Showing papers in "IEEE Latin America Transactions in 2018"
TL;DR: The stability proof using the well-known Lyapunov methodology, for the proposed artificial neural network trained with an algorithm based on the extended Kalman filter, is included.
Abstract: This paper presents the results of the use of training algorithms for recurrent neural networks based on the extended Kalman filter and its use in electric energy price prediction, for both cases: one-step ahead and n-step ahead In addition, it is included the stability proof using the well-known Lyapunov methodology, for the proposed artificial neural network trained with an algorithm based on the extended Kalman filter Finally, the applicability of the proposed prediction scheme is shown by mean of the one-step ahead and n-step ahead prediction using data from the European power system
83 citations
TL;DR: Both the original (PDDB) and subdivided (XDB) databases are now being made freely available for academic research purposes, thus supporting new studies and contributing to speed up the advances in the area.
Abstract: Over the last few years, considerable effort has been spent by Embrapa in the construction of a plant disease database representative enough for the development of effective methods for automatic plant disease detection and recognition. In October of 2016, this database, called PDDB, had 2326 images of 171 diseases and other disorders affecting 21 plant species. PDDB size, although considerable, is not enough to allow the use of powerful techniques such as deep learning. In order to increase its size, each image was subdivided according to certain criteria, increasing the number of images to 46,513. Both the original (PDDB) and subdivided (XDB) databases are now being made freely available for academic research purposes, thus supporting new studies and contributing to speed up the advances in the area. Both collections are expected to grow continuously in order to expand their reach. PDDB and XDB can be accessed in the link https://www.digipathos-rep.cnptia.embrapa.br/.
63 citations
TL;DR: This work uses Particle Swarm Optimisation (PSO), a social-inspired algorithm, to solve the vehicle routing problem with simultaneous pick-up and delivery and time windows, which allows us to address the problem of removing goods after they have been labelled as obsolete.
Abstract: In logistics, planners' main goal is to reduce operational cost as much as possible. Keeping this in mind, other aspects such as recycling arise as important issues for customers. As a consequence, planners often need to find a balance among all these aspects and operational costs. In this work, the vehicle routing problem with simultaneous pick-up and delivery and time windows is considered. Simultaneous pick-up and delivery (also called reverse logistics) allows us to address the problem of removing goods after they have been labelled as obsolete. To solve this complex combinatorial optimisation problem we use Particle Swarm Optimisation (PSO), a social-inspired algorithm. PSO aims to find a set of paths that minimises the total distance of the paths while serving, simultaneously, customers delivery and pick-up demands. Further, time windows constraints are also considered in this paper, which make the problem harder to solve. Including time windows makes also the problem more realistic, though. Results show that the PSO algorithm can find solutions that are quite competitive w.r.t. previously reported algorithms in the literature. Furthermore, the PSO algorithm solves the problem within an acceptable time.
40 citations
TL;DR: This paper consists in a survey presenting to the reader concepts, architecture, applications, routing, simulators and possible integration among FANet and other technologies as well as challenges to be faced in the FANETs.
Abstract: The growing use of UAVs in the last years, has showed that is possible to use them like communication data network in several areas. Such network formed by UAVs is known in the literature as Flying Ad Hoc Networks (FANET). This kind of network has attracted, recently, attention of the research community, showing a promising future. This paper consists in a survey presenting to the reader concepts, architecture, applications, routing, simulators and possible integration among FANET and other technologies as well as challenges to be faced in the FANETs.
35 citations
TL;DR: Multi-layer perceptrons with two layers were the successful methods, reaching an accuracy rate of 96.20%, proving the possibility of building a computer-aided diagnosis system to improve accuracy of mammogram analysis, contributing to improve prognosis as well.
Abstract: Breast cancer is a worldwide public health problem, with a high rate of incidence and mortality. The most widely used to perform early on possible abnormalities in breast tissue is mammography. In this work we aim to verify and analyze the application of classifiers based on neural networks (multi-layer perceptrons, MLP, and radial bassis functions, RBF), and support vector machines (SVM) with several different kernels, in order to detect the presence of breast lesions and classify them into malignant of benign. We used the IRMA database, composed by 2,796 patch images, which consist of 128x128 pixels region of interest of real mammography images. IRMA database is organized by BI-RADS classification (normal, benign and malignant) and tissue type (dense, extremely dense, adipose, and fibroglandular), generating 12 classes. Each image was represented by texture patterns (Haralick and Zernike moments) extracted from the components of the two levels decomposition by morphological wavelets. Multi-layer perceptrons with two layers were the successful methods, reaching an accuracy rate of 96.20%, proving the possibility of building a computer-aided diagnosis system to improve accuracy of mammogram analysis, contributing to improve prognosis as well.
34 citations
TL;DR: The general aim of this investigative research is to develop a brain-computer based interface for the movement imagination of the left fist, right fist, both fists and both feet in order to control an intelligent wheelchair.
Abstract: The activities pertaining to body control performed by human beings utilize neuromuscular tracts Tasks' performance such as moving the arms or walking demand the planning of the task to be performed People diagnosed with clinical conditions such as Amyotrophic Lateral Sclerosis, Spine lesions or Cerebrovascular Accident, for instance, have their neuromuscular tracts damaged One of the alternatives to bypass that problem is the development of technologies which can partially replace the loss functioning of people with severe motor impairment The imagination of the movement is considered as a cognitive state which corresponds to the mental simulation of a given motor action The general aim of this investigative research is to develop a brain-computer based interface for the movement imagination of the left fist, right fist, both fists and both feet in order to control an intelligent wheelchair The electroencephalography signals were acquired through the database eegmmidb - EEG Motor Movement/Imagery Dataset Electroencephalography signals samples of 106 individuals were utilized in order to validate the computational model The proposed model obtained an efficiency of 74,96% in the correct classification of the events related to movement imagination The developed techniques are promising The model intends to contribute as a complementation of an improvement towards the mobility of people suffering from severe motor impairment
33 citations
TL;DR: The present work aims to analyze several methods of classification of digital images, including infrared (IR) or thermographic images, with the objective of investigating the feasibility of the use of these images as an auxiliary exam for the detection of breast cancer.
Abstract: Recent studies have shown that thermographic examination can be a very promising auxiliary tool in the early detection of breast cancer, a key factor in improving the patient's chances of cure. The present work aims to analyze several methods of classification of digital images, including infrared (IR) or thermographic images. It is intended to evaluate the results obtained with the objective of investigating the feasibility of the use of these images as an auxiliary exam for the detection of breast cancer. Initially IR images were acquired and processed. Then, the extraction of characteristics was performed, based on the appropriate temperature ranges from the thermograms. Thus, the input data were determined for the classification process. Several image classifiers were evaluated. Finally, 93.42% accuracy, 94.73% sensitivity and 92.10% specificity were the rates achieved for the Cancer Class in a binary (Cancer versus Non-Cancer) analysis. In a multiclass analysis (Malignant, Benign, Cyst and Normal), 63.46% accuracy, 80.77% sensitivity and 86.54% specificity were the rates achieved for the Malignant Class.
28 citations
TL;DR: In computer security audits, information security risk assessments (ISR) are performed to computer systems, within it to database management systems (DBMS), often using qualitative methodologies, and a model based on knowledge and fuzzy logic was developed for the evaluation of the ISR in the DBMS.
Abstract: In computer security audits, information security risk assessments (ISR) are performed to computer systems, within it to database management systems (DBMS), often using qualitative methodologies. In these methodologies, the evaluation of the ISR is classified according to its impact in linguistic terms such as: High, Medium or Low, so that ambiguities can be generated in the evaluation result. Security checklists are also used to review the configurations of the DBMS. They have a strong dependence on the presence of the expert auditor in DBMS for this analysis. In order to facilitate the work of the auditors, a model based on knowledge and fuzzy logic was developed for the evaluation of the ISR in the DBMS. In this way, the experience in previous audits of this type is useful and improves the results in the evaluation of the ISR.
28 citations
TL;DR: A systematic review of the literature about research involving sports data mining is presented, as systematic searches were made out in five databases, resulting in 21 articles that answered a question that grounded this article.
Abstract: Data mining technique has attracted attention in the information industry and society as a whole, because of the big amount of data and the imminent need to transform that data into useful information and knowledge Recently conducted studies with successfully demarcated results using this technique, to estimate several parameters in a variety of domains However, the effective use of data in some areas is still developing, as is the case of sports, which has shown moderate growth In this context, the objective of this article is to present a systematic review of the literature about research involving sports data mining As systematic searches were made out in five databases, resulting in 21 articles that answered a question that grounded this article
27 citations
TL;DR: A difference between computer simulation methods for characterizing the distribution of the electric field generated by power transmission lines can characterize the limiting exposure to theElectric field, as is show in this paper.
Abstract: This paper has the purpose of evaluating the application of computer simulation methods for characterizing the distribution of the electric field generated by power transmission lines. Analysis through computer simulation with the charge simulation method and the finite element method are performed for a comparison with measured practical data obtained from a 525 kV power transmission line, located in the Southern area of Brazil. The comparison methodology is quantitative and based on concepts of the literature. The results show a difference between the methods, and this difference can characterize the limiting exposure to the electric field, as is show in this paper.
25 citations
TL;DR: The results show that, as well as providing a high degree of accuracy in the classification, the STORm maintains a high stability in the decision-making process.
Abstract: This work proposes STORM, a solution for decision-making in a residential environment that combines fog computing and computational intelligence In this scenario, STORm is able to collect, treat, disseminate, detect and control information generated from the sensor nodes to the decision- making process With this in mind, STORm is based on the development of an ensemble of classifiers to enhance precision in the decision-making process, as well as on the use of the fog computing paradigm to manage and process the actions in the residence in real-time The idea is to provide computational resources closer to the end-users, processes them locally before transmits them to the cloud When compared with the classical approaches adopted in the literature for classification, the results show that, as well as providing a high degree of accuracy in the classification, the STORm maintains a high stability in the decision-making process
TL;DR: In this article, a simulation of the electrical field distribution on the surface of insulators, considering the ambient various conditions to which such components are subjected, has been performed and the finite element software has been used to evaluate the degree of changes due to such conditions in the distribution of the electric field.
Abstract: In the electrical distribution system, the insulating components may have their characteristics compromised by several reasons such as contamination, cracks caused by vandalism, nest of birds, among others. The influence that each type of variation can cause on its insulation capacity is difficult to determine and must be specifically studied. This article has as objective the evaluation through simulation of the electrical field distribution on the surface of insulators, considering the ambient various conditions to which such components are subjected. The finite element software will be used to evaluate the degree of changes due to such conditions in the distribution of the electric field on the surface of insulators. This evaluation shows that the conductive surface significantly reduces the distance from the electric potential to the ground, which makes this condition favorable for electric discharges and therefore must be controlled.
TL;DR: This work has combined two famous architectures of deep learning, the convolutional neural networks (CNN) for acoustic approach and a recurrent architecture with connectionist temporal classification (CTC) for sequential modeling, in order to decode the frames in a sequence forming a word.
Abstract: Systems using deep neural network (DNN) have shown promising results in automatic speech recognition (ASR), where one of the biggest challenges is the recognition in noisy speech signals. We have combined two famous architectures of deep learning, the convolutional neural networks (CNN) for acoustic approach and a recurrent architecture with connectionist temporal classification (CTC) for sequential modeling, in order to decode the frames in a sequence forming a word. Experimental results show that the proposed architecture achieves improved performance over classical models, such as hidden model Markov (HMM) for labeling in variable time sequences in BioChaves database.
TL;DR: A systematic review of academic and industrial literature regarding architectural patterns and architectural tactics for microservices yields 44 architectural patterns in academic sources and 74 in industrial ones, as well as a few architectural tactics originally proposed to address related problems.
Abstract: Microservices are an emerging trend for development of service-oriented software. This approach proposes to build each application as a collection of small services running on separate process and inter-communicating with lightweight mechanisms. Systematic development of microservices is hampered by the lack of a catalog of emerging recurrent architectural solutions (architectural patterns) and design decisions (architectural tactics). This article describes a systematic review of academic and industrial literature regarding architectural patterns and architectural tactics for microservices. The review yield 44 architectural patterns in academic sources and 74 in industrial ones, as well as a few architectural tactics originally proposed to address related problems. Most architectural patterns and tactics are associated to one of just five quality attributes: scalability, flexibility, testability, performance, and elasticity. Also, most microservices in academic (but not industrial) literature are related to DevOps and IoT. The findings lead to propose a new taxonomy of microservice architectural patterns.
TL;DR: A novel methodology to assess structural vulnerability was proposed and applied in IEEE test system and high voltage transmission networks of 94 buses, by using graph theory to investigate various risk scenarios that can trigger cascading failures is validated.
Abstract: In previous research a novel methodology to assess structural vulnerability was proposed and applied in IEEE test system and high voltage transmission networks of 94 buses, by using graph theory to investigate various risk scenarios that can trigger cascading failures. In this paper we validate the application of this methodology in larger networks by applying a case study on the transmission network 230 and 400 kV of Mexico. The events of cascading failures are simulated through two elimination strategies: by deliberate attacks on critical nodes or by random errors. Iterations are performed by running successive N-1 contingencies on a network that is constantly changing its structure with the elimination of each node. The power flows are not necessary and only the calculation of the graph statistical parameter “geodesic vulnerability” is required. This reduces the computation time and leads to a comparative analysis of structural vulnerability.
TL;DR: The PID controller is developed following an adaptive neuronal technique, and the discrete theory of Lyapunov verifies its stability, and an extended Kalman filter is used in order to filter the signals from the aerial vehicle that are contaminated by measurement noises, and that can affect the quality of the identification.
Abstract: In this paper, we present a novel trajectory tracking algorithm for a four-rotor air vehicle (quadrotor). The PID controller is developed following an adaptive neuronal technique, and the discrete theory of Lyapunov verifies its stability. Also, the neuronal identification of the UAV dynamic model is presented. Besides, an extended Kalman filter is used in order to filter the signals from the aerial vehicle that are contaminated by measurement noises, and that can affect the quality of the identification. Then, the output errors are re-propagated to adjust the PID gains to reduce the control errors. Finally, the experimental results are presented using a four-rotor aerial vehicle (quadrotor), by comparing the presented proposal with a classical fixed-gain PID.
TL;DR: Results shows that the fractional order PID controller with the computed torque control strategy has a better performance and robust stability against the presence of the analyzed external disturbances for tracking tasks.
Abstract: This paper presents the tracking tasks control for a robotic manipulator employing a fractional order PID controller with the computed torque control strategy. The proposed control strategy is contrasted with an integer order PID controller with the computed torque control strategy. Robotic system is a two degree of freedom manipulator simulated using a MSC-ADAMS/MATLAB cosimulation model. The manipulator dynamic model is identified employing the recursive least squares algorithm. Fractional order PID controller is tuned using optimization techniques. Controllers robustness is evaluated against the presence of external disturbances in the joints toques, random noise in the feedback loop and payload variations. Obtained results shows that the fractional order PID controller with the computed torque control strategy has a better performance and robust stability against the presence of the analyzed external disturbances for tracking tasks.
TL;DR: It is shown that the designed GPC controller allows controlling the plant under study with a high accuracy considering different real industrial operation scenarios.
Abstract: In this paper a generalized predictive controller (GPC) for effective control of clinkerization temperature in a cement rotary kiln is designed. By using methods for identification of dynamic systems a mathematical model of the plant under study is obtained, whose validation results showed a high adequacy degree. The design of the GPC controller is performed based on the mathematical model obtained. It is shown that the designed controller allows controlling the plant under study with a high accuracy considering different real industrial operation scenarios. The comparative simulation results of the control system with PID and GPC controllers showed a better performance when the GPC controller is applied.
TL;DR: Based on the well-known Big Five model of personality, a basic linguistic-computational resource is collected and used to build supervised models of personality recognition from Facebook status updates.
Abstract: This paper presents a study on the recognition of personality traits from text in Brazilian Portuguese. Based on the well-known Big Five model of personality, we collected a basic linguistic-computational resource - which can be seen as a parallel corpus of texts and personality inventories - and then use this resource to build supervised models of personality recognition from Facebook status updates.
TL;DR: It was found that, although studies still focus on higher education, in recent years there has been an increasing interest in programming teaching projects for children and teenagers, using gamification and tools such as Scratch.
Abstract: It has been frequent the discussion about the teaching and learning of Programming, from the initial series to the undergraduate courses. It is noticed that many students have difficulty to learn programming by several reasons: methodology, tools, programming languages, lack of programming logic in basic education, motivation, among others. Thus, this carries out a survey of the state of the art of existing and documented approaches in the literature, through a mapping of published works in the last five years (2012 to 2016) in two of Brazil's leading scientific computing platforms (CEIE and RENOTE), whose focus is to present solutions that address methodologies and tools that can be used in the different teaching modalities. As methodology was used the Systematic Review of Literature. As a result, it was found that, although studies still focus on higher education, in recent years there has been an increasing interest in programming teaching projects for children and teenagers, using gamification and tools such as Scratch. The results also demonstrate the growing interest of researchers in the search for approaches that provide better results in this area.
TL;DR: According to the obtained results, the fuzzy controller presented a smaller response time than the classic proportional controller and better controllability of the generated power.
Abstract: Wind energy is a renewable source that has grown in application and technological development. Moreover, the generation of electricity from these systems does not compromise the environment as traditional systems, such as hydraulic and thermal generation of electricity. Wind turbine is the main component of a wind generation system, it is responsible for capturing and converting wind energy into electricity. Due to the nonlinear behaviour of the wind, the development of a precise mathematical model of such systems is a complex and sometimes impractical task. However, Fuzzy Logic Controllers to controller this type of since once they do not need a mathematical model, differently from the classical techniques which needs a transfer functions of the controlled system. Therefore, this work analyses the performance of a Fuzzy controller, by using the Fuzzy Logic ToolboxTM available in MATLAB®, applied to a variable speed wind turbine with Permanent Magnetic Synchronous Generator (PMSG) connected to a power system. Also, the obtained results are compared to those provides for a Classical Proportional Controller. According to the obtained results, the fuzzy controller presented a smaller response time than the classic proportional controller and better controllability of the generated power.
TL;DR: An heuristic adjustment is proposed to improve the performance of the James-Stein State Filter (JSSF) even in the presence of measurement noise and the estimation of Chen and Lorenz attractors with uncertainties in their parameters is considered.
Abstract: In practical applications, the chances of having a mathematical model that accurately describes the dynamics of a complex real system are limited. Besides, the noise included in the measurements increases, even more, the problem of estimating the actual states of the system. It is well-known that the most recurring methods for estimating systems, in a practical way, are: Kalman Filter (KF) for linear case and the Extended Kalman Filter (EKF) for nonlinear case. Unfortunately, such estimation methods do not hold when the mathematical model and the real system do not coincide with each other. On the other hand, the James-Stein State Filter provides a robust approach to estimate linear and nonlinear systems under parametric uncertainties of the mathematical model; but, its performance degrades as the standard deviation of the measurement noise increases. Therefore, in this paper, an heuristic adjustment is proposed to improve the performance of the James-Stein State Filter (JSSF) even in the presence of measurement noise. In order to illustrate the applicability of the approach on nonlinear systems with complex dynamics, the estimation of Chen and Lorenz attractors with uncertainties in their parameters is considered.
TL;DR: A MIMO (Multiple-Input-Multiple-Output) adaptive neural PID (AN-PID) controller that can be applied to a nonlinear dynamics is proposed, and its use is shown in the control of a SCARA robot for two degrees of freedom.
Abstract: In this paper a MIMO (Multiple-Input-Multiple-Output) adaptive neural PID (AN-PID) controller that can be applied to a nonlinear dynamics is proposed, and its use is shown in the control of a SCARA robot for two degrees of freedom. The AN-PID controller, including a neural network of the dynamic perceptron type, is designed. The proposed controller uses a RBF network to identify the model and back propagates the control error to the AN-PID controller, unlike other controllers, that use direct methods to back propagate such error. With these properties, an AN-PID controller corrects the tracking errors due to the uncertainties and variations in the robot arm dynamics. It is robust and with adaptive capacity in order to achieve a suitable control performance. Experimental results on the SCARA robot were obtained to illustrate the effectiveness of the proposed control strategy, including comparison with a classical PID. By using Lyapunov's discrete-time theory, it was demonstrated that the control error is semi-global uniformly ultimate bounded (SGUUB).
TL;DR: This work proposes the development of a computational tool that goal at forecasting the daily generation capacity of electric power in a distributed system based on micro generators that use renewable and seasonal sources.
Abstract: This work proposes the development of a computational tool that goal at forecasting the daily generation capacity of electric power in a distributed system based on micro generators that use renewable and seasonal sources. In the specific case, wind and photovoltaic microgenerators are used, which can be found in smart homes. The forecasting tool is based in Extreme Learning Machine (ELM) which is an artificial neural network model. The parameter selection implemented for ELM is based on the Particle Swarm Optimization (PSO). The forecasting system used the mathematical models of the seasonal micro-generators and a meteorological database of the geographic region where the distributed system is located. The tests performed indicate that the Mean Square Error Root (REQM) of the forecast is 7.3 percent.
TL;DR: A mapping of the main technologies and methodologies for the design of power filters that can be customized for the mitigation of harmonic disturbances and improvement of the power factor is presented.
Abstract: Considering the current scenario of the advance of the use of renewable energy sources, mainly wind and solar photovoltaic, in the Brazilian energy matrix. At the same time, there is concern about the stability of the power system and its respective power quality (PQ), due to this type of generation of non-linear loads, such as current converters, which in turn cause distortion Of the current and voltage waveform. The generation of energy by renewable sources does not participate in the control of the voltage and frequency of the electrical system, and in the occurrence of disturbances that exceed the pre-established limits and that affect the system's PQ, are disconnected until the PQ is restored, to be Reconnected after normal operation returns. In order to standardize and quantify the electromagnetic disturbances of the QEE, as well as ensure the quality of the product and services, standards and procedures were designed to instruct from generators to energy consumers. In Brazil, the Electric Power Distribution Procedure in the National Electric System (PRODIST) - module 8. The disturbances of the PQ in the generating units, mainly wind and solar photovoltaic are caused by the harmonics of current, generated by the converters, causing the distortion in the waveform, which in turn, increases the losses in transformers and transmission lines, reduction of cable and equipment life, flicker. Among the possible techniques to minimize the effects of harmonics, the use of passive, active and hybrid filters stands out. However the applicability of each type of filter depends on each system where it will be applied according to the characteristics of the electrical system. Thus, in order to contribute to the advancement of the topic, it can be observed that the use of power filters has a good functionality in harmonic attenuation as well as economical solution. Therefore, this work presents a mapping of the main technologies and methodologies for the design of power filters that can be customized for the mitigation of harmonic disturbances and improvement of the power factor. Finally, it presents a comparative analysis between the topologies of power filters, followed by an alternative proposal of an automated filter for microgeneration of energy connected in low voltage.
TL;DR: An analysis of power quality using a quality analyzer on a photovoltaic system connected to the local distribution grid, based on the limits established in PRODIST Module 8 and IEEE 519 at the connection point of common coupling.
Abstract: Solar Energy is clean, renewable and applicable in every place with a good solar incidence such as Brazil. The photovoltaic system is generally composed by photocells which are able to produce direct current through solar light incidence and, when connected to the grid, it needs an inverter to convert direct current in alternate current. This equipment can bring negative impact in the power quality provided from the main grid, such as voltage oscillation and harmonic distortion. In this context, this article presents an analysis of power quality using a quality analyzer on a photovoltaic system connected to the local distribution grid. All measurements were realized in a residential 3.38 kWp-photovoltaic system with an inverter. Harmonic spectrum were measured and analyzed based on the limits established in PRODIST Module 8 and IEEE 519 at the connection point of common coupling (PCC).
TL;DR: An integrated methodology for planning and operation of distribution systems in an environment of smart grids using a specialized genetic algorithm for location of devices, a particles swarm optimization algorithm for dimensioning of elements and a non-dominated sorting genetic algorithm to solve the multi-objective problem associated with the location of reclosers.
Abstract: This paper presents an integrated methodology for planning and operation of distribution systems in an environment of smart grids The decision variables considered are location and sizing of technologies such as: distributed generation (wind, solar and small scale hydroelectric), energy storage, protection elements for fault isolation and automatic reclosers for load transfer The integration of these technologies enable a higher automation of the distribution system the which in turn brings advantages such as reduction in the energy and active power losses and improvement in the voltage profile and service quality The methodology consists of three stages: i) a specialized genetic algorithm for location of devices, ii) a particles swarm optimization algorithm for dimensioning of elements and iii) a non-dominated sorting genetic algorithm to solve the multi-objective problem associated with the location of reclosers The tests were performed on a system of 102 nodes to verify the performance of the proposed methodology
TL;DR: The HHT approach and its new methodologies for improvement of the analysis, such as the masking process are exposed and a mode mixing separation technique is presented.
Abstract: Time and frequency localizations are of crucial importance in the analysis of nonlinear and non-stationary processes, especially in systems with high level of complexity where detection of information/events, estimation of parameters and classification of signals in classes is necessary to take decisions. The Hilbert Huang Transform (HHT) offers an adaptive approach to analyze no-linear and non-stationary processes. This paper exposes the HHT approach and its new methodologies for improvement of the analysis, such as the masking process. Two examples are given to show the techniques, first a synthetic signal, representing a typical behavior of an electrical signal immersed in a power electronic environment and second a brain signal to extend the acknowledgment to a biological process. Finally a mode mixing separation technique is presented.
TL;DR: In this work the trajectory tracking control is solved for the “DC/DC Boost converter‑inverter‑DC motor” and the experimental results show the good performance of the designed control.
Abstract: In this work the trajectory tracking control is solved for the “DC/DC Boost converter‑inverter‑DC motor”. In the control design, the exact tracking error dynamics passive output feedback (ETEDPOF) methodology is used. So, a control that does not require electromechanical sensors for its implementation is yield. The generation of the reference trajectories, required by the control based on the ETEDPOF, is achieved via differential flatness. Finally, the experimental validation of the control is performed in a built system, along with the use of Matlab-Simulink and a DS1104 board. The obtained experimental results show the good performance of the designed control.
TL;DR: By the statistical comparison of the prediction accuracies of the neural networks with that of a statistical lineal regression, the three neural networks obtained higher prediction accuracy.
Abstract: Failure rates in online higher education raises as a major problem for the modal instruction, thus, prediction of student performance is proposed as a preventive strategy to diminish student failure. Research in student performance prediction is of a wide variety that complicates the replication of studies in order to take advantage of research results. This study tackles this issue by proposing three types of neural networks for student performance prediction, which are built from standardized variables more easily obtained than those used in most of studies identified. By the statistical comparison of the prediction accuracies of the neural networks with that of a statistical lineal regression, the three neural networks obtained higher prediction accuracy. As conclusion, neural networks are proposed as techniques for an early prediction and identification of students in risk of failure.