Tomasz A. Rutkowski

Simplified, multiregional fuzzy model of a nuclear power plant steam turbine

Power systems, including steam turbines and synchronous generators, are complex nonlinear systems with parameters varying over time. The paper presents the developed simplified, multiregional fuzzy model of the steam turbine of a nuclear power plant turbine generator set and compares the results with a full nonlinear model and commonly used linear input-output model of a steam turbine. The proposed model consist of series of linear input-output models defined for specific steam turbine operating points and one fuzzy switching module with Takagi-Sugeno reasoning interface. Provided simulation results indicate that the response of presented simplified fuzzy multiregional steam turbine model is very close to the reference nonlinear steam turbine model such that the maximum absolute error is around few percent for considered operating point changes, 30-100% of nominal mechanical power. Hence steam turbine model in such form may increase the accuracy of developed control algorithms, which use system model for adaptation or prediction purposes.

Use of Neural Networks in Diagnostics of Rolling-Element Bearing of the Induction Motor

Bearing defect is statistically the most frequent cause of an induction motor fault. The research described in the paper utilized the phenomenon of the current change in the induction motor with bearing defect. Methods based on the analysis of the supplying current are particularly useful when it is impossible to install diagnostic devices directly on the motor. The presented method of rolling-element bearing diagnostics used indirect transformation, namely Clark transformation. It determines the vector of the spatial stator current based on instantaneous current measurements of the induction motor supply phases current. The analysis of the processed measurement data used multilayered, one-directional neural networks, which are particularly attractive due to their nonlinear structure and ability to learn. During the research 40 bearings: undamaged, with damages of three types and various degrees of fault extent, were used. The conducted research proves the efficiency of neural networks for detection and recognition of faults in induction motor bearings. In case of tests of the unknown state bearings, an efficiency approach to failure detection equaled 77%.

Comparison of tuning procedures based on evolutionary algorithm for multi-region fuzzy-logi PID controller for non-linear plant

The paper presents a comparison of tuning procedures for a multi-region fuzzy-logic controller used for nonlinear process control. This controller is composed of local PID controllers and fuzzy-logic mechanism that aggregates local control signals. Three off-line tuning procedures are presented. The first one focuses on separate tuning of local PID controllers gains in the case when the parameters of membership functions of fuzzy-logic mechanism are know a priori. The second one consists of two major steps. In the first step, the local PID controllers gains are calibrated and in the next step the parameters of membership functions of fuzzy-logic mechanism are tuned. The third one focuses on tuning of PID controllers gains and membership functions parameters jointly. These procedures exploit evolutionary algorithm to optimize the performance of the multi-region fuzzy-logic controller with respect to integral quality criterion such as Integral Absolute Error (IAE). Effectiveness of these procedures is verified based on a well known from literature continuous nonlinear pH neutralization reactor model. For comparison a single PID controller tuned for specific work point is also taken into account.

Multi-nodal PWR reactor model — Methodology proposition for power distribution coefficients calculation

In the paper the multi-nodal Pressurized Water Reactor (PWR) heat transfer model based on Manns model is presented. This model is used for heat transfer from fuel to coolant modelling purposes in a reactor core. The authors expand the approach widely used in literature by defining additional coefficients for heat transfer model. These parameters approximate the power generation distribution in the PWR reactor core according to the reactor control rod bank movement. Authors describe in details the proposed methodology for calculation of introduced power distribution coefficients. In the simplest case, those coefficients may be equal and constant over time and space of a reactor. On the other hand, they can be treated as variables that are time and reactor region dependent. By introducing these specific coefficients, the nodal model of heat transfer gains advanced capabilities that can be efficiently used in design and synthesis of more advanced and complex power control system in nuclear reactor with respect to temperature distribution within the reactor core.

The excitation controller with gain scheduling mechanism for synchronous generator control

The power systems, including the synchronous generators and power systems networks, are complex nonlinear systems with configuration and parameters which change through time. That leads to electromechanical oscillations occurring in that system. Thus synchronous generator excitation controller must be capable of providing appropriate stabilization signal over broad range of operating conditions and disturbances. In this paper, an excitation controller (EC) with gain scheduling mechanism is proposed to deal with those phenomena. This controller is designed and implemented based on the gain scheduling approach according to the power system stabilizer (PSS) and the conventional PID controller in the case of automatic voltage regulator (AVR). The proposed PSS accommodate the nonlinear nature of power systems and adopt with parametric gain surfaces its parameters to the actual operating conditions. It is parameterized with two stages tuned procedure based on the nonlinear optimization problem solving according to the linearized power system model. The obtained result shows that the proposed excitation controller with gain scheduling mechanism can provide effective electromechanical oscillation damping and lower control signal effort in comparison with the a classic excitation system through wide range of operating conditions.

Supervised model predictive control of wastewater treatment plant

An optimizing control of a wastewater treatment plant (WWTP), allowing for cost savings over long time period and fulfilling effluent discharge limits at the same time, requires application of advanced control techniques. Model Predictive Control (MPC) is a very suitable control technology for a synthesis of such a truly multivariable controller that can handle constraints and accommodate model-based knowledge combined with hard measurements. As it is impossible to efficiently control the plant by one universal control strategy under all possible influent conditions, it is proposed in the paper to on-line adapt the nonlinear MPC control strategy in order to best adapt the control actions to actual and predicted WWTP conditions. Adjusting the MPC control strategy is carried out by suitable manipulating the components of performance index and constraints. This process is supervised by Mamdani reasoning system. The supervised MPC controller performance was tested by simulations within large range of plant operating conditions and then compared with classic MPC without such mechanism. The simulation model of the benchmark WWTP utilizes ASM2d model.

The distributed model predictive controller for the nuclear power plant turbo-generator set

Typically there are two main control loops with PI controllers operating at each turbo-generator set. In this paper a distributed model predictive controller DMPC, with local QDMC controllers for the turbine generator, is proposed instead of a typical PI controllers. The local QDMC controllers utilize step-response models for the controlled system components. These models parameters are determined based on the proposed black-box models of the turbine and synchronous generator, which parameters are identified on-line with the RLS algorithm. It has been found that the proposed DMPC controller realize the reference trajectories of the effective power and the angular velocity, and damp generator voltage oscillations with satisfactory quality in comparison to the typical control structure with the PID controllers.

Implementation of hierarchical control of drinking water supply system: Didactic project — Computer controled system

The paper presents the outline of the didactical project of a complex computer controlled system realized by the first degree students of the Automatics and Robotics on the Faculty of Electrical and Control Engineering (FoEaCE) in Gdansk University of Technology (GUT). The synthesis, implementation and analysis of a multilayer hierarchical control system for drinking water supply system (DWSS) are main topics of that project. The scope of the project covers many aspects of automation: mathematical modeling, dynamic object identification, real-time simulation, acquisition and scaling of signals, estimation, PID control, definition of optimization criteria, predictive control technology, control synthesis, SCADA and HMI application development, OPC communication, hardware-in-the-loop technology, and many others tools and methods needed to obtain a functional control system structure. The DWSS system model is simulated with quasi-real time using the Simulink RealTime Desktop toolbox of Matlab/Simulink environment. The proposed control system structure consists of three layers — direct control layer with PLC/PAC controller, supervisory control layer with SCADA/HMI system and optimization control layer with Optimization toolbox of Matlab/Simulink environment for optimizing control purposes with MPC controller. The undoubted advantage of presented project is the integration into one coherent, functional system many previously separately recognized during the studies methods and tools.

The QDMC Model Predictive Controller for the Nuclear Power Plant Steam Turbine Control

There are typically two main control loops with PI controllers operating at each turbo-generator set. In this paper a model predictive controller QDMC for the steam turbine is proposed - instead of a typical PI controller. The QDMC controller utilize a step-response model for the controlled system. This model parameters are determined, based on the simplified and linear model of turbo-generator set, which parameters are identified on-line with RLS algorithm. It has been found that the proposed QDMC controller realize the reference trajectories of the effective power and the angular velocity, and damp the electromechanical oscillations with satisfactory quality in comparison to the typical PI and DMC controllers.