Cookies Policy
We use cookies to improve our site and your experience. By continuing to browse our site you accept our cookie policy. Find out More
Close
  • Menu
About

About

I was born in Oporto, Portugal, in 1958. I was graduatedi n Electrical Engineering in 1981 at Opoto University, and received the MSc degree in Computers and Digital Systems in 1987, the Ph.D. degree in Electrical Engineering in 1994 and the Agregado degree in 2016, all from
Oporto University. From 1983 to 1994 I worked as an assistant lecturer in the Electrical Engineering Department of the Oporto University. In 1985 I started a full-time academic career, and presently I am a Lecturer in Electrical Engineering at the same University.

From 1985 to 1989  I developed his research in INESC. I moved d to the Institute for Systems and Robotics- Oporto (ISRP) in 1989 where I stayed until 2018. I have joined the INESC TEC in 2018.

My research interests are Control, Estimation, Dynamical Systems Identification including multi-dimensional systems, with applications ranging from Bimedical Systems to Energy Systems.

I am author and co-author of dozens of papers published in international journals and proceedings of international conferences. I am a member of the the Portuguese Association of Automatic Control (APCA), IEEE CST and of the IEEE CST International Technical Committee on Systems Identification and Adaptive Control,  the IEEE CST International Technical Committee on Health and Medical Systems TC  and of the IFAC (International Federation on Automatic Control) Technical Committee on Signal Processing.

Interest
Topics
Details

Details

001
Publications

2019

A Dynamic Mode Decomposition Approach with Hankel Blocks to Forecast Multi-Channel Temporal Series

Authors
Filho, EV; Dos Santos, PL;

Publication
IEEE Control Systems Letters

Abstract
Forecasting is a task with many concerns, such as the size, quality, and behavior of the data, the computing power to do it, etc. This letter proposes the dynamic mode decomposition (DMD) as a tool to predict the annual air temperature and the sales of a stores' chain. The DMD decomposes the data into its principal modes, which are estimated from a training data set. It is assumed that the data is generated by a linear time-invariant high order autonomous system. These modes are useful to find the way the system behaves and to predict its future states, without using all the available data, even in a noisy environment. The Hankel block allows the estimation of hidden oscillatory modes, by increasing the order of the underlying dynamical system. The proposed method was tested in a case study consisting of the long term prediction of the weekly sales of a chain of stores. The performance assessment was based on the best fit percentage index. The proposed method is compared with three neural network-based predictors. © 2017 IEEE.

2018

The secrets of Segway revealed to students: revisiting the inverted pendulum

Authors
Azevedo Perdicoulis, TPA; Lopes dos Santos, PL;

Publication
2018 13TH APCA INTERNATIONAL CONFERENCE ON CONTROL AND SOFT COMPUTING (CONTROLO)

Abstract
This article revisits the inverted pendulum-in particular, analyses a simplified model of a Segway, with a view to exploring its capabilities in Control Systems Engineering education. The integration between the theoretic and practical side is achieved through simulation, and in particular by using MathWorks software. We also present a structure for the work to be done in the Laboratory class and propose a solution for the problem.

2018

An extended instrument variable approach for nonparametric LPV model identification

Authors
Lima, MML; Romano, RA; dos Santos, PL; Pait, F;

Publication
IFAC PAPERSONLINE

Abstract
Linear parameter varying models (LPV) have proven to be effective to describe non-linearities and time-varying behaviors. In this work, a new non-parametric estimation algorithm for state-space LPV models based on support vector machines is presented. This technique allows the functional dependence between the model coefficients and the scheduling signal to be "learned" from the input and output data. The proposed algorithm is formulated in the context of instrumental (IV) estimators, in order to obtain consistent estimates for general noise conditions. The method is based on a canonical state space representation and admits a predictor form that has shown to be suitable for system identification, as it leads to a convenient regression form. In addition, this predictor has an inherent filtering feature. In the context of vector support machines, such filtering mechanism leads to two-dimensional data processing, which can be used to decrease the variance of estimates due to noisy data. The performance of the proposed approach is evaluated from simulated data subject to different noise scenarios. The technique was able to reduce the error due to the variance of the estimator in most of the analyzed scenarios.

2018

An Iterative MOLI-ZOFT Approach for the Identification of MISO LTI Systems

Authors
Saraiva, PG; dos Santos, PL; Pait, F; Romano, RA; Perdicoulis, TP;

Publication
2018 13TH APCA INTERNATIONAL CONFERENCE ON CONTROL AND SOFT COMPUTING (CONTROLO)

Abstract
In this paper, a new system identification algorithm is proposed for linear and time invariant systems with multiple input and single output. The system is described by a state-space model in the canonical observable form and represented by a Luenberger observer with a known state matrix. Thence, the identification problem is reduced to the estimation of the system input matrix and the observer gain which can be performed by a simple Least Square Estimator. The quality of the estimator depends on the observer state matrix. In the proposed algorithm, this matrix is found by an iterative process where, in each iteration, a state matrix called curiosity is generated. A weight depending on the value of the Least Square Cost is associated to each curiosity. The optimal state matrix is the barycenter of the curiosities. This iterative process is a free derivative optimization algorithm with its roots in non-iterative barycenter methods previously introduced to solve adaptive control and system identification problems. Although the Barycenter iterative version was recently proposed as an optimization method, here it will be implemented in an identification algorithm for the first time.