The matrix converter is a new generation of power electronic converters and is an alternative to back-to-back converters in applications that dimensions and weight are important. In this paper, a simple control algorithm for a three-phase asynchronous motor based on a direct torque control technique, which is fed through a three-phase direct matrix converter, is presented. For direct matrix converters, 27 switching modes are possible, which using the predictive control technique and for the different modes of the matrix converter, the motor behavior is estimated at the next sampling interval. Then the objective function is determined and the optimal possible mode is selected. Finally, the best switching mode is applied to the direct matrix converter. In order to evaluate the proposed method, simulation of the system in Matlab/Simulink software environment is performed. The results show the effectiveness of the proposed method.
One of the main problems of multivariable cost functions in model predictive control is the choice of weighting factors. Two finite control set model predictive control algorithms, applied to the three-phase active rectifier with an LCL filter, are described in the paper. The investigated algorithms, i.e. PCicuc and PCigicuc, implement multivariable approaches applying line (grid) current, capacitor voltage and converter current. The main problem dealt with in the paper is the choice of optimum values of the cost function weighting factors. The values of the factors calculated using the method proposed in the paper are very close to the values represented by the lowest THDi of the line current. Moreover, simulations verifying the equations used in the prediction of controlled values, i.e. line current, capacitor voltage and converter current, are presented. Both simulation and experimental results are presented to verify effectiveness of the investigated control strategies under change of the load (P = 5 kW and 2.5 kW), during transient states, under unbalanced and balanced line voltage.
Electromagnetic mill installation for dry grinding represents a complex dynamical system that requires specially designed control system. The paper presents model-based predictive control which locates closed loop poles in arbitrary places. The controller performs as gain scheduling prototype where nonlinear model – artificial recurrent neural network, is parameterized with additional measurements and serves as a basis for local linear approximation. Application of such a concept to control electromagnetic mill load allows for stable performance of the installation and assures fulfilment of the product quality as well as the optimization of the energy consumption.
This paper investigates the application of a novel Model Predictive Control structure for the drive system with an induction motor. The proposed controller has a cascade-free structure that consists of a vector of electromagnetics (torque, flux) and mechanical (speed) states of the system. The long-horizon version of the MPC is investigated in the paper. In order to reduce the computational complexity of the algorithm, an explicit version is applied. The influence of different factors (length of the control and predictive horizon, values of weights) on the performance of the drive system is investigated. The effectiveness of the proposed approach is validated by some experimental tests.
This paper proposes a practical tuning of closed loops with model based predictive control. The data assumed to be known from the process is the result of the bump test commonly applied in industry and known in engineering as step response data. A simplified context is assumed such that no prior know-how is required from the plant operator. The relevance of this assumption is very realistic in the context of first time users, both for industrial operators and as educational competence of first hand student training. A first order plus dead time is approximated and the controller parameters immediately follow by heuristic rules. Analysis has been performed in simulation on representative dynamics with guidelines for the various types of processes. Three single-input-single-output experimental setups have been used with no expert users available in different locations – both educational and industrial – these setups are representative for practical cases: a variable time delay dominant system, a non-minimum phase system and an open loop unstable system. Furthermore, in a multivariable control context, a train of separation columns has been tested for control in simulation, followed by experimental tests on a laboratory system with similar dynamics, i.e. a sextuple coupled water tank system. The results indicate the proposed methodology is suitable for hands-on tuning of predictive control loops with some limitations on performance and multivariable process control.
Model predictive control (MPC) algorithms brought increase of the control system performance in many applications thanks to relatively easily solving issues that are hard to solve without these algorithms. The paper is focused on investigating how to further improve the control system performance using a trajectory of parameters weighting predicted control errors in the performance function of the optimization problem. Different shapes of trajectories are proposed and their influence on control systems is tested. Additionally, experiments checking the influence of disturbances and of modeling uncertainty on control system performance are conducted. The case studies were done in control systems of three control plants: a linear non- minimumphase plant, a nonlinear polymerization reactor and a nonlinear thin film evaporator. Three types of MPC algorithms were used during research: linear DMC, nonlinear DMC with successive linearization (NDMC–SL), nonlinear DMC with nonlinear prediction and linearization (NDMC–NPL). Results of conducted experiments are presented in greater detail for the control system of the polymerization reactor, whereas for the other two control systems only the most interesting results are presented, for the sake of brevity. The experiments in the control system of the linear plant were done as preliminary experiments with the modified optimization problem. In the case of control system of the thin film evaporator the researched mechanisms were used in the control system of a MIMO plant showing possibilities of improving the control system performance.
The model predictive control (MPC) technique has been widely applied in a large number of industrial plants. Optimal input design should guarantee acceptable model parameter estimates while still providing for low experimental effort. The goal of this work is to investigate an application-oriented identification experiment that satisfies the performance objectives of the implementation of the model. A- and D-optimal input signal design methods for a non-linear liquid two-tank model are presented in this paper. The excitation signal is obtained using a finite impulse response filter (FIR) with respect to the accepted application degradation and the input power constraint. The MPC controller is then used to control the liquid levels of the double tank system subject to the reference trajectory. The MPC scheme is built based on the linearized and discretized model of the system to predict the system’s succeeding outputs with reference to the future input signal. The novelty of this model-based method consists in including the experiment cost in input design through the objective function. The proposed framework is illustrated by means of numerical examples, and simulation results are discussed.
This paper presents simulation and laboratory test results of an implementation of an infinite control set model predictive control into a three-phase AC/DC converter. The connection between the converter and electric grid is made through an LCL filter, which is characterized by a better reduction of grid current distortions and smaller (cheaper) components in comparison to an L-type filter. On the other hand, this type of filter can cause strong resonance at specific current harmonics, which is efficiently suppressed by the control strategy focusing on the strict control input filter capacitors voltage vector. The presented method links the benefits of using linear control methods based on a space vector modulator and the nonlinear ones, which result in excellent control performance in a steady state as well as in a transient state.
This paper presents an analysis and simulation studies of three-phase matrix converter with GaN HEMT bidirectional switches with predictive control of grid currents and converter output currents. Two methods of grid currents shaping are described and compared. The first method is based on calculations of instantaneous grid reactive power and the second one uses the active power of the load. The analyzed converter works with the resistive-inductive load, and from the grid side the LC filter with damping resistor has been used.