Abstract: In this work, the implementation of an optimization algorithm in an FPGA is presented with the objective of maximizing the hydrogen flow rate produced by the degradation of organic matter in a microbial electrolysis cell. Through numerical simulation tests, the correct performance of the digital architecture was verified. The feasibility of implementing this optimization algorithm in an FPGA to replace a computer is verified through an analysis of hardware resources, execution time and power consumed.

Abstract: The periodic operation in a CSTR for the hydrolysis of acetic anhydride was studied by a first-harmonic balance approach. Two temperature conditions were evaluated for the system (low and high temperature) and the performance was examined for different periodic signals (i.e., feed concentration, feed temperature and jacket temperature) over a range of frequencies and amplitudes. The first-harmonic balance approach simplified the differential equations model to a set of nonlinear algebraic equations and the nonlinearities, like reaction kinetic, were approximate considering the first bases of the Fourier series. The results showed that the first-harmonic balance estimated the response of the time-average variables (i.e., concentration and reactor temperature) for the periodic operation of the CSTR, and it was possible determine the optimal conditions to improve performance of the acetic anhydride hydrolysis.

Resumen: Los invernaderos constituyen un entorno cerrado en el que se pueden controlar las variables climáticas para permitir el crecimiento óptimo y desarrollo de los cultivos. El control de la humedad y temperatura se consideran fundamentales para tal fin. El control de invernaderos es complicado por la complejidad e incertidumbre del modelo, así como las interacciones entre variables ambientales externas y del invernadero. En este trabajo se aborda el diseño de esquemas de control multivariable del clima de un invernadero usando dos enfoques de control simples y prácticos: (i) el control basado en compensación de error de modelado, y (ii) el control de modo predictivo no-lineal.

Resumen: Los procesos de cultivo de células animales se han consolidado como una tecnología madura y de gran relevancia dentro de la industria biofarmacéutica. La gran mayoría de los procesos industriales de cultivo en biorreactores son operados por lote alimentado o semi-lote. Debido a su naturaleza de operación dinámica, perturbaciones, no-linealidades e incertidumbres, los biorreactores lote y semi-lote presentan retos interesantes para fines de su operación. En el presente trabajo se plantea la optimización dinámica y el diseño e implementación numérica de un controlador de modo predictivo no-lineal para un caso de estudio que describe el modelo de un bioproceso semi-lote para el cultivo de células de mamífero tipo CHO con la finalidad de optimizar su productividad.

Resumen: Se estudia el problema de estabilidad y estabilización de sistemas lineales positivos, a través de las propiedades de las matrices Metzler y Hurwitz, se establece la invariancia de politopos convexos, de tal manera que es posible describir la geometría de la dinámica estable para dichos sistemas lineales positivos.

Abstract: Estimated state feedback is a well-known control strategy for many types of systems. In the case of delayed systems, the general problem with estimated state feedback is that the necessary and sufficient stability conditions cannot be easily obtained. Many works in the literature make use of linear matrix inequalities where only sufficient conditions for the stability problem are available. This work proposes an observer-based state feedback control strategy for first-order systems with an integrator, unstable and delay. The proposed strategy is based on an observer, which is an extension of the work done previously. Necessary and sufficient conditions for the proposed control scheme are obtained and numerical simulations are presented to evaluate the behaviour of the proposed control scheme.

Abstract: This work considers stabilization and control problems of linear systems with time delay in the direct loop. It is well known that time-delays complicate the stability analysis and controller design. In this work a methodology to design a continuous predictor for delayed systems is proposed. The key point in the proposal is to guarantee the existence of the predictor, regardless of issues such as the order system, its instability or the delay size. The main contribution of this work is based on a hybrid predictor that consider the case when the size delays in the plant is not conveniently divisible by an integer. The behavior of the control strategy is illustrated through numerical simulations.

Abstract: This work presents a model-based predictive control (MPC) scheme using a multi-predictor for high-order linear systems with an unstable pole with long delay in direct loop, the stability analysis of this type of system is difficult due to the delay size considered. To solve the stabilization problem, as a first step a digital multi-predictor scheme is presented that estimates the intermediate variables of the process that will later be used for the design of an MPC controller to control the plant. The convergence of the multi-predictor scheme is based on a static output feedback, for this reason, the conditions that ensure the stability of an output feedback scheme are presented. The performance of the method is shown through numerical examples in simulation, likewise, some advantages of the methodology with respect to other controllers are illustrated.

Abstract: This paper aims to design a robust output-based control to regulate the output for a class of uncertain linear systems with input saturation and state constraints, in the presence of external disturbances. The proposed robust control approach is composed of a homogeneous observer, that guarantees nite-time convergence of the state estimation error to a neighborhood of the origin, and a linear control law, that is designed based on the attractive ellipsoid method and a barrier Lyapunov function approach, taking into account the input and state constraints. The synthesis of the robust-output-based control is given in terms of linear matrix inequalities. Simulation results show the workability of the proposed robust control approach.

Abstract: This paper describes the Attractive Ellipsoid Method (AEM) application, which uses the state estimates obtained by a sliding mode observer (SMO) for a wide class of quasi-Lipschitz nonlinear stochastic discrete-time systems. For the extended vector, containing state estimation and tracking errors as its components, we prove the mean square convergence to an attractive ellipsoid, which "size" is done as small as possible by the corresponding optimal selection of the gain matrices in both the SMO and in the linear feedback, using obtained current state estimates. It is shown that the procedure of the gain matrices optimization consists of the numerical solution of a corresponding matrix optimization problem subject to a set of bilinear matrix inequalities (BMIs), which by a special transformation procedure can be converted to a set of linear matrix inequalities (LMIs). An illustrative example shows the effectiveness of the suggested approach.

Resumen: En este artículo se propone un controlador por retroalimentación de estado y un controlador por retroalimentaci&ocute;n de salida, ambos proporcionan estabilidad en tiempo finito para la dinámica del doble integrador. En comparación con otros esquemas reportados, el sistema retroalimentado no es homogéneo de grado negativo y no es posible utilizar los resultados establecidos para sistemas homogéneos. Este problema se resuelve proponiendo funciones de Lyapunov estrictas. También se incluye una simulación numérica para mostrar el desempeño de los controladores propuestos.

Resumen: En este artículo, un diseño de controlador super-twisting multivariable para sistemas de dos salidas con matriz de desacoplamiento incierta es presentado. De igual forma que el super-twisting clásico para sistemas de una salida, este controlador puede rechazar perturbaciones Lipschitz, es decir, cuya derivada se encuentre acotada por una constante. La estabilidad del origen del sistema en lazo cerrado es probada utilizando una función estricta de Lyapunov. Asimismo, eL funcionamiento del controlador es mostrado a través de simulaciones.

Abstract: This work presents the design of a fault accommodation control for trajectory tracking in a Quad-Rotor under multiple actuator faults and external disturbances. The faults are modeled as partial loss of effectiveness. The proposed scheme is composed of a fault identification module, based on a finite-time sliding mode observer, and a robust nominal controller. Such a strategy uses the fault identification to partially compensate for the effect of faults, while the robust nominal controller deals with external disturbances. The performance of the proposed scheme is validated through numerical simulations.

Abstract: The present work proposes a novel methodology for failure analysis using knot theory. In this particular case a LC circuit of order four is studied, where the current and voltage signals are analyzed by means of the above mentioned method, and where a topological invariant is obtained under normal conditions. This procedure is also performed when a fault occurs in the dielectric of the capacitors. As a result, it is shown how the changes in the dielectric are manifested and associated to a certain topological invariant. This provides justification for this case based on topological invariants. An advantage of the proposed approach is that no equations describing the system are required to be implemented.

Abstract: This paper deals with the detection and localization of damage in pipelines. The purpose is to characterize the transient wave propagation in a faulty pipeline with a viscous absorption fluid and the pattern response to an external acoustic source as a function of the fault type and position. The study assumes an infinite-dimensional linear model for the wave propagation in the pipeline and lumped models for the leak and blockage faults. The key to the study is to analyze the wave propagation model through graph theory and by the reflection coefficient to solve these tasks. The system decomposition according to the fault scenarios allows the characterization of the impulse input response for a leak and blockage in the pipeline. For the specific case of a leak and a blockage, the impulse patterns are established and validated by simulation.

Abstract: The development of fault tolerant control systems is the subject of this paper. By using supplementary control it is possible to increase the fault-tolerant capability of a control system. The supplementary control has the advantage of not modifying the nominal control, because the control action that is added to compensate the effect of faults on the system is implemented by an additional control loop. The nominal control loop remains as an inner loop and the new control loop as an outer loop of a cascade control. The main advantages of the proposal are that fault tolerance is achieved without modifying the nominal control, the correction to the effect of faults is done automatically and it is possible to improve the behavior of the system. Simulation results confirm the described results.

Abstract: In this work, the solution of the trajectory tracking control problem for a class of port Hamiltonian systems is presented. The considered class is characterized by the fact that the interconnection matrix exhibits a particular structure that corresponds to the condition that the internal interconnection of the components of the vector state is modulated by the state itself. An additional feature of the presented result is the possibility to deal with under-actuated system. Regarding the stability analysis, the structure of the approached class allows to formulate the control error dynamic in such a way that using well-known results from the perturbed systems theory it is possible to obtain asymptotic stability properties for the closed-loop system. The usefulness of the contribution is illustrated by solving the speed tracking control problem of the Permanent Magnet Synchronous Motor.

Abstract: This paper proposes a pressure management method for water distribution systems considering the mathematical model under the water hammer equations. Two water systems are analyzed: a simple pipeline and a hydraulic network with two branches. A Pressure Reducing Valve (PRV) is modeled together with the dynamical model for controlling the pressure. This PRV works as the primary actuator, which controls the water ows under a free-leak and leakage scenario. The control strategy is based on a proportional Integral (PI) controller. The simulation results show a small but signicant reduction of water losses over long periods.

Abstract: This work presents a comparison of two nonlinear speed-tracking controllers for three-phase permanent-magnet synchronous motor (PMSM) drives. The chosen algorithms are the Interconnection and Damping Assignment Passivity-Based Control (IDA-PBC) and the classical Field-Oriented Control (FOC). Furthermore, a detailed and realistic simulation study is performed, taking into account the power electronic converter. Finally, some figures of merit are used to evaluate the controllers' performance.

Abstract: This work addresses the problem of stator short-circuit fault detection for an in-wheel Brushless Direct Current (BLDC) motor using a perceptron artificial neural network. The proposed FD scheme was compared with the motor current signature analysis based on discrete Fourier transform and discrete wavelet transform. The algorithms were validated on a test rig with an in-wheel BLDC motor for light electric vehicles.

Resumen: Este artículo presenta un enfoque para la detección de daño en edificios utilizando observadores intervalo. Primero, se diseña el observador intervalo para establecer las cotas superiores e inferiores de las variables de desplazamientos y velocidades de la respuesta estructural, bajo condiciones nominales. Suponiendo que durante la actividad sísmica los parámetros estructurales cambian, es de esperarse que con la respuesta también suceda lo mismo. Por consiguiente, los pisos que presentan daño estructural tendrán mayor cambio en su respuesta. De modo que si tal comportamiento es mayor o menor que los umbrales adaptables, establecidos por el observador intervalo, se asume como un claro indicador de daño. Esto es consecuencia de la pérdida de rigidez en el piso dañado. Los resultados de simulación confirman que el método propuesto es prometedor para aplicaciones prácticas.

Abstract: In this work, an open-circuit fault detection algorithm is presented for the power electronic converter of a wind generation system under voltage sags and swells. The system is based on a doubly-fed induction generator (DFIG) and supplies power to the grid through a three-phase back-to-back neutral-point clamped (NPC) converter. The power electronic converter is controlled using a field-oriented control (FOC) scheme. The fault detection method is based on the discrete wavelet transform of the DFIG current signals.