Doctorado en Ingeniería

URI permanente para esta colecciónhttp://hdl.handle.net/11349/17

Examinar

Mostrando 1 - 20 de 45

Modelo Computacional Cognitivo para la predicción de preferencias de usuario, basado en análisis de emociones y preclasificación de contenido,orientado a la visualización de datos
(Universidad Distrital Francisco José de Caldas) Gélvez Garcia, Nancy Yaneth; Gaona García, Paulo Alonso; Gaona García, Paulo Alonso [0000-0002-8758-1412]
The rapid expansion of the digital ecosystem has intensified people’s exposure to massive volumes of content, complicating the timely selection of what is truly relevant. This thesis addresses that challenge through affective personalization: it proposes and validates a cognitive computational model able to anticipate user preferences from emotion analysis measured with electroencephalography (EEG) and integrated with standardized selfreport (SAM). The approach articulates three components: (i) reproducible preclassification of visual stimuli via an MLOps pipeline,(ii) a formalized and time synchronized EEG→SAM transformation heuristic, and (iii) personalized prediction with Gaussian Process Regression (GPR). All of this runs in the EPRA web prototype, backed by APIs and a traceable experimental database. Empirical validation was conducted using a case study of information-rich images with high affective load related to the Colombian armed conflict, presented in a controlled web environment. A total of 15 students (16–25 years) completed 5 sessions each, with 5 stimuli per session (375 observations in total). The timing per stimulus (5s baseline, 30s exposure, 15s SAM) guaranteed deterministic alignment between EEG and self-report. The results show positive concordance between measurement routes: global EEG–SAM correlation of r ≈ 0,53 for valence and r ≈0,36 for arousal, with mean absolute discrepancies MAE≈1,6 (1–9 scale). We observed inter-subject heterogeneity consistent with individualized affective profiles, supporting user-specific GPR models. In addition, there was evidence of early adaptation (improvement across the first sessions), while maintaining end-to end data traceability and reproducibility. The thesis contributions are fourfold: (i) a conceptual integration of neurophysiological (EEG) and subjective (SAM) routes within a preference oriented model; (ii) methodological formalization of the EEG→SAM heuristic, bias diagnostics, and longitudinal evaluation with robust metrics; (iii) technical materialization in EPRA (MLOps, Clean Architecture, component-based frontend, APIs, and data schema); and (iv) empirical evidence with a documented, replicable dataset. Taken together, the model and prototype demonstrate feasibility and utility for personalizing information visualization based on emotional state, and provide a transferable foundation for domains such as visual analytics, education, and neuromarketing.
Arquitectura tecnológica para la implementación de una plataforma de telemedicina interoperable sustentada en el Internet de las cosas y la seguridad en profundidad
(Universidad Distrital Francisco José de Caldas) Morales Valenzuela, Olguer Sebastian; Tarazona Bermúdez, Giovanny Mauricio; Tarazona Bermúdez, Giovanny Mauricio [0000-0001-5012-1466]
This dissertation developed a technological reference framework for the implementation of telehealth platforms based on the seven-layer Internet of Things model and a viable financial scheme, aimed at emerging cities with low technological adoption. The study addressed the structural problems of the Colombian health system, evidenced by the persistent non-compliance with the quality indicators established by Constitutional Court Ruling T-760 of 2008, linking these shortcomings to global limitations typical of emerging countries such as hospital overcrowding, insufficient infrastructure, and low technological acceptance. The research analyzed the fragmentation of health information systems, the absence of interoperable mechanisms, and the lack of service traceability, and identified critical technological variables required to ensure the integrity, availability, and security of biomedical data. Based on this, it designed and implemented an architecture validated through a functional prototype capable of processing millions of records in minutes, packaged as Infrastructure as Code to enable its deployment and scalability. The results allowed the characterization of structural failures in the health system—such as gaps in disability records, clinical safety issues, and financial inconsistencies—demonstrating that the proposed technological architecture is a robust instrument for monitoring compliance with the indicators established in Ruling T-760 of 2008 and for strengthening the transparency, efficiency, and traceability of the Colombian health system.
Modelado de estructuras FSS mediante resonadores sublambda para el diseño de filtros y antenas integrados en guías de onda
(Universidad Distrital Francisco José de Caldas) Diaz Pardo, Ivan Eduardo; Suarez Fajardo, Carlos Arturo; Suarez Fajardo, Carlos Arturo [0000-0002-1460-5831]
This thesis presents the development, modeling, and validation of frequency-selective surfaces (FSS), microwave filters based on metamaterial structures, and waveguide antennas. A theoretical model is proposed for the implementation of filters in waveguides, employing bandwidth control through the use of metallic layers with CSRR and Omega-type cells, as well as leveraging mirror symmetry for coupling and enhanced selectivity. The study analyzes the electric and magnetic polarizability of the metamaterial elements and their impact on the transmission and reflection of waves, integrating these parameters into the design of compact and highly selective devices. Based on the mathematical description of scattering parameters, models are established that allow for the prediction and adjustment of the resonant response of the cells, optimizing both the selectivity and performance of the filters, as well as beam control in antennas. The model validation is carried out through electromagnetic simulations, fabrication, and experimental characterization of filters and antennas in WR340 waveguide. For antennas, arrays of six CELC cells were implemented for beam steering, and for filters, CSRR and Omega cells with mirror symmetry were employed in both single-layer and double-layer configurations. The experimental and numerical results obtained demonstrate the viability of the proposed approach for controlling bandwidth and beam direction in microwave devices. The use of mirror symmetry and the optimization of the cell polarizability enable robust tuning of the frequency response and efficient design of compact filters and antennas. In conclusion, the work presented constitutes an advanced tool for the design and optimization of filters and antennas, providing innovative solutions for the development of microwave devices, with experimental validation and applications in communication systems, radars, and advanced sensors.
Modelado de estructuras FSS mediante resonadores sublambda para el diseño de filtros y antenas integrados en guías de onda
(Universidad Distrital Francisco José de Caldas) Diaz Pardo, Iván Eduardo; Suarez Fajardo, Carlos Arturo; Suarez Fajardo, Carlos Arturo [0000-0002-1460-5831]
This thesis presents the development, modeling, and validation of frequency-selective surfaces (FSS), microwave filters based on metamaterial structures, and waveguide antennas. A theoretical model is proposed for the implementation of filters in waveguides, employing bandwidth control through the use of metallic layers with CSRR and Omega-type cells, as well as leveraging mirror symmetry for coupling and enhanced selectivity. The study analyzes the electric and magnetic polarizability of the metamaterial elements and their impact on the transmission and reflection of waves, integrating these parameters into the design of compact and highly selective devices. Based on the mathematical description of scattering parameters, models are established that allow for the prediction and adjustment of the resonant response of the cells, optimizing both the selectivity and performance of the filters, as well as beam control in antennas. The model validation is carried out through electromagnetic simulations, fabrication, and experimental characterization of filters and antennas in WR340 waveguide. For antennas, arrays of six CELC cells were implemented for beam steering, and for filters, CSRR and Omega cells with mirror symmetry were employed in both single-layer and double-layer configurations. The experimental and numerical results obtained demonstrate the viability of the proposed approach for controlling bandwidth and beam direction in microwave devices. The use of mirror symmetry and the optimization of the cell polarizability enable robust tuning of the frequency response and efficient design of compact filters and antennas. In conclusion, the work presented constitutes an advanced tool for the design and optimization of filters and antennas, providing innovative solutions for the development of microwave devices, with experimental validation and applications in communication systems, radars, and advanced sensors.
Modelado y estrategia de control para obtener la máxima producción de energía eléctrica en un sistema de generación a partir de residuos sólidos orgánicos
(Universidad Distrital Francisco José de Caldas) Harker Sánchez, Orlando; Jaramillo Matta, Adolfo Andrés; Jaramillo Matta,Adolfo Andrés [ 0000-0002-9743-5638 ]; García Barreto, Germán Alberto (Catalogador)
The environmental detriment and the need for alternative sources of energy today and in the future mean that renewable energies are beginning to be a fundamental part of the energy matrix worldwide, therefore, the generation of electrical energy from organic solid waste is positioning itself as one of the main alternatives in the global energy panorama. The development of control strategies that lead to the improvement of the efficiency of energy production from organic solid waste plays an important role in the progress towards the implementation of this technology. This process is developed through two major stages, anaerobic digestion and the generation of energy from biogas. This Thesis is developed with the objective of improving the production rates of electrical energy generated from RSO; It begins with the determination of the Anaerobic Digestion model based on the AM2 model, subsequently the model for generating electrical energy from methane is proposed, the reactor is designed and implemented for validation tests and the conditions required for the maximum energy production. Subsequently, a control strategy for the variable dilution rate is designed and validated, which allows the process to be kept stable through linear control. The designed compensator is validated by comparing the responses of the linearized model with the responses of the nonlinear theoretical model around its operating point, through simulation in MATLAB. Finally, a PID-fuzzy control strategy for the stoichiometric concentration of methane is designed and implemented, which is validated through simulation in MATLAB and through implementation in the components at the laboratory level. Among the most relevant results of this Thesis are: • A method for determining the parameters of the DA AM2 process model, • A method for to calculate an observer for the dilution rate, D. • A method for to design a linear control strategy for the dilution rate to keep the biodigestion process stable, and • A method for to design a strategy that improves the production of electrical energy by controlling the stoichiometric concentration of methane. These results allow improving the efficiency of electrical energy production from organic solid waste and can be implemented in systems that are already operating, Salitre PTAR, as well as in systems that are designed with the solution incorporated.
Desarrollo de un modelo de inteligencia artificial para ayudar en la toma de decisiones en política pública aplicado a la vocación agrícola desde la perspectiva sociodemográfica en el territorio colombiano
(Universidad Distrital Francisco José de Caldas) Sánchez Céspedes, , Juan Manuel; Espitia Cuchango, Helbert Eduardo; Rodríguez Miranda, Juan Pablo; Sánchez Céspedes, Juan Manuel [0000-0001-9101-2936]; Espitia Cuchango, Helbert Eduardo [0000-0002-0742-6069]
The main purpose of this doctoral thesis is to develop an artificial intelligence model as a mechanism for public policymakers to improve decision-making processes and enhance the agricultural vocation in Colombia, considering sociodemographic variables. Therefore, three main objectives were defined: first, the development of a conceptual model of the agricultural vocation; second, the design and implementation of the model; and third, the evaluation of the model. The first step included a conceptual revision of public policies and artificial intelligence, and a state-of-the-art revision. On the one hand, public policies related to the agricultural sector and employment in Colombia between 2006 and 2022 were analyzed; on the other hand, a systematic revision of artificial intelligence applied to research on agricultural policies was also carried out. An evolution was observed in the agricultural policies: the focus was rooted in the productivity of the sector, to later prioritize sustainable development. In addition to agricultural processes, the use of artificial intelligence has reached evaluation and prevision processes for efficient use of natural resources like water and land, aiming to anticipate economic and environmental impacts and promotion of sustainable development. After the conceptual review, which included a systematic revision of relevant scientific publications followed by a qualitative-quantitative analysis, a conceptual model and a state-of-the-art for the agricultural vocation was developed. The development of the mode revealed that the agricultural vocation embodies three essential perspectives for sustainable development: sociodemographic, economic, and environmental. Consequently, the model integrates such perspectives as well as technological and public policy variables to support public agrarian policy processes. During the design, the most significant findings show agrarian public policies must be comprehensive and include healthcare issues, education, security, and infrastructure in rural zones to guarantee efficient and sustainable production. The next step was developing the computational model, which included creating the database. The first step involved generating various indexes using the sociodemographic variables taken from the conceptual model. The sources of information were mainly governmental and an imputation process was carried out to complete the missing data. A correspondence analysis was completed using the database and the conceptual model as references to identify relationships between variables starting from the Colombian data. Thus, a more suitable computational design for the conditions set was established. From this design, more suitable techniques and configurations were implemented for both predictions and phenomenon analysis. The techniques employed included artificial neural networks, support vector machines, and neuro-fuzzy inference systems. One of the main findings taken from the model implementation revealed that public policies intended to promote agricultural employment must be based on promoting secondary school or higher education for the rural population; besides, agricultural employment must be formalized with open-ended contracts and decent wages. During the evaluation of the computational model, it was concluded that artificial neural networks are one of the most efficient techniques for making predictions, followed by neuro-fuzzy inference. Nevertheless, the neuro-fuzzy inference system proved to be the most suitable technique useful for understanding the phenomenon, and thus providing public policy guidelines focused on agricultural vocation.
Metodología para la toma de decisiones de proyectos de electrificación en zonas rurales aisladas, desde un enfoque sistémico y de desarrollo sostenible
Garcia Miranda, Diana Stella; Trujillo Rodriguez, Cesar Leonardo; Santamaria Piedrahita, Francisco; Trujillo Rodriguez, César Leonardo [0000-0002-0985-1472]; Santamaria Piedrahita, Francisco [0000-0002-0391-4508]
The doctoral thesis proposes a methodology for decision-making in rural electrification projects in isolated areas, adopting a systemic approach toward sustainable development. The research focuses on designing a model that integrates technical, economic, social, and environmental factors, addressing rural communities' energy needs through renewable technologies and promoting responsible consumption practices. Using tools such as Systems Thinking, Design Thinking, Behavioral Economics, and Participatory Action Research (PAR), this methodology fosters collaboration and community empowerment at every project stage, from design to implementation and monitoring. Additionally, the study presents analyses of the selection, use, and energy services required to meet the diverse needs of rural households. The developed model evaluates households' energy consumption behavior, classifying needs into three dimensions: Existence, Relationship, and Growth. This approach enables the assessment of individual household demands and the impact of shared consumption on community energy infrastructure, facilitating continuous monitoring of community impact and ensuring equitable and sustainable use of energy resources. The research concludes that rural electrification projects can be more sustainable and efficient if they integrate community management practices and adapt to the specific characteristics of each context. Implementing this methodology expands access to electricity and related services in remote areas and contributes to communities' integral and sustainable development, enhancing their capacity to manage and optimize energy resources autonomously and collaboratively.
Modelo de inferencia difusa evolutivo aplicado al pronóstico no estacionario de humedad interna en invernaderos
(Universidad Distrital Francisco José de Caldas) Vanegas Ayala, Sebastián Camilo; Barón Velandia, Julio
Automated greenhouse agriculture serves as a tool for developing crops with high yield and quality indices, requiring proper management of internal climatic variables through prediction and control models. Due to the existence of measurable but uncontrollable disturbances that characterize the non-stationary behavior of these variables, models that can evolve and adapt to these changes are necessary. This document presents the development of a doctoral research project in which an evolutionary fuzzy inference model is constructed to forecast the non-stationary behavior of internal relative humidity in greenhouses for short-cycle crops, providing automatic adjustment of system parameters over time. The proposed methodology consists of three stages: the first is data analysis using the ETL (Extraction, Transformation, and Loading) process employed in data mining; the second is the definition of the model structure based on a systematic review of compatible structures using the PRISMA methodology; and the third is the development of the model based on the software prototyping paradigm. A software prototype of the model is established through various structures of evolutionary fuzzy inference systems, implementing optimized fuzzy systems with a hybrid Mamdani-type algorithm that handles floating-point values. This facilitates monitoring and control for individuals interested in agricultural activities, yielding results with high levels of interpretability and maximum precision in all proposals, with a mean squared error (MSE) of 1.20E-02.
Análisis de vulnerabilidad y resiliencia en sistemas de energía eléctrica ante la ocurrencia de un evento disruptivo deliberado
(Universidad Distrital Francisco José de Caldas) Mosquera Palacios , Darin Jairo; Rivas Trujillo, Edwin; Rivas Trujillo,Edwin [0000-0003-2372-8056]
Electrical Power Systems (EPS) face service interruptions caused by deliberate disruptive events, such as cyberattacks, which significantly impact the system's vulnerability and resilience. These attacks compromise the system's ability to recover and restore normal operation, reducing its responsiveness and adaptability, leading to greater economic losses and prolonged impact on critical loads. This Doctoral thesis proposes an optimization model based on the interdiction problem, implementing a Genetic Algorithm (GA) that will identify the system's vulnerable points (interdiction vectors) and prioritize mitigation actions, minimizing vulnerability and maximizing resilience. The vulnerability analysis identified the most susceptible elements to attacks, such as lines and generators. The development of the Genetic Algorithm (GA) extracted the Interdiction Matrix (IM) and the Interdiction Vector (IV), which generate higher costs for the system in the event of a disruptive incident. The most effective strategies of the attacker were identified, targeting lines and generators, as well as the response of the Network Operator (NO). Four scenarios were established, including the base scenario, in which Demand Response (DR) mechanisms and Distributed Generation (DG)/Generation Plants (GP) were applied, aiming to improve the system's resilience. In the transmission system scenario, by applying DR and GP, the system managed to increase the load served by 69%, while applying DR and GP separately resulted in increases of 64% and 47%, respectively. In the distribution system, by applying DR and DG, the system managed to increase the load served by 56%, while applying DR and DG separately resulted in increases of 46% and 44%, respectively. Additionally, a topological reconfiguration strategy of the electrical network was implemented, suggesting alternative configurations after a disruptive event to maximize resilience and minimize operational costs and vulnerability. Finally, metrics were established to quantify and qualify the resilience of the NO's mitigation actions, reducing load loss and its costs, validated through case studies with IEEE test networks. Among the contributions, the most important contribution of this Doctoral thesis is the development of an optimization technique based on Genetic Algorithms (GA) that allows identifying the most damaging attack plans and determining the optimal restoration actions for system elements, such as lines and generators.
Prototipado rápido de control no lineal aplicado a convertidores conmutados de corriente continua de orden cuadrático
Acosta Rodríguez , Rafael Antonio; Martínez Sarmiento, Fredy Hernán; Martínez Sarmiento, Fredy Hernán [0000-0002-7258-3909]
This document presents a rapid control prototyping methodology to apply the disciplinary concept in the industrial field. Nonlinear control in quadratic-order DC-DC is used to reduce material and electronic device costs, as well as to enhance power quality in the system's response. The study begins with two types of switched quadratic converters. On one hand, the quadratic Boost type undergoes a dynamic analysis starting from its model and leading to the characterization of its behavior, including the procedure for calculating commercial-type elements, which involves the construction and design of its formal plant for behavioral research. Similarly, the quadratic Reductor type is also characterized based on a model, whose functionality is validated in an experimental physical plant to create a functional model ranging from 380V to 48V up to 500W. This is achieved using the dSPACE CP1103 rapid prototyping tool to implement the development of Model in the Loop (MIL), Software in the Loop (SIL), and Hardware in the Loop (HIL) with real-time simulation via Real-time Simulation (RTS). Furthermore, modern robust control techniques are employed, such as sliding mode control and passivity-based control, which are compared with classical control through a PID (Proportional-Integral-Derivative) arrangement established through optimization algorithms. This is done to propose a general framework for the desired response to parameter stability in the presence of disturbances. Finally, a results analysis is conducted based on performance indices, including response time, signal accuracy, system stability, and resource utilization efficiency.
Modelo para la gestión de una identidad digital descentralizada y autogobernada bajo tecnología de registro distribuido
Pava Díaz, Roberto Albeiro; López Sarmiento, Danilo Alfonso; Niño Vásquez , Luis Fernando; López Sarmiento, Danilo Alfonso [0000-0002-6148-3099]
Digital identity is conceived as a fundamental human right; therefore, it should facilitate an individual’s timely access to goods and services, control over their privacy, and mitigation of impersonation risks. Moreover, it is crucial for the economic, political, and social development of a country and its citizens. Every person should be capable of proving, managing, and preserving their digital identity without access barriers and with the autonomy to manage their personal identity attributes. On the other hand, the Internet has evolved without a digital identity layer, compelling each web service or application to implement an identification scheme, generally based on username-password access credentials. This obliges users to remember and manage multiple passwords and utilize various validation methods. Within this context, digital identity management is one of the significant challenges associated with large-scale digital infrastructures in contemporary society. It is a complex problem, as individuals’ information is currently stored in a dispersed manner, with storage and custody by third parties and authentication attributes associated with public personal or biometric data. Self-Sovereign Identity (SSI) supported by distributed ledger technology provides the necessary digital identity layer in the current web and enables an entity to create, own, and control a verifiable and persistent identity in a user-centric data ecosystem. This research project proposes an SSI model called Aletheia, which provides a verifiable representation of digital identity resistant to manipulation due to the availability of a distributed ledger. This model preserves user privacy and control of personal identification attributes through the im- plementation of decentralized identifiers and verifiable credentials, along with the integration of a decentralized storage system with an additional document encryption method. Aletheia was compared with a set of SSI frameworks by analyzing its adherence to digital identity principles, and due to its nature, it demonstrated notable compliance in persistence, protection, access, portability, and interoperability. Additionally, a proof of concept was implemented to enable the creation of decentralized identities with secure and protected storage of verifiable credentials, fully under the control of the identity rights holder. This proof of concept allows users to manage their credentials and perform verifiable presentations within an environment that ensures privacy preservation. It facilitates a comprehensive analysis of the information flow within the model, encompassing the entire process from the creation of the digital wallet, the request and issuance of a credential, and its subsequent storage, to the presentation of the credential to a verifier, culminating in cryptographic verification by this.
Propuesta de un modelo para la predicción de ruido por tráfico rodado a partir de aprendizaje automático que ofrezca información potencialmente útil en los procesos de gestión ambiental del ruido en entornos urbanos de Bogotá
Acosta Agudelo, Óscar Esneider; Montenegro Marín, Carlos Enrrique
This research proposed a model based on machine learning for predicting road traffic noise for the city of Bogota. The model considers conditions typical of vehicular traffic in the city. The input variables of the model were: vehicle capacity, speed, type of flow and number of lanes. Measurement campaigns involving audio and video recordings were carried out to obtain the input data. The audio recordings allowed the calculation of the noise levels through software processing, since they were taken with a measuring microphone calibrated at a height of 4 meters. On the other hand, the video data were used to count and classify the number of vehicles in four categories: motorcycles, light, medium and heavy vehicles. This process was done using a classifier trained with images of vehicles taken in the field and from free databases. Similarly, a processing algorithm based on an image classifier was used to estimate the speed of the vehicles from the video data. Then, the analysis of the measurements was carried out for some measuring points characterizing the noise emission of vehicle categories, arterial and secondary roads, traffic situations and pavements. Then, through exploratory data analysis, correlations were found, and a regression study was performed between noise levels and predictor variables. To determine the machine learning algorithm to be used, five models were compared, configured with their respective hyperparameters obtained through mesh search. The results showed that the Multilayer Perceptron (MLP) regression had the best fit with MAE = 0,86 dB for the test dataset. Finally, the proposed MLP regressor was compared with classical statistical models for traffic noise prediction. In conclusion, the MLP regressor obtained the best error and fit indicators with respect to statistical models.
Propuesta de modelo de planificación turística apoyado en tecnología para la anticipación de impactos socioculturales en comunidades étnicas
Lara Silva, Marcia Ivonne; Rincón Rojas, Edgar Jacinto; Rodríguez Rojas, Luz Andrea; Rodríguez Rojas, Luz Andrea [0000-0003-0312-1177]
A smart destination requires both planning focused on attracting tourists - to generate economic growth and positioning of the destination - and preserving all the elements that compose it, including the community that inhabits it. In this sense, tourism planning approaches have been developed and implemented within the framework of growth in demand, specialization of supply, changes in travelers' behaviour patterns, economic development of the place, and community participation through its productive units. Transformations and advances have been directed toward market elements. Still, none of them have identified aspects of identity and social valuation, such as the well-being or quality of life of the hosts, especially the ethnic, indigenous, and black communities. It is necessary to understand how the host communities, particularly the ethnic ones, have been affected by the development of the activity, since there has been evidence of negative impacts due to displacement, deculturation, and affectation of their dignity, among others, actions that go against the main objectives of tourism. Having said this, the starting premise of this research project is the construction of a tourism planning model that involves the identification of the possible negative impacts on the inhabitant population as a result of the development of tourism activities. To achieve this, a documentary review of tourism planning models, the negative impacts on communities at the case study level, and international policies framed in tourism, which include a code of ethics, was carried out. Similarly, legislation on the protection of intangible cultural heritage was analyzed as a tool for the protection and conservation of cultural assets that are part of the cosmovision of the different peoples that inhabit tourist destinations. As a result of the review, all those elements that provide information for the analysis of negative impacts on the host community were identified, characterized, and arranged in criteria, dimensions, and valuations. With this purpose, it was possible to complement the tourism supply system and the conformation of a new system within the traditional tourism planning model, the host community system. Finally, a tourism destination approach is provided that focuses on the well-being and quality of life of the host community.
Estimación y localización de fugas en sistemas de distribución de agua a través de algoritmos de optimización combinatoria
(Universidad Distrital Francisco José de Caldas) Ladino Moreno, Edgar Orlando; García Ubaque, César Augusto; García Ubaque, César Augusto [0000-0002-6959-6610]
The K-BiLSTM-MC model developed for the detection and localization of leaks in water supply systems proved to be effective, achieving an accuracy of 93.08% in validation and 87.11% in real-world implementation. The approach integrated advanced techniques, such as Hamiltonian equations and IoT, and was validated in a hydraulic sector in Bogotá, Colombia. The results revealed consistent pressure patterns, the presence of weak chaos in the signals, and precise leak detection, improving prediction and reducing uncertainty in hydraulic systems.
El modelo Integra-Ciudad: la gestión del tráfico vial apoyado por la gestión del conocimiento
Romero Villalobos, Oswaldo Alberto; Rincón Rojas, Edgar Jacinto; Romero Villalobos, Oswaldo Alberto [0000-0001-8308-3128]; Rincón Rojas, Edgar Jacinto [0000-0002-2997-8075]
This document presents a research proposal to enable the construction of a model based on knowledge management that supports road traffic management in a city, taking the city of Bogotá as a case study, to understand the interactions between the city, the citizens and mobility that cause the problem of road congestion and have as a consequence problems of an economic, social and environmental nature. The structure developed provides as a starting point a review of the criteria to identify the problem and a justification is provided. Afterwards, some antecedents around the subject of transit and transport are considered, as well as the implication of technology in the mobility of cities to later propose how knowledge management could contribute to the management of vehicular traffic. Finally, we will use the Internet of Things (IoT) and its interrelation with Smartcities as a general theoretical framework and describe some general knowledge management models where we will frame the central theme of the research with documentary analysis to identify the relevant factors and thus build the indicators of knowledge management activities to create a generalized model.
Propuesta de una metodología para el desarrollo sostenible de las actividades económicas en los territorios rurales de Colombia a partir de modelos de inteligencia
Suárez Roldán, Carolina; Méndez Giraldo, Germán Andrés
Over time, it has been noted that the conditions for the development of rural areas have not been equivalent to the resources allocated to urban and/or intermediate spaces, being one of the reasons that explains the current lag in rural development. That is to say, rural areas face enduring challenges related to limited employment opportunities, low coverage of basic services, poor internet connectivity, limited income, among other aspects. These are developmental conditions that do not favor the rural population. Therefore, there has been an increase in rural population migration towards cities in search of new job opportunities and stability for their families. Therefore, rural population migration to cities in search of new job opportunities and stability for their families has increased in recent years. Consequently, urban areas present an imbalance because these spaces do not have the infrastructure and resources to cover these new demands. Rural territories have a crucial function that consists of supplying food, providing inputs to the secondary sectors of the economy, and especially preserving natural resources that are essential for the survival of living beings. Therefore, the potential that rural territories have in activating the development of nations is recognized, and generating a balance between rural, intermediate and urban areas. Consequently, the priority of contributing to the sustainable development of economic activities in rural territories is recognized. Based on the above, it is proposed to design a methodology that guides decision-making in the sustainable development of economic activities in rural territories of Colombia. This methodology is applied to a case study conducted in a rural territory in the Tolima department. In the methodology, a series of development phases is designed to enable the recognition of functions, variables, relationships, components, and elements that must be considered in decisions regarding which economic activities to undertake in these geographic areas. The main result of the doctoral thesis is the design of a methodology based on intelligence models. This methodology guides decision-making around the selection of the best activities that contribute to the sustainable development of rural territories. Consequently, this methodology contributes to the design of rural development programs that are more efficient and effective for the development of the regions.
Modelo de un sistema de administración de energía autónomo operado desde la nube para optimizar la gestión de un grupo de microrredes
Rosero Bernal, David Gustavo; Diaz Aldana, Nelson Leonardo; Trujillo Rodríguez, Cesar; Diaz Aldana, Nelson Leonardo [0000-0003-0202-0489]; Trujillo Rodríguez, Cesar [0000-0002-0985-1472]
Organizing the generation, storage, and management of electrical energy from the perspective of renewable energies, as well as the parameterization of the energy consumption characteristics of communities with limited access to the interconnected electricity supply, has taken more relevance in recent years due the demands that define the social welfare of this century. Complementary to the demand increase, other factors require the improvement and updating of the utility grid infrastructure and its opening to other technologies that meet the needs of end users. The interest in renewable energy sources, the evolution of energy storage technologies, the continuous research in microgrid management systems, and the massification of technologies and tools available in cloud computing, machine learning, big data, and the internet of things environment motivated the development of this doctoral research. This doctoral research focuses on three tasks linked to the operation of a cluster of microgrids. The first task is the fluctuating integration of heterogeneous energy generation devices and objects whose mobility and distribution characteristics are particular over various geographical areas. The second is the need for real-time operation and extensive information processing and storage capabilities. Finally, the third task considers the conservation factors for critical applications linked to advanced optimization techniques, especially the operational cost and the battery's lifespan. An autonomous and scalable energy management model that follows the hierarchical control structure and bases its operation on cloud computing, the internet of things, machine learning, and big data solves the aforementioned tasks. This research defines the elements considered by the real-time autonomous and scalable energy management system framework in a cluster of microgrids. For this, it is necessary to emulate the behavior of a group of interconnected microgrids and test the framework under real scenarios with the assistance of power-hardware-in-the-loop platforms connected to a cloud server. The server programming must include the implementation of the framework management protocol that exploits the optimization algorithm and state of charge equalization. Also, the framework takes advantage of machine learning and big data tools available in a cloud computing environment, ensuring the scalability of the framework's operation based on the fluctuation of the available resources in a microgrid or extending its operation to a cluster microgrids in a transparent way by the incorporation of IoT sensors or other tools. This doctoral thesis summarizes the framework research results and the published evidence released in one book, two journal papers, two international conferences, and one national conference.
Modelo de gestión de alertas tempranas para la predicción de riesgos en el contexto de Smart City para la Ciudad de Bogotá D.C.
Aguirre Buenaventura, Edgar Alirio; Ferro Escobar, Roberto; Ferro Escobar, Roberto [0000-0002-8978-538X]
Risk management is made up of three elements: the first is risk awareness, the second is risk reduction, and the third is disaster management. This research focused on the component of risk knowledge and on proposing a model for the management of early warnings in the context of Smart City that allows contributing to the prediction and mitigation of risks in the city of Bogotá DC, for which a risk knowledge model was designed and a business architecture based on TOGAF was proposed that allows understanding, from a business scenario, what the objectives, goals and processes of this component are, as well as allowing the identification of the business focuses to which an organization is dedicated. risk awareness, ten experiments were carried out to test different components of the architectures, where information, modeling, artificial intelligence, and hardware components were tested, so the results show the potential of the model not only for events such as floods and mass movement, but they are also applicable in scenarios of climate change and agricultural risks, expanding the context of applications to different types of events.
Modelo algorítmico para alta disponibilidad en transporte de volúmenes crecientes de tráfico variable en redes ópticas
Aguirre Moreno, Diego Fernando; BARON VELANDIA, JULIO
The Internet has shown exponential growth in the last decade, generating that users demand solutions to the requirements instantly. These requirements constitute the main problem related to the performance and characterization of the access and aggregation network infrastructure. Among the promising technologies are optical networks, their characteristics in capacity, quality of service and performance allow to support the traffic generated by future applications and technologies such as high-definition video, 5G networks and ultra-definition transmissions. Today, the implementation of all-optical networks presents several challenges such as: device maturity, optical storage buffers, optical packet switching, lack of effective network administration and management methods, and high costs. These limitations have delayed the development of fully optical networks, promoting research into elastic optical networks (EON), which dynamically adjust their resources according to the requirements of each demand. This research addresses the benefits in the use of elastic optical networks as support in the transport of the increasing volume of traffic, supported with machine learning techniques to face the problems of routing, spectrum allocation, modulation format and core selection. Providing a technological vision for the exploitation of the infrastructure that ensures completely optical switching and routing processes, optimizing resources based on demands.

Examinar

Envíos recientes