Skip to main content

Jorge Henrique

Followers

0

Following

1

Public Views

Vrije Universiteit Brussel

Armağan Karahanoğlu

University of Twente

Kansas State University

University of Derby

Rodrigo Lins Rodrigues

Universidade Federal de Pernambuco

Universidad de Sevilla

Viacheslav Kuleshov

Stockholm University

Nelson Mauro Maldonato

Università degli Studi di Napoli "Federico II"

Ricardo Montero

Universidad Autónoma de Occidente

ivanildo Araujo

UFCG - Universidade Federal de Campina Grande

Interests

Uploads

Papers by Jorge Henrique

Iterative Multidimensional Scaling for Industrial Process Monitoring

Monitoring high dimensional industrial processes for performance analysis and improvement should ... more Monitoring high dimensional industrial processes for performance analysis and improvement should be done in a low-dimensional space, where human-computer interaction is effective and where the human operator can easily identify the main actual features of the process. For this purpose space reduction must keep the relevant and informative geometric characteristics of the original space, using proper metrics. In this work

MULTIVARIABLE ADAPTIVE CONTROL USING A RECURRENT NEURAL NETWORK

An indirect adaptive controller based on a hybrid solution is presented and applied. Using data c... more An indirect adaptive controller based on a hybrid solution is presented and applied. Using data concerning the process, a dynamic recurrent neural network (Elman&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s type) is trained to model a general nonlinear discrete time system. By assuming a linearisation of this neural model, a time varying standard linear state space model is derived. As a consequence, some well-established conventional technique

Supervision and c-Means clustering of PID controllers for a solar power plant

International Journal of Approximate Reasoning, 1999

A hierarchical control strategy consisting on a supervisory switching of PID controllers, simpli ... more

Constrained neural model predictive control with guaranteed free offset

An extended model-based predictive control scheme is proposed and implemented on a bench three-ta... more An extended model-based predictive control scheme is proposed and implemented on a bench three-tanks system. This structure is based on a constrained local instantaneous linear model-based predictive controller complemented with a static offset compensator for guaranteeing that tracking errors converge to zero in a finite time. A non-linear state-space neural network architecture trained offline is used for modelling purposes and forming a seed from where linear models are extracted at each sampling time. Results from experiments show that this extended MPC scheme ensures a good tracking performance with zero steady-state offsets, in spite of modelling errors.

Adaptive RBFNN versus conventional self-tuning: comparison of two parametric model approaches for non-linear control

Control Engineering Practice, 2000

In this work a practical study evaluates two parametric modelling approaches — linear and non-lin... more In this work a practical study evaluates two parametric modelling approaches — linear and non-linear (neural) — for automatic adaptive control. The neural adaptive control is based on a developed hybrid learning technique using an adaptive (on-line) learning rate for a Gaussian radial basis function neural network. The linear approach is used for a self-tuning pole-placement controller. A selective forgetting

STATE SPACE NEURAL NETWORKS IN NON-LINEAR ADAPTIVE SYSTEM IDENTIFICATION AND CONTROL

Non-linear control design methods, such as feedback linearisation and output regulation theory, a... more Non-linear control design methods, such as feedback linearisation and output regulation theory, are effective techniques for solving non-linear control problems. However, they assume the exact knowledge of the true model and that the states are completely accessible, which is not always true in practice. The starting point for this paper is to develop a viable practical control strategy by combining the modelling capabilities of state space neural networks with the effectiveness of the output regulation theory. By using input and output measurements and based on Lyapunov stability and non-linear observation theories a stable on-line learning methodology for the network parameters is proposed. A practical implementation for controlling a multivariable process is included to illustrate the effectiveness of the proposed adaptive control methodology.

Scheduling of PID controllers by means of a neural network with application to a solar power plant

This paper concerns the application of a neural network control strategy to the distributed colle... more This paper concerns the application of a neural network control strategy to the distributed collector field of a solar power plant. The neural network is trained based on measured data from the plant providing a way of scheduling between a set of PID controllers, a priori tuned in different operating points by means of Takahashi rules. The present work consists in a hybrid scheme combining the potentialities of neural networks for approximation purposes with the well-know theory and widespread industrial application of PID techniques.

Adaptive neural model-based predictive control of a solar power plant

This paper describes the application of a non-linear adaptive constrained model-based predictive ... more This paper describes the application of a non-linear adaptive constrained model-based predictive control scheme to the distributed collector field of a solar power plant at the Plataforma Solar de Almería (Spain). This methodology exploits the intrinsic non-linear modelling capabilities of nonlinear state-space neural networks and their online training by means of an unscented Kalman filter. Tests on the ACUREX field illustrate the great engineering potential of the proposed control strategy. Index Terms-Solar power plant; state-space neural networks; online training; constrained model-based predictive control; dual unscented Kalman filter. Adaptive Neural Model-Based Predictive Control of a Solar Power Plant.

Adaptive neural output regulation control of a solar power plant

Control Engineering Practice, 2010

This work proposes an indirect adaptive nonlinear control scheme based on a recurrent neural netw... more This work proposes an indirect adaptive nonlinear control scheme based on a recurrent neural network and the output regulation theory.The neural model is first trained off-line, being further improved by means of an on-line learning strategy using the Lyapunov and nonlinear observation theories.The regulation problem is solved by an iterative strategy, formulated as an eigenvalue assignment problem, ensuring the convergence

Reset-Driven Fault Tolerance

A common approach in embedded systems to achieve faulttolerance is to reboot the computer wheneve... more A common approach in embedded systems to achieve faulttolerance is to reboot the computer whenever some non-permanent error is detected. All the system code and data are recreated from scratch, and a previously established checkpoint, hopefully not corrupted, is used to restart the application data. The confidence is thus restored on the activity of the computer. The idea explored in this paper is that of unconditionally resetting the computer in each control frame (the classic read sensors → calculate control action → update actuators cycle). A stable-storage based in RAM is used to preserve the system’s state between consecutive cleanups and a standard watchdog timer guarantees that a reset is forced whenever an error crashes the system. We have evaluated this approach by using fault-injection in the controller of a standard temperature control system. The experimental observations show that the Reset-Driven Fault Tolerance is a very simple yet effective technique to improve reliability at an extremely low cost since it is a conceptually simple, software only solution with the advantage of being application independent.

Choosing Real-Time Predictors for Ventricular Arrhythmia Detection

International Journal of Pattern Recognition and Artificial Intelligence, 2007

The risk of developing life-threatening ventricular arrhythmias in patients with structural heart... more The risk of developing life-threatening ventricular arrhythmias in patients with structural heart disease is higher with increased occurrence of premature ventricular complex (PVC). Therefore, reliable detection of these arrhythmias is a challenge for a cardiovascular diagnosis system. While early diagnosis is critical, the task of its automatic detection and classification becomes crucial. Therefore, the underlying models should be efficient, albeit ensuring robustness. Although neural networks (NN) have proven successful in this setting, we show that kernel-based learning algorithms achieve superior performance. In particular, recently developed sparse Bayesian methods, such as, Relevance Vector Machines (RVM), present a parsimonious solution when compared with Support Vector Machines (SVM), yet revealing competitive accuracy. This can lead to significant reduction in the computational complexity of the decision function, thereby making RVM more suitable for real-time applications.

Real-time Adaptive Training of RBFNN using a Selective Forgetting Algorithm

In this work, we investigate the real-time identification problem of dynamic non-linear systems u... more In this work, we investigate the real-time identification problem of dynamic non-linear systems using a neural network approach. The neural network model structure is chosen to be the Gaussian Radial Basis Function (RBF) type, due to its universal approximation property and also to the fact that the parameters are linearly related to the outputs, allowing linear learning algorithms, with fast convergence speed and low computational time, suitable for real-time applications. We propose a new hybrid learning, where we used first an adaptive learning rate with process monitoring. Secondly, taking advantage of the locality property of this type of networks, we apply a selective forgetting algorithm. Simulation examples are provided, applied to the identification of a theoretical model and to a real laboratory process.

FUZZY SUPERVISOR AND FEEDFORWARD CONTROL OF A SOLAR POWER PLANT USING ACCESSIBLE DISTURBANCES

Page 1. FUZZY SUPERVISOR AND FEEDFORWARD CONTROL OF A SOLAR POWER PLANT USING ACCESSIBLE DISTURBA... more

APPLICATION OF A RECURRENT NEURAL NETWORK IN ON LINE MODELLING OF REAL-TIME SYSTEMS

Page 1. APPLICATION OF A RECURRENT NEURAL NETWORK IN ON-LINE MODELLING OF REAL-TIME SYSTEMS Jorge... more

ADAPTIVE RBFNN VERSUS SELF-TUNING CONTROLLER: AN EXPERIMENTAL COMPARATIVE STUDY

In this paper a comparative study evaluates two different techniques: neural adaptive control and... more In this paper a comparative study evaluates two different techniques: neural adaptive control and self tuning (ST) control. The neural adaptive control is based on a new hybrid learning technique using an adaptive learning rate for the on-line learning of a Gaussian Radial Basis Function Neural Network (RBFNN) type. In the self tuning structure the control parameters are updated from

Iterative Multidimensional Scaling for Industrial Process Monitoring

Monitoring high dimensional industrial processes for performance analysis and improvement should ... more Monitoring high dimensional industrial processes for performance analysis and improvement should be done in a low-dimensional space, where human-computer interaction is effective and where the human operator can easily identify the main actual features of the process. For this purpose space reduction must keep the relevant and informative geometric characteristics of the original space, using proper metrics. In this work

MULTIVARIABLE ADAPTIVE CONTROL USING A RECURRENT NEURAL NETWORK

An indirect adaptive controller based on a hybrid solution is presented and applied. Using data c... more An indirect adaptive controller based on a hybrid solution is presented and applied. Using data concerning the process, a dynamic recurrent neural network (Elman&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s type) is trained to model a general nonlinear discrete time system. By assuming a linearisation of this neural model, a time varying standard linear state space model is derived. As a consequence, some well-established conventional technique

Supervision and c-Means clustering of PID controllers for a solar power plant

International Journal of Approximate Reasoning, 1999

A hierarchical control strategy consisting on a supervisory switching of PID controllers, simpli ... more

Constrained neural model predictive control with guaranteed free offset

An extended model-based predictive control scheme is proposed and implemented on a bench three-ta... more An extended model-based predictive control scheme is proposed and implemented on a bench three-tanks system. This structure is based on a constrained local instantaneous linear model-based predictive controller complemented with a static offset compensator for guaranteeing that tracking errors converge to zero in a finite time. A non-linear state-space neural network architecture trained offline is used for modelling purposes and forming a seed from where linear models are extracted at each sampling time. Results from experiments show that this extended MPC scheme ensures a good tracking performance with zero steady-state offsets, in spite of modelling errors.

Adaptive RBFNN versus conventional self-tuning: comparison of two parametric model approaches for non-linear control

Control Engineering Practice, 2000

In this work a practical study evaluates two parametric modelling approaches — linear and non-lin... more In this work a practical study evaluates two parametric modelling approaches — linear and non-linear (neural) — for automatic adaptive control. The neural adaptive control is based on a developed hybrid learning technique using an adaptive (on-line) learning rate for a Gaussian radial basis function neural network. The linear approach is used for a self-tuning pole-placement controller. A selective forgetting

Iterative Multidimensional Scaling for Industrial Process Monitoring

Monitoring high dimensional industrial processes for performance analysis and improvement should ... more Monitoring high dimensional industrial processes for performance analysis and improvement should be done in a low-dimensional space, where human-computer interaction is effective and where the human operator can easily identify the main actual features of the process. For this purpose space reduction must keep the relevant and informative geometric characteristics of the original space, using proper metrics. In this work

MULTIVARIABLE ADAPTIVE CONTROL USING A RECURRENT NEURAL NETWORK

An indirect adaptive controller based on a hybrid solution is presented and applied. Using data c... more An indirect adaptive controller based on a hybrid solution is presented and applied. Using data concerning the process, a dynamic recurrent neural network (Elman&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s type) is trained to model a general nonlinear discrete time system. By assuming a linearisation of this neural model, a time varying standard linear state space model is derived. As a consequence, some well-established conventional technique

Supervision and c-Means clustering of PID controllers for a solar power plant

International Journal of Approximate Reasoning, 1999

A hierarchical control strategy consisting on a supervisory switching of PID controllers, simpli ... more

Constrained neural model predictive control with guaranteed free offset

An extended model-based predictive control scheme is proposed and implemented on a bench three-ta... more An extended model-based predictive control scheme is proposed and implemented on a bench three-tanks system. This structure is based on a constrained local instantaneous linear model-based predictive controller complemented with a static offset compensator for guaranteeing that tracking errors converge to zero in a finite time. A non-linear state-space neural network architecture trained offline is used for modelling purposes and forming a seed from where linear models are extracted at each sampling time. Results from experiments show that this extended MPC scheme ensures a good tracking performance with zero steady-state offsets, in spite of modelling errors.

Adaptive RBFNN versus conventional self-tuning: comparison of two parametric model approaches for non-linear control

Control Engineering Practice, 2000

In this work a practical study evaluates two parametric modelling approaches — linear and non-lin... more In this work a practical study evaluates two parametric modelling approaches — linear and non-linear (neural) — for automatic adaptive control. The neural adaptive control is based on a developed hybrid learning technique using an adaptive (on-line) learning rate for a Gaussian radial basis function neural network. The linear approach is used for a self-tuning pole-placement controller. A selective forgetting

STATE SPACE NEURAL NETWORKS IN NON-LINEAR ADAPTIVE SYSTEM IDENTIFICATION AND CONTROL

Non-linear control design methods, such as feedback linearisation and output regulation theory, a... more Non-linear control design methods, such as feedback linearisation and output regulation theory, are effective techniques for solving non-linear control problems. However, they assume the exact knowledge of the true model and that the states are completely accessible, which is not always true in practice. The starting point for this paper is to develop a viable practical control strategy by combining the modelling capabilities of state space neural networks with the effectiveness of the output regulation theory. By using input and output measurements and based on Lyapunov stability and non-linear observation theories a stable on-line learning methodology for the network parameters is proposed. A practical implementation for controlling a multivariable process is included to illustrate the effectiveness of the proposed adaptive control methodology.

Scheduling of PID controllers by means of a neural network with application to a solar power plant

This paper concerns the application of a neural network control strategy to the distributed colle... more This paper concerns the application of a neural network control strategy to the distributed collector field of a solar power plant. The neural network is trained based on measured data from the plant providing a way of scheduling between a set of PID controllers, a priori tuned in different operating points by means of Takahashi rules. The present work consists in a hybrid scheme combining the potentialities of neural networks for approximation purposes with the well-know theory and widespread industrial application of PID techniques.

Adaptive neural model-based predictive control of a solar power plant

This paper describes the application of a non-linear adaptive constrained model-based predictive ... more This paper describes the application of a non-linear adaptive constrained model-based predictive control scheme to the distributed collector field of a solar power plant at the Plataforma Solar de Almería (Spain). This methodology exploits the intrinsic non-linear modelling capabilities of nonlinear state-space neural networks and their online training by means of an unscented Kalman filter. Tests on the ACUREX field illustrate the great engineering potential of the proposed control strategy. Index Terms-Solar power plant; state-space neural networks; online training; constrained model-based predictive control; dual unscented Kalman filter. Adaptive Neural Model-Based Predictive Control of a Solar Power Plant.

Adaptive neural output regulation control of a solar power plant

Control Engineering Practice, 2010

This work proposes an indirect adaptive nonlinear control scheme based on a recurrent neural netw... more This work proposes an indirect adaptive nonlinear control scheme based on a recurrent neural network and the output regulation theory.The neural model is first trained off-line, being further improved by means of an on-line learning strategy using the Lyapunov and nonlinear observation theories.The regulation problem is solved by an iterative strategy, formulated as an eigenvalue assignment problem, ensuring the convergence

Reset-Driven Fault Tolerance

A common approach in embedded systems to achieve faulttolerance is to reboot the computer wheneve... more A common approach in embedded systems to achieve faulttolerance is to reboot the computer whenever some non-permanent error is detected. All the system code and data are recreated from scratch, and a previously established checkpoint, hopefully not corrupted, is used to restart the application data. The confidence is thus restored on the activity of the computer. The idea explored in this paper is that of unconditionally resetting the computer in each control frame (the classic read sensors → calculate control action → update actuators cycle). A stable-storage based in RAM is used to preserve the system’s state between consecutive cleanups and a standard watchdog timer guarantees that a reset is forced whenever an error crashes the system. We have evaluated this approach by using fault-injection in the controller of a standard temperature control system. The experimental observations show that the Reset-Driven Fault Tolerance is a very simple yet effective technique to improve reliability at an extremely low cost since it is a conceptually simple, software only solution with the advantage of being application independent.

Choosing Real-Time Predictors for Ventricular Arrhythmia Detection

International Journal of Pattern Recognition and Artificial Intelligence, 2007

The risk of developing life-threatening ventricular arrhythmias in patients with structural heart... more The risk of developing life-threatening ventricular arrhythmias in patients with structural heart disease is higher with increased occurrence of premature ventricular complex (PVC). Therefore, reliable detection of these arrhythmias is a challenge for a cardiovascular diagnosis system. While early diagnosis is critical, the task of its automatic detection and classification becomes crucial. Therefore, the underlying models should be efficient, albeit ensuring robustness. Although neural networks (NN) have proven successful in this setting, we show that kernel-based learning algorithms achieve superior performance. In particular, recently developed sparse Bayesian methods, such as, Relevance Vector Machines (RVM), present a parsimonious solution when compared with Support Vector Machines (SVM), yet revealing competitive accuracy. This can lead to significant reduction in the computational complexity of the decision function, thereby making RVM more suitable for real-time applications.

Real-time Adaptive Training of RBFNN using a Selective Forgetting Algorithm

In this work, we investigate the real-time identification problem of dynamic non-linear systems u... more In this work, we investigate the real-time identification problem of dynamic non-linear systems using a neural network approach. The neural network model structure is chosen to be the Gaussian Radial Basis Function (RBF) type, due to its universal approximation property and also to the fact that the parameters are linearly related to the outputs, allowing linear learning algorithms, with fast convergence speed and low computational time, suitable for real-time applications. We propose a new hybrid learning, where we used first an adaptive learning rate with process monitoring. Secondly, taking advantage of the locality property of this type of networks, we apply a selective forgetting algorithm. Simulation examples are provided, applied to the identification of a theoretical model and to a real laboratory process.

FUZZY SUPERVISOR AND FEEDFORWARD CONTROL OF A SOLAR POWER PLANT USING ACCESSIBLE DISTURBANCES

Page 1. FUZZY SUPERVISOR AND FEEDFORWARD CONTROL OF A SOLAR POWER PLANT USING ACCESSIBLE DISTURBA... more

APPLICATION OF A RECURRENT NEURAL NETWORK IN ON LINE MODELLING OF REAL-TIME SYSTEMS

Page 1. APPLICATION OF A RECURRENT NEURAL NETWORK IN ON-LINE MODELLING OF REAL-TIME SYSTEMS Jorge... more

ADAPTIVE RBFNN VERSUS SELF-TUNING CONTROLLER: AN EXPERIMENTAL COMPARATIVE STUDY

In this paper a comparative study evaluates two different techniques: neural adaptive control and... more In this paper a comparative study evaluates two different techniques: neural adaptive control and self tuning (ST) control. The neural adaptive control is based on a new hybrid learning technique using an adaptive learning rate for the on-line learning of a Gaussian Radial Basis Function Neural Network (RBFNN) type. In the self tuning structure the control parameters are updated from

Iterative Multidimensional Scaling for Industrial Process Monitoring

Monitoring high dimensional industrial processes for performance analysis and improvement should ... more Monitoring high dimensional industrial processes for performance analysis and improvement should be done in a low-dimensional space, where human-computer interaction is effective and where the human operator can easily identify the main actual features of the process. For this purpose space reduction must keep the relevant and informative geometric characteristics of the original space, using proper metrics. In this work

MULTIVARIABLE ADAPTIVE CONTROL USING A RECURRENT NEURAL NETWORK

An indirect adaptive controller based on a hybrid solution is presented and applied. Using data c... more An indirect adaptive controller based on a hybrid solution is presented and applied. Using data concerning the process, a dynamic recurrent neural network (Elman&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s type) is trained to model a general nonlinear discrete time system. By assuming a linearisation of this neural model, a time varying standard linear state space model is derived. As a consequence, some well-established conventional technique

Supervision and c-Means clustering of PID controllers for a solar power plant

International Journal of Approximate Reasoning, 1999

A hierarchical control strategy consisting on a supervisory switching of PID controllers, simpli ... more

Constrained neural model predictive control with guaranteed free offset

An extended model-based predictive control scheme is proposed and implemented on a bench three-ta... more An extended model-based predictive control scheme is proposed and implemented on a bench three-tanks system. This structure is based on a constrained local instantaneous linear model-based predictive controller complemented with a static offset compensator for guaranteeing that tracking errors converge to zero in a finite time. A non-linear state-space neural network architecture trained offline is used for modelling purposes and forming a seed from where linear models are extracted at each sampling time. Results from experiments show that this extended MPC scheme ensures a good tracking performance with zero steady-state offsets, in spite of modelling errors.

Adaptive RBFNN versus conventional self-tuning: comparison of two parametric model approaches for non-linear control

Control Engineering Practice, 2000

In this work a practical study evaluates two parametric modelling approaches — linear and non-lin... more In this work a practical study evaluates two parametric modelling approaches — linear and non-linear (neural) — for automatic adaptive control. The neural adaptive control is based on a developed hybrid learning technique using an adaptive (on-line) learning rate for a Gaussian radial basis function neural network. The linear approach is used for a self-tuning pole-placement controller. A selective forgetting