Snejana Yordanova

Robust stability of single input fuzzy system for control of industrial plants with time delay

Most industrial plants are complex, nonlinear, time-varying with time delay, difficult to control. The single input fuzzy controllers (SI FCs) are promising in satisfying the high performance demands to the control systems of such plants and ensuring the fuzzy system stability, coping with system nonlinearity, plant model uncertainty and SI FC design. The aim of the present investigation is to derive robust stability requirements for an incremental PI-like SI FC closed loop system that serve the development of a SI FC design method for the control of industrial plants with time delay and model uncertainty. The main contributions are: 1) modification of the Popov stability approach to study the SI FC system stability and to derive design rules; 2) derivation of robust stability criterion for a SI FC system and on its basis development of a SI FC tuning algorithm. The results are applied for the stabilisation of the air temperature in a furnace using MATLAB™. The advantages of the designed SI FC system are assessed in comparison with ordinary PI control system via simulations. The SI FC system preserves stability for a great range of plant uncertainties. The tradeoff is longer settling time of the nominal system transient responses.

A Two-Variable Fuzzy Control Design With Application to an Air-Conditioning System

Air-conditioning systems provide optimal conditions for work, human health, and quality of living. At the same time, they are one of the greatest household energy consumers. Therefore, the Building Energy Management System sets high demands on their control, aiming to reduce energy consumption while ensuring indoor comfort. The classic control technique cannot satisfactorily deal with the energy-efficient stabilization of basic interacting microclimate parameters. The fuzzy logic approach offers intelligent means for the stabilization of the coupled temperature, humidity, and fresh air supply with low energy requirements. The aim of this study is to develop a model-free design method for a two-variable PI fuzzy controller for temperature and humidity control that ensures indoor comfort and reduces energy consumption by supervisory fuzzy tuning.

Design and implementation of robust multivariable PI-like fuzzy logic controller for aerodynamic plant

Most modern plants are multivariable, non-linear and time-variant. Therefore, they are difficult to model and control especially when measurements are insufficient or imprecise due to variables interactions, disturbances and noise impact. A perspective model-free approach is based on fuzzy logic, which has proved efficient in ensuring both robust and economic control, employing linguistic expert knowledge in production of simple and feasible non-linear controllers that outperform their linear counterparts. The aim of this investigation and the main contributions are related to the development of a method for the design of multivariable robust fuzzy logic controllers and its implementation for the control of a laboratory-scale aerodynamic plant in real time using MATLAB.

Computerized investigation of robust measurement systems

This paper deals with the development of software for investigating the robust properties of measurement systems and for their design and tuning in order to improve their robustness. The software constitutes Simulink models and m-files as extensions of the libraries of MATLAB. The investigations on continuous measurement systems (a self-balancing system) and discrete systems (ADCs) with improved robustness by using the internal model controller technique revealed new properties-fast dynamics, high accuracy, and discretization error reduction via multiple measurements.

Development of measurement systems with robustness feedback

The aim of this paper is to the design and study of a robustness feedback for an intelligent measurement system and more specifically for the main constituent of its digital part-the analog-to-digit converter. The robustness feedback is based on a simplified reference model of the converter. It is introduced by simple means in order to reduce the quantization error. The main contributions are: design of a new type of structure of analog-to-digit converter; development of a model of the robust converter; quantization error reduction by multiple measurements. The behavior of the proposed converter is studied in environment SIMULINK of the MATLAB program package.

Computerised investigation of robust measurement systems

The paper deals with the development of software for investigating the robust properties of measurement systems and for design and tuning in order to improve their robustness. The software consists of Simulink models and m-files as extensions of the libraries of MATLAB. Investigations of continuous measurement systems (a self-balancing system) and discrete systems (ADCs) with improved robustness using the internal model controller technique revealed new properties-fast dynamics, high accuracy, and discretisation error reduction via multiple measurements.

Design of Fuzzy Supervisor-Based Adaptive Process Control Systems

The modern industrial processes are difficult to model and control by classical means for their nonlinearity, inertia, model uncertainty and varying parameters. The adaptive fuzzy logic controllers (AFLCs) improve the system performance but are computationally hard to design and embed in programmable logic controllers (PLCs) for wider industrial applications. In this chapter a design approach for simple AFLCs is suggested, based on main controllers—linear, FLC or parallel distributed compensation (PDC), and fuzzy logic supervisors (FLSs) for on-line auto-tuning of their gains or scaling factors. The effect is a continuous adaptation of the control surface in response to plant changes. Approximation of the designed AFLC to a PDC equivalent on the basis of neuro-fuzzy and optimization techniques enables the stability analysis of the AFLC system using the indirect Lyapunov method and also its PLC implementation. The AFLC is applied for the real time control of the processes in a chemical reactor, a dryer, a two-tank and an air-conditioning systems, decreasing overshoot, settling time, control effort and coupling compared to classical FLC and linear control systems.

A fuzzy model for automated precision robot assembly of parts

Abstract The fuzzy logic theory is suitable for solving the vagueness, nonlinearity and complexity problems of an assembly process by implanting ’human reasoning and action’ in the automated devices. The aim of the paper is to develop a simple and fast fuzzy rule-based algorithm and to explore via simulation the possibility of its implementation for precision assembly of chamferless parts. The main contribution is a simple, fast and reliable fuzzy algorithm, ready for imbedding in the PLC of a SCARA type robot.

Intelligent approaches for control of nonlinear plants with significant time delay

The nonlinear plants with significant time delays are difficult to be controlled by classical means. In the present paper intelligent approaches are applied for the design of a nonlinear Smith predictor for compensation of the plant time delay based on a Takagi-Sugeno-Kang plant model and a fuzzy logic parallel distributed compensation (PDC). The design and the advantages of the PDC-Smith are illustrated in temperature control. The good compensation of both the plant time delay and nonlinearity results in faster system step responses in regard to a linear PI, a Smith and a PDC control systems.

Fuzzy logic approach to coupled level control

ABSTRACT Liquid level in coupled tanks is difficult to control by classical techniques because the plant is nonlinear, multivariable, often with no self-regulation and no model. Fuzzy logic (FL) enables a successful stable and robust control by simple means and a model-free design. This paper suggests a procedure for the design of two-variable FL controllers (FLCs) for the levels in a laboratory coupled-tank system. First a model-free two-variable Mamdani FLC is empirically developed and applied for real-time levels control. The plant input and output experimental data are then used for derivation via genetic algorithms optimization of a Takagi–Sugeno–Kang (TSK) plant model needed for FLC improvement. The TSK model is validated on a different set of experimental data and used in designing of two-variable linear proportional-plus-integral (PI) controller and parallel distributed compensation with local linear PI controllers. The performance of the systems with the designed controllers is compared in real-time levels control.