Rosa M. Fernandez-Canti

Multiplatform virtual laboratory for engineering education

An autonomous and multiplatform virtual laboratory for educational purposes is presented. The implemented platform includes a server with a SSH (Secure SHell) connection and a separated repository containing the virtual experiments. The programming of the experiments is implemented in Java language based tool, the EJS (Easy Java Simulation) and uses an external computation engine, for example Matlab. The virtual laboratory provides control system experiments at University level. Two application examples are described, namely, a magnetic levitator and an inverted pendulum-cart system. The virtual laboratory has been successfully used for education and training of Electronics Engineering students. A discussion of the results of this e-learning experience is also presented.

Set-membership identification and fault detection using a Bayesian framework

This paper deals with the problem of set-membership identification and fault detection using a Bayesian framework. The paper presents how the set-membership model estimation problem can be reformulated from a Bayesian viewpoint in order to determine the feasible parameter set and, in a posterior fault detection stage, to check the consistency between data and the model. The paper shows that, assuming uniform distributed measurement noise and flat model prior probability distribution, the Bayesian approach leads to the same feasible parameter set than the set-membership strips technique and, additionally, can deal with models nonlinear in the parameters. The procedure and results are illustrated by means of the application to a quadruple tank process.

Nonlinear set-membership identification and fault detection using a Bayesian framework: Application to the wind turbine benchmark

This paper deals with the problem of nonlinear set-membership identification and fault detection using a Bayesian framework. The paper presents how the set-membership model estimation can be reformulated from a Bayesian viewpoint in order to determine the feasible parameter set and, in a posterior fault detection stage, to check the consistency between the model and the data. The paper shows that the Bayesian approach, assuming uniform distributed measurement noise and flat model prior probability distribution, leads to the same feasible parameter set as the set-membership technique. To illustrate this point a comparison with the subpavings approach is included. Finally, by means of the application to the wind turbine benchmark problem, it is shown how the Bayesian fault detection test works successfully.

Fault detection and isolation for a wind turbine benchmark using a mixed Bayesian/set-membership approach

This paper addresses the problem of fault detection and isolation of wind turbines using a mixed Bayesian/Set-membership approach. Modeling errors are assumed to be unknown but bounded, following the set-membership approach. On the other hand, measurement noise is also assumed to be bounded, but following a statistical distribution inside the bounds. To avoid false alarms, the fault detection problem is formulated in a set-membership context. Regarding fault isolation, a new fault isolation scheme that is inspired on the Bayesian fault isolation framework is developed. Faults are isolated by matching the fault detection test results, enhanced by a complementary consistency index that measures the certainty of not being in a fault situation, with the structural information about the faults stored in the theoretical fault signature matrix. The main difference with respect to the classical Bayesian approach is that only models of fault-free behavior are used. Finally, the proposed FDI method is assessed against the wind turbine FDI benchmark proposed in the literature, where a set of realistic fault scenarios in wind turbines are proposed.

Monitoring and remote control of energy consumption by WiFi networks

In this paper we present a modular system for remote monitoring and control of energy consumption. In particular, we demonstrate the application of the proposed system for electric energy monitoring and control in domestic or medium size office installations. This whole system is integrated in a single platform and consists of different modules: one or several smart plugs including a wireless communication interface for connection to a WiFi network, and a centralized application server. The developed system allows a real time monitoring of the energy consumption, and the control and scheduling of remote switches, providing the tools for an optimal energy saving. This monitoring and control leads to a significant cost reduction and to a sustainable management of energy resources beneficial for the environment. The consumption results are also presented for a typical scenario with five power outlets.

Sensor placement for classifier-based leak localization in water distribution networks using hybrid feature selection

This paper presents a sensor placement approach for classifier-based leak localization in water distribution networks. The proposed method is based on a hybrid feature selection algorithm that combines the use of a filter based on relevancy and redundancy with a wrapper based on genetic algorithms. This algorithm is applied to data generated by hydraulic simulation of the considered water distribution network and it determines the optimal location of a prespecified number of pressure sensors to be used by a leak localization method based on pressure models and classifiers proposed in previous works by the authors. The method is applied to a small-size simplified network (Hanoi) to better analyze its computational performance and to a medium-size network (Limassol) to demonstrate its applicability to larger real-size networks.

Optimal sensor placement for classifier-based leak localization in drinking water networks

This paper presents a sensor placement method for classifier-based leak localization in Water Distribution Networks. The proposed approach consists in applying a Genetic Algorithm to decide the sensors to be used by a classifier (based on the k-Nearest Neighbor approach). The sensors are placed in an optimal way maximizing the accuracy of the leak localization. The results are illustrated by means of the application to the Hanoi District Metered Area and they are compared to the ones obtained by the Exhaustive Search Algorithm. A comparison with the results of a previous optimal sensor placement method is provided as well.

Sensor Placement for Classifier-Based Leak Localization in Water Distribution Networks

This chapter presents a sensor placement method for the classifier-based approaches for leak localization in water distribution networks introduced in the previous chapter. The proposed approach formulates the sensor placement problem as a binary optimization problem. Because of the complexity of the problem, it is solved by applying genetic algorithms. In order to reduce the number of sensor configurations to test, a binary matrix that identifies pairs of sensors providing similar information is added as a constraint. The sensors are placed in an optimal way, which maximizes the accuracy of the leak localization . The proposed approach is first illustrated by means of the application to an academic example based on the reduced version of the Hanoi water distribution network . A more realistic case study is then proposed based on the Limassol district metered area.

Leak localization in water distribution networks using model-based Bayesian reasoning

This paper presents a new method for leak localization in Water Distribution Networks that uses a model-based approach combined with Bayesian reasoning. Probability density functions in model-based pressure residuals are calibrated off-line for all the possible leak scenarios by using a hydraulic simulator, being leak size uncertainty, demand uncertainty and sensor noise considered. A Bayesian reasoning is applied online to the available residuals to determine the location of leaks present in the Water Distribution Network. A time horizon method combined with the Bayesian reasoning is also proposed to improve the accuracy of the leak localization method. The Hanoi District Metered Area case study is used to illustrate the performance of the proposed approach.

Nonlinear Set-membership Identification using a Bayesian approach*

This paper deals with the problem of set-membership identification of nonlinear-in-the-parameters models. To solve this problem a Bayesian approach is presented. The paper illustrates how the Bayesian approach can be used to approximate the feasible parameter set (FPS) by assuming uniform distributed estimation error and flat model prior probability distributions. The methodology leads to an approximation of the FPS consisting of a set of boxes, where two regions can be identified. The inner region constitutes an inner approximation of the FPS whereas the external region can be viewed as an outer approximation of the FPS. Also, the boxes in the border give information about the percentage of consistent models inside each box and it can be used to iteratively refine the inner and outer approximations.