Rómulo Antão

Model-based control using interval type-2 fuzzy logic systems

Type-2 fuzzy logic introduced new formalisms capable of overcoming the inherent uncertainties of approximating real-world processes by computational models. Yet, despite increasingly present in the nonlinear modeling literature, model-based control theory does not seem to be taking full advantage of the improvements that type-2 fuzzy logic provides. Therefore, the present work proposes the development of a new control methodology based on the generalized predictive control theory supported by interval type-2 Takagi–Sugeno fuzzy logic systems. The developed control system is based on locally linear approximations of the type-2 Takagi–Sugeno model and will be evaluated using a nonlinear process based on the yeast fermentation reaction. For comparison purposes, two additional generalized predictive control implementations based on a linear auto-regressive model with exogenous inputs and a type-1 Takagi–Sugeno fuzzy model will be used. The performance of the closed loop systems will be evaluated by subjecting the process to quick changes in the operation regime and to unmeasured external disturbances. The mean squared error, control effort and typical transient step response metrics (overshoot and settling time) will provide the benchmark criteria. The results achieved demonstrate that at the expense of a small increase in the computational effort, type-2 fuzzy logic systems improve the transient behavior of the closed loop system, presenting significant advantages when the controlled process is subject to unmodeled disturbances.

Adaptive control of a buck converter with an ARM Cortex-M4

The capability of dealing with unpredictable variations of a process under manipulation is one of the most sought features when implementing digital control loops. Self tuning regulators are one type of control systems with such capability, existing several successful application of it in industry. However implementations of such systems are typically based on computationally intensive algorithms that, when applied to processes with fast dynamics, require high performance but complex and expensive embedded systems to cope with the required control-loop turnaround times. With the performance improvements brought by the new ARM Cortex M4, general purpose microcontrollers and advanced digital signal processing are no longer disjoint domains, becoming now possible to develop more versatile and robust control algorithms on affordable embedded systems with less restrictive computational limitations. Taking advantage of this architecture, this paper will present a real-time embedded adaptive controller applied to a Buck DC-DC converter. To assess the capabilities of this new architecture, comparative measurements of the algorithms CPU usage under different system configurations and results relative to the setpoint tracking capability of the adaptive controller under time-varying dynamics will be presented.

A self tuning regulator based on the ARM cortex-M4

The versatility and closed-loop performance resulting from the use of self-calibrating control systems are two of the most sought features when implementing a digital feedback loop based on embedded systems. However, such methods are typically based on computationally intensive algorithms that, when executed in low-cost embedded systems, severely restrict their usability to application with relatively slow dynamics in order to cope with the control-loop calculations turnaround time. Taking advantage of the new ARM Cortex-M4 microcontroller performance improvements in the digital signal processing field, this paper will present a real-time self-tuning regulator designed for a generic second order dynamic process. To assess the capabilities of this new architecture, a Buck DC-DC converter will be used as test scenario to present comparative measurements of the algorithms turnaround time and CPU load under different system configurations and results relative to the setpoint tracking capability of the adaptive controller under time-varying dynamics.

Fuzzy Logic Systems

The human mind has always shown a remarkable capability of coordinating a wide variety of physical and mental tasks without using any explicit measurements and computations.

System Modeling Using Type-2 Takagi-Sugeno Fuzzy Systems

The development of computational models capable of accurately describe a process’s dynamic response is a task ultimately dependent on its physical phenomena complexity and the type of disturbances that may affect its operation.

Takagi-Sugeno Fuzzy Logic Systems

The achievements obtained by Fuzzy Logic undoubtedly changed the way expert information is represented, manipulated, and interpreted in computational systems. Nevertheless, the initialization of Mamdani FLSs’ main parameters, namely its membership functions and their interdependency relations, is a process that depends on the knowledge of an expert (which may be subjective and is ultimately limited by its know-how). Takagi and Sugeno [1] were among the first researchers who recognized that FLSs could be further enhanced with autonomous learning techniques. Together, they proposed a new structure for the consequent part of the rules, introducing also methodologies to autonomously create and improve the FLSs’ performance. Their method uses heuristic and non-linear optimization algorithms for the antecedent part of the rule-base and a Kalman Filter for the consequent one. It is however for their innovative FLS’s structure supporting their work that Takagi and Sugeno are nowadays known in FLSs’ literature (effectively coining the concept of Takagi-Sugeno FLSs), serving their work as the stepping stone for many successful research topics.

Model Predictive Control Using Type-2 Takagi-Sugeno Fuzzy Systems

Model Predictive Control (MPC) is one of the most researched synthesis approaches which, based on a model of a process, computes the best control strategy according to a set of predefined goals over a future time horizon.

Processor-In-the-Loop Simulation

Empowered by the increasing computational power broadly available in current computer technology, the use of simulation software is an ubiquitous approach both in academia and industry during the development path of a large number of systems.

Assessing the performance of inkjet-printed coils on paper substrate for WPT

In recent years, wireless power transmission (WPT) has received special attention as a very valuable and convenient technology to power a wide range of equipment and devices, such as mobile phones, laptops, integrated circuits, electric vehicles[1], medical devices and other passive sensors. Magnetic resonance-based WPT is a directional coupled type of near-field power transmission suitable for short distances. Design of inductive coupling coils is crucial to the performance of WPT[2]. The aim of this work is to assess the potential of inkjet printing technology for cheaper and faster prototyping of inductively coupled coils on paper substrate.

Generalized Predictive Control using Interval Type-2 Fuzzy models

The development of Interval Type-2 Fuzzy Logic Systems has brought great improvements in the non-linear system modeling domain. However, in what concerns to the development of control systems, the approaches found in literature of Type-2 Fuzzy Sets do not seem to be taking fully advantage of the advances achieved by adaptive self-tuning algorithms, already well established in both academic and industrial communities. This work presents how a controller based on Generalized Predictive Control (GPC) theory can be developed based on an Interval Type-2 Takagi-Sugeno Fuzzy Model, providing details regarding the online model training mechanisms and controller parameters synthesis. This approach is then compared with two additional GPC implementations based on an Auto-Regressive model with eXogenous inputs (ARX) and Type-1 Takagi-Sugeno Fuzzy models. A Multiple-Input-Single-Output (MISO) Coupled Tank System will serve as benchmark system to evaluate the reference tracking capability and robustness of the controllers when subjected to different operation points and several unmodeled disturbances.