these parameters, such as stator resistances, are sometimes given by constructors, but they vary according to the operating conditions. This paper presents and compares two parameter identification methods of an actuator used in haptic interfaces which, in this case, is a permanent magnet synchronous machine (PMSM). Using these methods, the electrical parameters of the PMSM can be determined during different operating conditions. Physical parameters estimation of a dynamic 4-parameter model is performed on the one hand, using an output error identification method based on a non linear programming algorithm and on the other hand, using an identification method based on least-squares techniques and inverse models

Sensorless Force/Position Control of a Single-Acting Actuator Applied to Compliant Object Interaction

This paper deals with a sensorless control approach of a lightweight electric actuator. Force and motion control are necessary to perform manipulation tasks with object interaction of different stiffnesses. The controller must ensure stability in a wide range of rigidity, particularly in soft contact cases. The approach proposed here is based on a force/position control scheme. First, a nonlinear position controller is performed through a linear parameter-varying state feedback. A state observer is added to estimate unmeasurable variables coming from an absence of terminal sensors. Subsequently, an output feedback force control is determined using an H∞ framework. The controller is experimentally applied on a single-acting actuator to demonstrate the efficiency of this approach dedicated to compliant object interaction.

Speed estimation improvement using full order state observer for a haptic interface

A haptic system is an articulated mechanical structure with motors, sensors as well embedded electronics allowing force feedback. It enables the user to interact with virtual reality through the sense of touch and sight. Speed control of such systems usually becomes unstable at low speed range due to the imposed speed by the operators hand. This paper describes implementation of full order state speed observer for a single degree of freedom haptic interface driven by permanent magnet synchronous motor. This method allows accurate speed estimation. The observer design criteria for haptic interface are analyzed in this paper. In addition, its influence on system stiffness is illustrated.

Amplification of Single Mechanical Fault Signatures Using Full Adaptive PMSM Observer

This paper presents a localized mechanical fault detection method using a full adaptive permanent-magnet synchronous machine (PMSM) observer in a d/q reference frame. It appears that the fault sensitivity of PMSM stator currents and rotor speed measurements is very low with regard to single mechanical faults. That is why an amplification of single mechanical fault signatures in frequency domain is proposed by adjusting the adaptive PMSM observer gains. To study the influence of the adaptive PMSM observer gains on fault signatures, the observer model is first of all linearized at an operating point. Then, the static gain, the resonance, and the quality factor Q of different transfer functions are analyzed. A dedicated experimental setup based on an original mechanical fault emulator and a 7.8 kW PMSM drive is designed to validate the simulation results. The simulation and the experimental results show that the proposed method is effective to amplify single mechanical fault signatures and the calculated fault indicator.

LPV Modeling and Experimental Identification of a New Self-Sensing Finger Joint

Abstract This paper deals with the development of a new finger articulation prototype used as a joint for anthropomorphic robotic hand. We focus on modeling strategies applied to the finger prototype and its experimental identification. Self-sensing characteristic is achieved by an optimal design of highly back-driveable and efficient electromechanical actuators based on a DC motor, which requires an accurate modeling due to the absence of external sensors. The non-linear model of a finger joint can be reformulated as a Linear Parameter-Varying (LPV) model. On the contrary of classic method used for modeling, a quasi-LPV model is derived from Linear Time Invariant (LTI) interpolation, using a simulated annealing algorithm. Then, our proposed quasi-LPV model is validated from the Inverse Dynamic Model (IDM), and shows a good efficiency in predicting the non-linear behavior of the end effector.

Speed estimation comparison between full order state observer & Kalman filter for a haptic interface

A haptic system is an articulated mechanical structure with motors, sensors as well embedded electronics allowing force feedback. It enables the user to interact with virtual reality through the sense of touch and sight. Speed control of such systems usually becomes unstable at low speed range due to the imposed speed by the operators hand. This paper describes the implementation of a full order state speed observer as well as a Kalman filter for a single degree of freedom haptic interface driven by a permanent magnet synchronous motor. Those two methods allow accurate speed estimation. The observer and the filter design procedure for haptic interface are analyzed in this paper. In addition, their influence on system stiffness is illustrated, taking into consideration variations of the system inertia.

Bearing Fault diagnosis using a pre-filtering and a spectral identification algorithm

The Time Synchronous Averaging (TSA) is a well known method used for early detection of defects in rolling elements. This paper applies and evaluates the effectiveness of this method, spefifies its limitations, and improves it using a pre-filtering of the envelop signal associated with an identification method based on Marquardt algorithm. The proposed procedure is evaluated and applied theoretically and practically on simulated and real signals.

Influence of speed estimation methods and encoder resolutions on the stiffness of a haptic interface

A haptic system is an articulated mechanical structure with motors, sensors, embedded electronics as well computer software allowing force feedback. It enables the user to interact with virtual reality through the sense of touch and sight. The control of such systems can become unstable at low speed because of mediocre speed estimation quality. This paper describes the influence of diminution of the position encoder resolution on the estimated speed, since the speed is calculated using the position information. Further more, this paper illustrates the implementation of various speed estimation techniques, such as a full order state speed observer, Kalman filter and the synchronous constant elapsed time method (direct digital position derivation) for a single degree of freedom haptic interface driven by a permanent magnet synchronous motor. In addition, their influence on system stiffness is illustrated.

Speed Estimation Improvement After Decreasing the Encoder Resolution for a Haptic Interface

A haptic system consists in an articulated mechanical structure with motors and position sensors, as well as embedded electronics allowing force feedback. It is driven by a haptic interface, which enables the user to interact with an image in a virtual reality application through the sense of touch. This paper describes a single degree of freedom haptic interface and particularly presents a low speed estimation method with a reduced resolution encoder, in order to optimize the application cost. It leads to an important relative error of the speed estimation and the deterioration of the obtained stiffness. This paper presents a simple speed estimation method which improves the relative error of the measure and increases the virtual stiffness to almost the value obtained with a high resolution encoder. It keeps the synchronism between the speed and the other variables measurements. Moreover, the measurement accuracy at low speeds using the proposed method is improved, compared to currently known methods