Karol Kyslan

Dynamic Emulation of Mechanical Loads with Backlash Based on Rapid Control Prototyping

The paper discusses advanced control of a dynamometer performing dynamic emulation of mechanical loads. Assumed is certain class of nonlinear load with backlash. Paper presents and describes the basic control structure, which can be used for validation of speed algorithms and its practical implementation with rapid control prototyping. Validation of experimental results against the simulations has been performed with experimental test bench containing permanent magnet DC machines and DC-DC converters with high sampling times in current control loops.

Finite Control Set Model Predictive Speed Control of a DC Motor

The paper describes the design procedure for a finite control set model predictive control (FCS-MPC) of brushed permanent magnet DC (PMDC) machine supplied from DC-DC converter. Full order linear Kalman filter is used for estimation of an unmeasured load torque and reduction of speed measurement noise. A new cost function has been introduced with a feedforward dynamic current component and a feedforward static load current component. The performance of the proposed control strategy is compared to the conventional PI-PWM cascade speed control through the experimental verification on the 250 W laboratory prototype. Obtained results show excellent dynamic behaviour and indicate possible energy savings of the proposed speed control.

Dynamic emulation of mechanical loads — Analysis and implementation into industrial drive

The paper presents an approach to the dynamic emulation of mechanical loads. The main goal was to implement known control algorithm to the industrial converter SINAMICS S120 with the control unit CU320 using Drive Control Chart programming tool. The reference torque of a load machine is calculated through the closed-loop control algorithm using PI estimator. Proposed implementation enables the user the emulation of various mechanical loads with the industrial drive dynamometer, although only variation of the inertia and the viscous friction is reported here.

Implementation of Robot Control Algorithms by Real-Time Control System

The paper deals with development and implementation of the direct and inverse kinematics to control of 6 DOF industrial robot SEF-ROBOTER SR25 by a real time control system. To obtain the angular position of each joint an iterative algorithm is applied that is developed in the Simulink program. This solution creates a basis for real time control of the robot drives utilizing features of SIEMENS SINAMICS family of frequency converters. The developed control system presents a universal platform enabling to debug any robot control algorithm and also easy to change a desired trajectory of the end effector. The equipment is suitable for testing different trajectories of the robot and is suitable also for educational purposes.

Tension Controllers for a Strip Tension Levelling Line

Strip material processed in continuous production lines causes a mechanical coupling among tension rolls driven by a multi-motor drive system. Thus, the drives are mutually mechanically coupled and influenced in their operation. One of the key techniques, to guarantee the output product quality in the fibre, paper, plastic and metal plating industries, consists in controlling the strip tension on a preset value that should be set differently for each section of the line. This paper describes two newly developed types of tension controllers: a tension controller with the ramp and a stepper tension controller, which are suitable for the line sections with high and low level of the tension, respectively. The proposed controllers were verified experimentally on a real tension levelling line.

Design and analysis of torque control for load drive with dynamic emulation

This paper analyses a torque control structure for load machine used as a dynamometer. Typical area of the use is a mimicking of a mechanical load behaviour in case the mechanical load is not available. Instead of that, the mechanical load is emulated by the load machine with specific torque control. This approach belongs to dynamic emulation of mechanical load methods. Reference torque calculation with particular components for compensation of mechanical dynamics and disturbance torque in the test bench is presented.

Design of load torque and mechanical speed estimator of PMSM with unscented Kalman filter — An engineering guide

This paper presents the design of Unscented Kalman Filter (UKF) for estimation of state space variables of permanent magnet synchronous machine (PMSM). The UKF is shown together with the field oriented speed control. At first, the position and the speed of PMSM are measured, and UKF is used only for a load torque estimation. It is indicated how differences in sampling time of the speed and the current loop affects overall estimation performance. Subsequently, speed sensorless performance of the UKF with the same parameters is shown for comparison. Designed filter is verified only by Matlab simulation.

A HARDWARE-IN-THE-LOOP SIMULATOR BASED ON REAL SKODA SUPERB VEHICLE AND RT-LAB/CARSIM

This paper describes the design and realization of a hardware-in-the-loop simulator made from a real Skoda Superb vehicle. A combination of RT-LAB and CarSim software is used for real-time control and for handling the sensoric subsystems. The simulator provides almost realistic testing of driving cycles with on-line visualization. This unique device can be used in various fields of research.

DC motor identification based on Recurrent Neural Networks

The paper describes system identification by using Artificial Neural Networks that is applied to a permanent magnet DC motor. To identify its dynamic behavior an experimental setup has been developed that enables to measure data of the system input (armature voltage) and output (current and rotor speed). Generally, the identification methods can be classified as parametric and non-parametric. We use a non-parametric method (black box). A recurrent neural network was used and the Nonlinear AutoRegressive network with eXogenous inputs network (NARX) has been selected. Parallel architectures have been used in training the NARX network. The scaled conjugate gradient training algorithm, using the first and second derivatives of error to train the network to minimize the error function, has been selected. The network architecture which has been used to create the dynamic model of the motor consists of three hidden layers, a single input neuron, and two output neurons. The modeled and measured normalized data were compared with good conformity.