Search

The paper describes a procedure using the first principle modelling and experimental identification of the Magnetic Levitation Model CE 152. It is a modified version of the paper [1]. The difference is that the identification and verification is done in open loop and constraints logic is added in the current paper. The author optimized and simplified dynamic model to a minimum to what is needed to characterize given system for the simulation and control design purposes. Only few open-loop experiments are needed to estimate the unknown parameters. Model quality is verified in open loop where the real and simulated data are compared. The model can serve as a simulation model for some standard control algorithms or as a process model for advanced control method design.

Model Order Reduction (MOR) challenges a high dimensional problem and plays a key role in areas where dynamic simulation studies are necessary for modern simulation strategy. Many conventional reduction methods namely, reduced order models based on Least Square Method (LSM), Balanced Truncation, Hankel Norm reduction, Dominant Pole Algorithm (DPA) and CDPA method have been developed in the field of control theory. Among these, recently proposed Clustering Dominant Pole Algorithm (CDPA) is able to compute the full set of dominant poles and their cluster center efficiently. In this paper, a hybrid algorithm for model order reduction known as Clustering Dominant Pole-Zero Algorithm (CDPZA) is proposed to identify and preserve the dominant zeros of the processes exhibiting non-minimum phase behaviour. The CDPZA method combines the features of clustering method and DPA. Further, the cluster centers of the dominant zeros in the numerator polynomial are determined using factor division algorithm. The Benchmark HiMAT system of 6th order is considered for testing and validation of the proposed algorithm. The simulation studies are carried out to show the efficacy of the proposed algorithm over conventional MOR algorithms.

Paper describes an implementation of a low cost but nontrivial laboratory system for automation and control theory teaching purposes. The system called RCDue consists of two parts – a dynamic system module and a control unit providing measurement, control and communication with MATLAB through USB serial port. Wiring of the dynamic system module can be modified to get a system of the desired behavior. Application example named as R5C4 with four RC circuits is described together with its mathematical model and calculated step responses. The system can be used in laboratory for practical education of C or MATLAB programming, modelling and identification or control theory subjects.

Precise object localization in an industrial environment is a significant task affecting follow-up processes for a pick and place application. One of the solutions to effectively ensure the success of this task is to use modern methods of machine vision. Machine vision is still a highly evolving topic, in which the use of approaches based on convolutional neural networks is rising. And so in this contribution, an innovative engineering approach based on convolutional neural networks is proposed for an object localization task. The approach is based on an atypical image segmentation, where the individual objects are represented by two colored gradient circles. These circles represent significant parts of the object like its center or ending. Each object type (class) is determined by a specific color. By use of a local maxima finder, all circles in an image are transformed to points. With knowledge of these points the coordinates and rotations are calculated. The proposed approach was tested on a legitimate localization problem with 100% precision, more than 99.52% recall on the positioning task and with an average of 6 minutes angle variance per object.

Grasp point detection is a necessary ability to handle for industrial robots. In recent years, various deep learning-based techniques for robotic grasping have been introduced. To follow this trend, we introduce a convolutional neural network-based approach for model-free one step method for grasp point detection. This method provides all feasible grasp points suitable for parallel grippers, based on a single RGB image of the scene. A case study, which shows the outstanding accuracy of the presented approach as well as its acceptable response time, is presented at the end of this contribution.

Robot guidance in an industrial environment is an important task to be solved in modern production facilities. A pick and place task is definitely one of the most common robot guidance issues to solve. In the beginning of the pick and place task, we need to perform a precise positioning of the objects of interest. In this contribution, an innovative engineering approach to multiple object positioning is proposed. The approach consists of two consecutive steps. At first, the original scene with objects of interest is transformed using a neural network. The output of this transformation is a schematic image, which represents the positions of the objects with gradient circles of various colors. Then, the positions of the gradient circles are determined by finding local maxima in the transformed image. The proposed approach is tested on a legitimate positioning problem with more than 99.8 % accuracy.

Together with smart homes, cities and factories, energy hubs and self-driving cars the smart agriculture or farming could be a way how to increase yields and efficiency as well as improve the welfare of farm animals, grow high quality crops and preserve the natural resources. Smart, precision and digital agriculture and farming current state survey with the technical challenges, interesting applications and future prospects is the aim of the paper. Worldwide and UE view is presented and compared with the situation in the Czech Republic. Authors are seeking for used or at least potential agriculture and farming applications of soft computing methods like fuzzy logic, machine learning and evolutionary computation.

Nowadays, the pressure on prevent invasive methods for diagnostics is still increasing in the health care sector. In the case of the lower urinary tract, early diagnosis can play a significant role to prevent a surgery. Here, the widely used non-invasive test, the uroflowmetry, is observed. As the new measurement devices are being created, new algorithms for basic urinary flow classification must be developed. There, the feature extraction methods are developed and introduced for further use in combination with standard classifiers based on machine learning. In the further work, the methods will be reviewed on extensive dataset, which is currently being created. As the credible dataset verified by several urologist will be obtained, the proposed methods should be examined. Direction of further development will depend on the results of introduced methods.

The usage of the renewable energy sources and electric vehicles is rising which might result in a problem with the energy grid stabilization. Some works state that buildings contribute to the overall energy consumption by 20–40% which makes them a great subject to study. Finding solutions for better energy management can have a significant impact on the energy sector. This work focuses on proposing a simplified model of a home considering main appliances, heating and cooling, a photovoltaic system, an electric vehicle and weather and solar irradiance forecasts. The model is created in MATLAB as a simulation script with defined parameters. It uses a 5 day weather forecast to predict energy demands of a home as well as expected energy generation by the photovoltaic panels. This model can be used to better understanding the energy demand of either a single building or a community of buildings put together based on the information about the outside environment. Another possibility is to use the model for predictive optimization in order to find better ways of energy usage in a building.

The repeating of heating/cooling cycle (75 times) simulating the heat energy storage in magnesium nitrate hexahydrate confirmed supercooling of 27 K, and thus many non-isostructural nucleating agents were tested in concentration of 1 mass% in four charge/discharge cycles. From the whole group, four promising nucleating salts were subjected to detailed study with addition of 0.5, 1, and 2 mass% of nucleating salt. The results show that the addition of Mg(OH)2, BaO, MgO and Sr(OH)2 suppressed supercooling below 5 K during 50 melting/crystallization cycles. At the similar conditions (temperature range and repeating of heating and cooling), MNH and MNH with 0.5 mass% Mg(OH)2 or Sr(OH)2 underwent corrosion tests in aluminium alloy tubes to discover applicability of this material for the containers or capsules for heat storage. The long-term contact (>49 days) of aluminium alloy with melted heat storage material for 664 h provided information about suitability of alloy for storing of magnesium nitrate hexahydrate with nucleating agent.

Software for thermogravimeter constructed at the Department of Inorganic Technology, University of Pardubice is presented in the paper. Used hardware and design of the experiment is described briefly. Main part of the paper focusses to the software solution in MATLAB environment. Except control, plotting and data exporting tasks it was necessary to communicate with Eurotherm nanodac Recorder / Controller through the TCP/IP Modbus protocol and with Sartorius weigh cell through the serial USB ASCII-based protocol. Simple example of Constant Rate Thermal Analysis experiment is given at the end of the paper.

Pest birds are considered as a special kind of vermin, since, in most of countries, their legal position does not enable their direct extermination. Therefore, in order to protect the agricultural areas indirectly from pest birds, the robust and highly selective pest bird sensor is necessary to design. In this contribution, the pest bird detection unit, based on a convolutional neural network, is presented. The convolutional neural network itself is used for the decision making about the pest bird occurrence, while sound recordings are used as input data. The testings, presented at the end of the contribution, proved a very high accuracy of the detection unit, with the results indispensably improved in comparison to previously presented approaches.

This paper deals with a nonlinear Model Predictive Control with a special form of the process model. Controller uses for the prediction purposes a locally valid linear sub-models. The sub-models are obtained from a neural model with the rectifier activation function in hidden neurons. Simulation example is given to demonstrate proposed solution - neural model design and predictive controller application.

Predictive Controller of a laboratory thermal process is presented in the paper. Process model is approximated by a neural network. On-line optimization is done by a genetic algorithm. Control algorithm is tested on the laboratory thermal process and compared to the standard control methods like predictive controller with the transfer and state-space linear model and the quadratic programming optimization method or a PI controller.

The paper deals with the decentralized control problem for linear multivariable systems with the number of manipulated variables equal or greater than the number of controlled variables. Proposed static compensator ensures automatic creation of input/output pairs for individual control loops. The compensator provides steady state autonomy and unit gain for the control loops. Steady state gain matrix of the controlled system and vector of offsets are sufficient information for the compensator design. Estimation of the gain matrix and offsets from the measured data is also proposed in the paper.

The paper deals with an online optimization control method for dynamical processes called Model Predictive Control (MPC). It is a popular control method in industry and frequently treated in academic areas as well. The standard predictive controllers usually do not guarantee stability especially for the case of short horizons and large control error penalization. Terminal state is one way to ensure stability or at least increase the controller robustness. In the paper, deviation of the predicted terminal state from the desired terminal state is considered as one term of the cost function. Effect of the stability and control quality is demonstrated in the simulated experiments. The application area for online optimization methods is very broad including various logistics and transport problems. If the dynamics of the controlled processes cannot be neglected, the optimization problem must be solved not only for steady state but also for transient behaviour, e.g., by MPC. Článek se zabývá online optimalizační metodou řízení označovanou jako prediktivní řízení. Standardní prediktivní regulátory negarantují stabilitu. Koncový stav je jednou z možností, jak stabilitu zajistit. V příspěvku je v kritériu uvažována odchylka predikovaného koncového stavu od žádaného koncového stavu. Efekt tohoto řešení na stabilitu a kvalitu regulace je ukázán na simulovaných experimentech. Aplikační oblast pro navrhované řešení je velmi široká včetně logistických a přepravních problémů.

How to explain Model Predictive Control (MPC) to students? How to practise it? The paper deals with chain of actions involving teaching, practicing and laboratory application of MPC at University of Pardubice in Czech Republic and at Anna University in India. Individual steps are presented and discussed with examples from educational experience - e.g. modelling and identification, derivation of MPC controller, simulations and laboratory applications. Every phase has a key and weak point as well. Desired results is that students understand better the theoretical concepts and they are able to apply predictive controllers at least for laboratory processes. Derivations and MATLAB scripts are available online.

In this paper, an efficient model-predictive control strategy that can be applied to complex multivariable process is presented. A reduced order generalized predictive algorithm is proposed for online applications with reduction in complexity and time elapsed. The complex multivariable process considered in this work is a binary distillation column. The reduced order model is developed with a recently proposed hybrid algorithm known as Clustering Dominant Pole Algorithm and is able to compute the full set of dominant poles and their cluster centre efficiently. The controller calculates the optimal control action based on the future reference signals, current state and constraints on manipulated and controlled variables for a high-order dynamic simulated model of nonlinear multivariable binary distillation column process. The predictive control algorithm uses controlled auto-regressive integrated moving average model. The performance of constraint generalized predictive control scheme is found to be superior to that of the conventional PID controller in terms of overshoot, settling time and performance indices, mainly ISE, IAE and MSE. V příspěvku je prezentována účinná prediktivní strategie řízení, kterou lze aplikovat na komplexní vícerozměrné procesy.

The heat capacity and enthalpy of fusion of calcium nitrate tetrahydrate and magnesium nitrate hexahydrate were determined from 234.15 K to melting temperature by DSC. The modified stepwise method was used for heat capacity measurement, and enthalpy of fusion was determined from continuous heating by the rate of 10 K min−1 (Pilař et al. in J Therm Anal Calorim 118:485–491, 2014). Determined values were used for the calculation of entropy and Gibbs energy in the experimental temperature range. Melting point and enthalpy of fusion of calcium nitrate tetrahydrate are 317.1 ± 0.3 K and 36.6 ± 0.2 kJ mol−1, and for the magnesium nitrate hexahydrate, the values are 362.9 ± 0.4 K and 40.8 ± 0.5 kJ mol−1. One solid–solid phase transformation was observed for the magnesium salt at 345.7 ± 0.9 K with enthalpy of transition 3.1 ± 0.2 kJ mol−1. The available accumulated energy composed of sensible (heating) and latent heat (phase transformations) is 43.4 and 63.8 kJ mol−1 for the hydrated calcium and magnesium nitrate, respectively. The kinetics of solid–solid phase transformation for magnesium salt was studied under non-isothermal conditions by DSC, and the process was described using autocatalytical model with parameters in the range of 0.50–0.85 for m and range of 2.58–1.48 for n, respectively.

The isobaric heat capacities of solid explosive materials were measured using differential scanning calorimeter. The values of C p were determined at temperature 298.15 K and compared to theoretical data calculated using the group additivity approach. Several group contribution methods were tested, showing that certain method for some explosives provides estimation of C p with error only 1 %. Generally, the average error limit for calculated C p was about 11 %, which is insufficient.

The paper deals with predictive control of non-holonomic mobile robot. The basic nonlinear kinematic equation is linearized into two different linear time varying models based on frame of reference — world coordinates and local coordinate of mobile robot. The non-linear model predictive control is applied to the trajectory tracking problem of a non-holonomic mobile robot with these models. The control law is derived from a cost function which penalizes the state tracking error, control effort and terminal state deviation error. Various simulation experiments are conducted and a comparative analysis has been made with respect to state-of-the-art approaches.

This paper presents modelling of ball and plate systems based on first principles by considering balance of forces and torques. A non-linear model is derived considering the dynamics of motors, gears, ball and plate. The non-linear model is linearized near the operating region to obtain a standard state space model. This linear model is used for discrete optimal control of the ball and plate system — the trajectory of the ball is controlled by control voltages to the motor.