Search

t New technologies and system communications are being applied in the industry, improving the efficiency and effectiveness. This paper is focused on novel technologies, software and materials that allow to explore deep ocean floor. Autonomous underwater vehicles require planning navigation models and algorithms. Sensors equipped in underwater vehicles allow to inspect and analyse inaccessible areas. Monitor and control measurement process is required to ensure suitable underwater operations. This paper presents a model using the main inspection process variables. The model calculates the field of view of the autonomous underwater vehicle to be determined according to the type of sensor, the orientation and the distance from the floor. This study aims at stabilising the fundaments to develop an autonomous route for the autonomous underwater vehicles and optimize its operation performance.

The wind energy industry is expanding in order to be able to meet the current and future energy demand, and is supported by governments in that renewable energy investment has been made. Optimal decision making (DM) in wind turbine manufacturing is required to guarantee the competitiveness of the business. This paper considers decision making for wind turbine manufacturing using a logical decision tree (LDT) and binary decision diagrams (BDD). A qualitative analysis of wind turbine manufacturing is carried out using logical decision trees. They are used for a qualitative study of the case study. Binary decision diagrams are used to obtain the Boolean function and, therefore, to carry out a quantitative analysis. Finally, an optimization of budgets is employed based on importance measures. There is no optimal method that can establish the importance measures. The following heuristic methods have been used to find a solution close to the optimal: Fussell-Vesely, Birnbaum and Criticality. The computational cost is reduced by ranking the events. The heuristic methods to establish the best rankings are: Top-Down-Left-Right, Level based method, AND based method, Breadth-First Search (BFS) and Depth First Search (DFS). A real case study is considered, in which a static and dynamic analysis is carried out.

Las políticas de Unión Europea en energía y medio ambiente están dirigidas a impulsar y desarrollar plataformas eólicas offshore. Ello hace que el sistema eléctrico español vaya a depender, cada vez más, de este tipo de sistemas de generación eléctrica. Los aerogeneradores para este caso son de mayor tamaño, más complejos y, requieren de unas altas exigencias de seguridad, fiabilidad, disponibilidad y mantenibilidad. En este proyecto se aborda dicha problemática planteando como fin último la gestión integral y óptima de este tipo de parques de aerogeneradores. El proyecto que da origen a este artículo parte con la intención de continuar y completar el proyecto Nacional WindSeaEnergy (DPI2012-31579), donde se analizaron las referencias científicas más importantes en revistas de alto impacto, observándose que existen grandes carencias de modelos matemáticos que permitan analizar las señales que se están monitorizando mediante Ensayos No Destructivos (END) para determinar el estado de las estructuras, así como en la gestión óptima de los aerogeneradores y los parques eólicos. OptiWindSeaPower ha continuado y completado este estudio iniciado en el ámbito de los elementos rotativos del aerogenerador y de la gestión del mantenimiento de los mismos, haciendo un estudio más exhaustivo en los sistemas de monitorización y métodos de procesamientos de señales para los elementos estructurales de dichos equipos. Se han tomado como referencia datos de los proyectos nacionales IcingBlades y WindSeaEnergy, además de los proyectos Europeos OPTIMUS y NIMO. Se ha diseñado y desarrollado un banco de ensayos de un sistema de monitorización basado en END, en ultrasonidos, para determinar el estado estructural de los aerogeneradores para completar dicho conjunto de señales. Se ha elaborado un modelo del coste de ciclo de vida para el sistema predictivo de mantenimiento. Así como se propone emplear modelos matemáticos basados en el análisis en el tiempo, la frecuencia y tiempo/frecuencia, así como Transformadas Wavellets, Redes Neuronales/Inteligencia Artificial, métodos basados en la extracción de características de la señal y derivados de la función de transferencia del sistema. El análisis multivariable se ha realizado mediante Árboles de Decisión Lógicos, que se analizarán empleando Diagramas de Decisión Binarios, y las medidas de importancia creadas mediante métodos heurísticos. Esto permite controlar y optimizar el estado de un aerogenerador de forma íntegra. Finalmente se han creado nuevos índices de significancia basados en costes y se formula el problema de optimización y su resolución mediante métodos metaheurísticos para determinar la política óptima de inversiones en la gestión de este tipo de parques. MINECO, Ministerio de Economía y Competitividad, Proyectos de I+D, del Programa Estatal de Fomento de la Investigación Científica y Técnica de Excelencia, Subprograma Estatal de Generación del Conocimiento (proyecto OptiWindSeaPower, ref.: DPI2015-67264-P). No data 2019 UEM

The constant increasing of renewable energy demand is leading wind turbines to become very complex and sophisticated devices. These technological developments imply new methods and tools to ensure the reliability of the systems. For this purpose, non-destructive testing techniques are widely employed in the field of wind turbine maintenance. This work presents the implementation of a walking robot-based system that allows non-destructive testing to be carried out in difficult access areas of wind turbines. The paper is divided as: a brief literature overview is done to identify those limitations of current procedures that could be overcome by using the proposed tool; a detailed explanation of the novel system is given, where the different components and features of the robot are described; several applications of the proposed systems are also shown. These applications can be classified regarding to the type of sensor and the area to inspect: Acoustic emission, visual inspection, guided wave testing, noise analysis or thermographic inspections are some of the non-destructive testing techniques that can be aided by this tool. Moreover, external and internal surfaces of blades, tower, nacelle and other difficult access areas can be reached by the robot. Finally, some advantages of this system are enhanced with respect to the conventional methodologies. The usefulness of the proposed system is demonstrated in terms of safety and efficiency with respect to other procedures.

Photovoltaic energy is one of the most relevant renewable energy due to its fast growth in the last years. The sizes of the solar panels and the solar power plants are also increasing. Therefore, the modern solar plants require new maintenance techniques that ensure the efficiency and rentability. In this field, condition monitoring systems play an essential role because they can increase the reliability, reduce cost and, therefore, improve the operation of the solar plant. The goal of this paper is to develop a new monitoring system composed of an unmanned aerial vehicle fitted with a thermal camera. This paper studies the capability of the system to detect defects in the solar panels or external factors that affect to the optimal operation. Firstly, the system is tested in an experimental platform under controlled conditions and, then, some experiments in a real solar plant are carried out. The data provided by the system is analyzed and converted into useful information of the state of the panels. It is demonstrated that this system can detect some specific failures and elements that affect the normal condition of the panels.

El objetivo principal del proyecto SCARAB es desarrollar un sistema de monitorización para la operación y el mantenimiento de plantas fotovoltaicas compuesto por los siguientes subsistemas: • Vehículo aéreo no tripulado (UAV o “dron”) adaptado para alojar diferentes sensores y seguir diferentes trayectorias según la tecnología y tipo de instalación del panel fotovoltaico. • Algoritmos que permiten detectar y clasificar las posibles causas de reducción del rendimiento de los paneles como roturas, suciedad o degradación. • Algoritmos que, en base a los sensores embarcados e información SCADA, facilitan indicadores fiables y robustos del rendimiento técnico y económico del sistema. • Aplicación de optimización de las actividades de mantenimiento. Con todo ello, se contribuye a dar respuesta al principal reto que afrontan las plantas solares fotovoltaicas: conseguir una alta rentabilidad que les permita competir con otras fuentes de energía tanto fósiles como renovables.

The complexity of solar power plants is constantly increasing. This sophistication includes the increasing number of solar panels installed and the technologies that are employed in the energy conversion systems. The new solar plants require advanced methods to ensure the availability of all the panels. This paper proposes a recurrent convolutional neural network algorithm for detecting failures, reducing the costs and the time of the inspections. The method is aimed to analyze the data provided by an unmanned aerial vehicle fitted with a thermographic camera. This system provides thermographic data and telemetry. A region-based recurrent convolutional neural network is trained by a previously created dataset. Once the neural network is trained, it is used as a hot spot detector. This detector will have employed the telemetry in order to identify the real panel that can be affected.

Nowadays, wind energy is one of the most important renewable energy sources for covering the current electrical demand. The sophistication and the complexity of the wind energy systems are constantly growing. Therefore, new maintenance management techniques are required for ensure the efficiency and rentability of these systems. This paper is focused on the extraction of information from the noise generated by wind turbines. The main causes of noise generation by wind turbines are analyzed in this paper. The noise information allows some types of failures to be detected incipiently and, consequently, preventive and corrective maintenance tasks can be improved. In this field, the use of unmanned aerial vehicles (UAV) can be very useful and facilitate the execution of the inspections. This paper presents a collection of methods for detecting failures through the processing of wind turbine noise. In addition, the possibility of employing UAVs for implement those methods is studied.

En primer lugar se aborda el concepto BIG DATA ofreciendo algunos algoritmos que permiten el tratamiento de datos masivos procedentes de sistemas de monitorización de aerogeneradores. No obstante, los procedimientos expuestos en la tesis son aplicables a otros sistemas que cumplan una serie de condiciones impuestas. Esto supone una primera fase de tratamiento de datos. Una vez los datos han sido tratados, se proponen dos herramientas (Árboles de fallo y Diagramas de Decisión Binarios) que permiten realizar un análisis cuantitativo y cualitativo de los datos aplicados sobre el propio sistema, es decir, teniendo en cuenta variables endógenas del sistema de donde proceden los datos. Porteriormente, se presentan varios métodos novedosos en los que se incorporan al análisis variables y restricciones exógenas, por ejemplo, de tipo económico, para construir una metodología de ayuda a la toma de decisiones.

Electrical and electronic components are very important subcomponents in modern industrial wind turbines (WT). Complex multi-MW WT are continuously being installed both onshore and offshore, continuously increasing the demand for sophisticated electronic and electrical components. The design and operational environment require to achieve maximum reliability, availability, maintainability and safety while in-service. The most critical electrical and electronic components in industrial WT have been identified and the applicability of appropriate condition monitoring processes simulated. A dynamic fault tree analysis has been carried out in order to determine the reliability of the components considered in this study over time. The fault tree has been built based on qualitatively data from the literature and important European collaborative research projects. In this paper, the binary diagram decision (BDD) method is proposed for solving the fault tree (FT). A set of algorithms for ranking events has been considered in order to reduce the computational costs; “Level”, “Top-down-Left-Right”, “AND”, “Depth First Search” and “Breadth-First Search”. A quantitative analysis approach is presented to find the occurrence probability of the top event, i.e. the generator, electrical and electronic components of the WTs failures. The novelty of the paper is to analyse the FT using the BDD approach in order to obtain an analytical expression that allows identifying the critical components of the converters and electrical components in different conditions, employing importance measures (IM) (Birnbaum, Criticality, Structural and Fussell-Vesely) over time. The approach proposed in this paper also allows to obtain the system probability failure over time and using different time increment to evaluate the system. This is a novelty regarding to the state of the art that increases the accuracy of the quantitative analysis when the system is about to exceed the threshold failure probability. The results provided by the dynamic probability of failure and the IMs is therefore employed to create a scheduled maintenance that contributes to improve the decision making and, consequently, to reduce the associate cost to the maintenance tasks.

Operation and maintenance tasks on the wind turbines have an essen- tial role to ensure the correct condition of the system and to minimize losses and increase the productivity. The condition monitoring systems installed on the main components of the wind turbines provide information about the tasks that should be carried out over the time. A novel statistical methodology for multivariable analysis of big data from wind turbines is presented in this paper. The objective is to analyse the necessary information from the condition monitoring systems installed in wind farms. The novel approach filters the main parameters from the collected signals and uses advanced computational techniques for evaluating the data and giving mean- ing to them. The main advantage of the approach is the possibility of the big data analysis based on the main information available.

For a qualitatively and quantitatively analysis of a complex Decision Mak- ing (DM) process is critical to employ a correct method due to the large number of operations required. This paper presents an approach employing Binary Decision Diagram (BDD) applied to the Logical Decision Tree. LDT allows addressing a Main Problem (MP) by establishing different causes, called Basic Causes (BC) and their interrelations. The cases that have a large number of BCs generate important computational costs because it is a NP-hard type problem.. This paper presents a new approach in order to analyze big LDT. A new approach to reduce the complex- ity of the problem is hereby presented. It makes use of data derived from simpler problems that requires less computational costs for obtaining a good solution. An exact solution is not provided by this method but the approximations achieved have a low deviation from the exact.

Nowadays logistical and operational systems are being more complex. A quantitative and qualitative analysis in Decision Making (DM) is presented in this paper. Decision Trees (DT) and Binary Decision Diagrams (BDD) are used to find the best solution to the main problem. The BDD method is used to perform the quantification. A real case related to the timeliness on the deliveries is studied in this paper. Importance Measures(IMs) have been considered to rank the basic events of the DT with respect to their contribution to the top event.Thereby, an improvement of the response of a company facing certain problems and the optimization of the company resources is done. Statistical data is used to obtain an approximate measure of the occurrence probability of the events involved.