This paper deals with a remote state estimation problem for a nonlinear system. In a typical networked control system (NCS) scenario, the estimator and controller are remotely located, and they are connected with the plant through a common communication network. Traditional Bayesian filters assume that the measurements are always available. However, that may not be the case in reality. As the sensor measurements are transmitted to the remotely located estimator through an unreliable communication channel, delay may be introduced. Similarly, the control signal is also applied remotely, and it reaches to the plant through a similar unreliable communication channel, and due to which here also delay may occur. In this paper, the authors develop a generalized framework of nonlinear filtering where the states can be estimated in presence of arbitrary random delay in (i) transmission of measurement from sensor to the estimator and (ii) transmission of input from the remotely located controller to the system. The filtering algorithm in such scenario is realized with deterministic sample points. The performance of the proposed method is tested experimentally on two simulation problems. With the help of the simulation results, it is shown that the developed method performs better than traditional non-delayed nonlinear filters in the presence of arbitrary delay in measurement and input.

This work reports the comparison of two different Electronic Tongues (ETs) approaches for the detection and quantification of the main chemicals responsible of Brett character in wines, namely 4-Ethylphenol (4-EP), 4-Ethylguaiacol (4-EG) and 4-Ethylcatechol (4-EC).

On the one hand, a sensor array based on Molecularly Imprinted Polymers (MIPs) was designed to be individually selective to each of the analytes which give the Brett character, i.e., 4-EP, 4-EG and 4-EC. These polymers were designed and synthesised using each of the analytes, respectively, as template molecules. Once obtained, these materials were characterised and integrated onto the Graphite Epoxy Composites sensors (GECs). Then, the readout was done by Differential Pulse Voltammetry (DPV), optimizing previously the measurement conditions. On the other hand, the voltammetric ET based on a chemically modified sensor array, was formed by 5 modified-GECs and 1 GEC, as bare electrode. The different sensors were modified with Cu nanoparticles, WO₃ nanoparticles, Co phtalocyanine, Bi₂O₃ nanoparticles and polypyrrole. This choice was intended as to maximize the differences in the obtained voltammograms for the different sensors using cyclic voltammetry (CV) as electrochemical technique.

Once the sensor arrays were developed Principal Component Analysis (PCAs) were done in order to discriminate the phenols among other interferent species. Finally, Artificial Neural Networks (ANNs) were used for the quantification of these analytes in aqueous samples in the case of the MIPs-based sensor array and in wine samples in the case of modified sensor array.

The Internet of Things (IoT) is an idea in which each device, building, etc. has a unique identifier and the ability to collect and then transfer data between them via wired or wireless networks. The most developed element of IoT are wearables, use to be located on or in the vicinity of the body or as part of the clothing. The aim of this study was to evaluate the absorption in the user’s head of an electromagnetic field (EMF) emitted by wearable IoT device from Wireless Sensor Network (WSN) for monitoring hazards in the work environment, in order to test the hypothesis that they have insignificant influence on humans. The modeled EMF source was the MIFA (Meandered Inverted-F Antenna) antenna emitting EMF of up to 100 mW at 2.45 GHz of radiofrequency module used in the model of wearable device, developed within reported study, using both Wi-Fi and/or Bluetooth communication technologies. To quantify the EMF absorption, the specific energy absorption rate (SAR) values were calculated in a multi-layer ellipsoidal model of the human head (involving skin, fat, and skull bones layers) while various scenarios of its use (headband or attached to the side of a helmet required e.g. in an industrial environment). The analysis of results of modelling is ongoing, therefore the results and their discussion will be presented in the manuscript.

As the world's population is growing significantly older, there is not enough nursing personnel in many countries for all the elderly people in need of care. To promote their autonomy while also reducing the burden of their care-givers, we propose a health monitoring system comprised of a social robot and various wearable and non-wearable sensors. Through the use of patient-reported outcome measurements (PROMs) in conversation with the social robot, the subjective health status of the user is determined. This is supplemented by the objective information gathered from various sensors, such as wearable devices used to measure numerous biosignals and non-wearable camera-based devices for activity detection and emotion recognition. By combining this subjective data obtained through interaction with the user and the objective data from the sensor network, a health report for both users and care-givers is generated. The data is visualized for the user and care-giver in a customizable and easily accessible IoT (internet of things) hub, which also warns the user and their emergency contacts when the data deviates from the expected values or ranges. The goal is to use this health-related information to avoid or decrease doctor visits and hospital stays as changes in the user’s health status can be determined more quickly. The proposed system establishes a good base for further testing and optimization together with the user, to ensure a useful and appropriate combination of sensors and technological devices that the user is comfortable with.

The advantages of Internet of Things (IoT) technology have led to the development of Real-Time Tracking Systems (RTLS) used to identifying, monitoring and tracking objects within indoor or confined environments (e.g. devices, things or people tracking in factories, warehouses or offices, and even patients, biological materials, pharmaceuticals tracking in medical centers). One of technologies which may be incorporated in RTLS is RadioFrequency IDentification (RFID) system. Health care workers and patients displacement often requires approaching to RFID readers and thus exposure to the electromagnetic field emitted there. The aim of this ongoing study was to evaluate the specific energy absorption rate (SAR) values in the body of person present near ultra-high frequency (UHF) RFID readers operating at frequency range 865 - 868 MHz, considering various exposure scenarios (various number of single antennas fixed to the walls and their different location against human body: readers’ plane or their side located in front of or on the side of human body). The modelled electromagnetic field (EMF) source was rectangular microstrip antenna designed at resonance frequency in free space of 865 MHz. The SAR values were calculated in anatomically based numerical model of adult male. Preliminary results of this ongoing study showed that the SAR values may reach significant levels in some exposure scenarios. The SAR values in the body exposed to EMF 5 cm away from UHF RFID readers need consideration with respect to general public exposure limits, when radiated power exceed 3 W.

This work focuses on the quantification of paracetamol, ascorbic acid and uric acid mixtures using electronic tongue principle. Five optimal electronic tongue sensors array were selected from a set of eight sensors using principal component analysis (PCA) and canonical variate analysis (CVA) in a combination of some clustering metric (F factor) for a given multianalyte resolution application. PCA and CVA allow to visually compare the performance of the different sensors, while the F factor allows to numerically assess the impact that the inclusion/removal of the different sensors does have in the discrimination ability of the ET towards the compounds of interest. The proposed methodology is based on the electrochemical analysis of a pure stock solution of each of the compounds under study, its posterior analysis by PCA/CVA and the stepwise iterative removal of the sensors that demote the clustering when retained as part of the array. Seven different graphite epoxy resin (GEC) electrodes modified with cobalt (II) phthalocyanine (CoPc), polypyrrole (PPy), Prussian blue (PB), oxide nanoparticles of bismuth (Bi₂O₃), titanium (TiO₂), zinc (ZnO) and tin (SnO₂) in addition to a Pt disc electrode, were used as the initial sensors array for the selection of five optimal sensors. After the optimal selection, the quantitative ANN model was built which successfully predicted the concentration of the three pharmaceutical compounds with a normalized root mean square error (NRMSE) of 0.00378and 0.0368 for the training and test subsets, respectively, and coefficient of correlation R² ≥0.971 in the predicted vs. expected concentrations comparison graph.

An electronic tongue is a device composed of a sensor array that takes advantage of the cross sensitivity property of several sensors to perform classification and quantification in liquid substances. In practice, electronic tongues generate a large amount of information that needs to be correctly analyzed, to define which interactions and features are more relevant to distinguish one substance from another. The present research focuses on implementing and validating feature selection methodologies in the liquid classification process of a multifrequency large amplitude pulse voltammetric (MLAPV) electronic tongue. Multi-layer perceptron neural network (MLP NN) and support vector machine (SVM) were used as supervised machine learning classifiers. Different feature selection techniques were used, such as Variance filter, ANOVA F-value, Recursive Feature Elimination and model-based selection. Both 5-fold Cross validation and GridSearchCV were used in order to evaluate the performance of the feature selection methodology by testing various configurations and determining the best one. The methodology was validated in an imbalanced MLAPV electronic tongue dataset of 13 different liquid substances, reaching a 93.85% of classification accuracy.

The development of sensors applied to failure detection systems for power transformers is a critical concern since this device stands out as a strategic component of the electric power system. Amongst the most issues is the presence of partial discharges (PD) in the insulation system of the transformer which can lead the device to total failure. Aiming to prevent unexpected damages, several PD monitoring approaches were developed. One of the most promising is the Acoustic Emission (AE) technique which captures the acoustic signals generated by PDs using piezoelectric sensors. Although many studies have proved the effectiveness of AE, most signal processing approaches are strictly related to the frequency analysis of PD signals, which can hide important information such as the repetition rate of the failure. This article presents a comparison between two types of piezoelectric transducers: the micro fiber composite (MFC) and the lead zirconate titanate (PZT). To ensure the detection of multiple PDs the time-frequency analysis was carried out by Short-time Fourier transform (STFT). Intending to compare the sensibility of the transducers, the AE signals were windowed, and the root mean square (RMS) value was extracted for each part of the signal. Results indicated that spectrogram and RMS analysis have great potential to detect multiple PD activity. Although MFC was 2 times more sensitive to PD detection compared with the PZT sensor, PZT presents a higher frequency response band (0 - 100 kHz) concerning MFC (80 kHz).

Under normal operation, insulation systems of high voltage electrical devices, like power transformers, are constantly subjected to multiple types of stresses (electrical, thermal, mechanical, environmental, etc) which can lead to degradation of the machine insulation. One of the main indicators of the dielectric degradation process is the presence of partial discharges (PD). Although it starts due to operational stresses, PD can cause a progressive insulation deterioration since it is characterized by localized current pulses that emit heat, UV radiation, acoustic and electromagnetic waves. In this sense, acoustic emission (AE) transducers are widely applied in PD detection. The goal is to reduce maintenance costs by predictive actions and avoid total failures. Due to the progressive deterioration, the assessment of the PD evolution is crucial to improve the maintenance planning and ensure the operation of the transformer. Based on this issue this article presents a new wavelet -based analysis to characterize the PD evolution. Three levels of failures were carried out in a transformer and the acoustic signals captured by a lead zirconate titanate piezoelectric transducer were processed by discrete wavelet transform. Experimental results revealed that the energy of the approximation levels increased with the failure evolution. More specifically, levels 4 and 6 presented a linear fit to characterize the phenomena, enhancing the applicability of the proposed approach to transformer monitoring.

Within a model of scientific and technical cooperation between the company Cerro
Matoso S.A. (CMSA) and the National University of Colombia (UNAL), a project was developed to
take advantage of the data obtained from a sensor network of a ferronickel electric arc furnace at
CMSA to improve the structural health monitoring process. Through this sensor network, online
data is obtained on the measurement of temperatures in the refractory lining of the electric
furnace, along with heat fluxes, and the chemical characterization of the minerals in each stage of
the process. These data are stored in a local database, which has several years of historical data
with valuable information for control and analysis tasks. These data reflect the behavior of the
industrial process and can be used in the development of machine learning models to predict the
operation of the electric arc furnace, and thus improve the decision-making process. Currently,
most of the data is analyzed by the experts of the structural control department but, due to the
large amount of data, the development of analytical tools is necessary to support their work. This
paper proposes a data cleaning approach to improve data quality by creating a set of rules and
filters based on both expert judgment and best practices in data quality. A statistical analysis was
also carried out to detect variables with anomalies and outliers, which do not represent a real
operation and belong to anomalous data that should not be considered for modelling. With the
proposed process, the quality of the data was improved and the data that were not useful were
eliminated, in order to consolidate a clean data set for later use in the development of machine
learning models. This work contributes to understanding the data cleansing rules that must be
considered to reflect the real behavior of the electric furnace operation for further analysis and
modeling tasks.