Javier Rojas Urbano

This paper presents a Systematic Literature Review (SLR) on the application of Artificial Intelligence (AI) techniques in smart water networks, analysing scientific production from 2015 to 2025. The primary objective is to identify predominant algorithms and existing technological gaps to propose a reference architecture. The methodology adhered to PRISMA guidelines, filtering an initial pool of 607 records down to 15 primary studies of high technical relevance. The results evidence a predominance of Convolutional Neural Networks (CNN) for automated meter reading and Reinforcement Learning algorithms for distribution optimization. However, a significant lack of ethical and regulatory frameworks for autonomous decision-making was detected. Finally, a conceptual “Cognitive Water Intelligence” architecture is proposed, integrating prediction and anomaly detection modules, offering a roadmap for the transition towards autonomous systems.

El secado de cacao constituye influye directamente en la calidad final del grano, estabilidad durante el almacenamiento y valor comercial. Los métodos convencionales de secado presentan dependencia de las condiciones climáticas y los generadores de calor conllevan un importante consumo de energía. La implementación de tecnologías térmicas eficientes basadas en fuentes renovables representa una alternativa de gran interés para el sector agroindustrial como el uso de concentradores solares para el intercambio de calor. La potencia máxima de un concentrador solar con geometría parabólica se obtiene al alcanzar la temperatura más alta en su punto focal; para ello es necesario que, en cualquier momento del día, el concentrador capte la mayor cantidad de energía solar, por lo que es habitual orientarlo; de hecho, la cara cóncava del colector siempre está dirigida hacia el sol. En presencia de obstáculos físicos o meteorológicos, es posible que no se pueda seguir la posición del sol; sin embargo, se puede orientar de tal forma que se obtenga la temperatura focal máxima. En este caso, es posible aprovechar la estrategia MPPT de los paneles solares para localizar este punto de máxima potencia, independientemente de las condiciones meteorológicas o de la hora del día. En este trabajo, el sistema de control MPPT se adapta a un concentrador parabólico con capacidad de rotación sobre un solo eje, creando un control de seguimiento del punto de máxima temperatura (MTPT), en el que se utiliza un algoritmo de perturbación y observación en función de la temperatura del punto focal; la rotación del concentrador solar se controla mediante un motor eléctrico hasta obtener la orientación de temperatura máxima; en este punto, se puede alcanzar el máximo rendimiento del sistema con la ayuda del MPPT, obteniendo datos de temperatura y posicionamiento con variaciones óptimas en función del tiempo. Los resultados del sistema muestran que el punto máximo no siempre se encuentra en la dirección del sol, por lo que es necesario un ajuste para mejorar la eficiencia del sistema a lo largo del día; el algoritmo puede alcanzar el punto máximo automáticamente con un error del 1,80 %, de modo que el sistema puede localizar fácilmente los puntos de máxima temperatura independientemente de las condiciones climáticas que se den durante el funcionamiento del concentrador parabólico.

En este artículo se desarrolla un prototipo de cultivo hidropónico NFT con un sistema de control de inyección de nutrientes que contiene la solución acuosa que alimenta un cultivo hidropónico NFT de lechugas como una alternativa de producción para la creciente demanda de alimentos, sustituyendo el uso de suelo por una solución acuosa para alimentar al cultivo. Se utiliza un controlador de entrada mediante un sistema de visión artificial potenciado por una red neuronal entrenada para la identificación de la etapa de crecimiento de las plantas de todo el lote de producción, de manera que establece el nivel adecuado de nutrientes de la solución acuosa, este valor hace de referencia para un sistema de control en lazo cerrado que regula los actuadores del sistema de inyección hasta alcanzar el nivel adecuado de ph, mediante el cual se miden los nutrientes contenidos en la solución acuosa. Los resultados obtenidos permiten una regulación precisa del nivel de pH del agua con variaciones mínimas en un rango de 5,5 - 6.5 en todas las etapas de crecimiento de las lechugas, además de un 93.33\% de precisión en la detección del tiempo de maduración y cosecha. Esto permite disponer de un sistema de cultivo sustentable, que optimiza el uso de nutrientes y suelo mediante la recirculación y la estructura de cultivo vertical, así como se reduce el consumo energético al activar el sistema cuando es necesario una regulación.

This paper evaluates the open- and closed-loop DC–DC converter operation within a DC coupling multilevel inverter architecture to obtain an infinite-level stepped sinusoidal voltage. Adding a cascade controller to the DC–DC converter should reduce the settling time and increase the number of levels in the output voltage waveform; it could decrease the speed error and phase shift concerning the sinusoidal reference signal. The proposed methodology consists of implementing an experimental multilevel inverter with DC coupling through a single-phase bridge inverter energized from a BUCK converter. Trigger signals for the two converters are obtained from a control circuit based in an ATMEGA644P microcontroller to explore its capabilities in power electronics applications. A digital controller is also implemented to evaluate the operation of the BUCK converter in open and closed loop and observe its influence in the stepped sinusoidal output voltage. The evaluation is performed to energize a resistive load with common output voltage in multilevel inverters, i.e., 3, 5, 7, 11, and infinity levels. Results show that during the design stage, fast dynamic elements, like the storage capacitor, can be used to obtain a minimum THD because the settling time is sufficiently fast, the speed error remains small, and there is no need for a controller. A digital controller requires processing time, and although in theory it can reduce the settling time to a minimum, the processor introduces latency in the control signals generation, producing the opposite effect. Controller complexity of the digital controller must be considered because it increases processing time and influences the efficiency of the closed-loop operation.

This paper evaluates the open and closed loop DC-DC converter operation within a DC coupling multilevel inverter architecture to obtain an infinite level stepped sinusoidal voltage. Adding a cascade controller to the DC-DC converter should reduce the settling time and increase the number of levels in the output voltage waveform, it could decrease the speed error and phase shift concerning the sinusoidal reference signal. The proposed methodology consists of implementing an experimental multilevel inverter with DC coupling, through a single-phase bridge inverter energized from a BUCK converter, trigger signals for the two converters are obtained from a control circuit based in an ATMEGA644P microcontroller, and a digital controller is also implemented to evaluate the operation of the BUCK converter in open and closed loop and observe its influence in the stepped sinusoidal output voltage. The evaluation is performed to energize a resistive load with common output voltage in multilevel inverters, that is 3, 5, 7, 11, and infinity levels. Results show that during the design stage, fast dynamic elements, like the storage capacitor, can be used to obtain a minimum THD because the settling time is sufficiently fast and the speed error remains small and there isn’t a need for a controller. A digital controller requires processing time and although in theory, it can reduce the settling time to a minimum, the processor introduces latency in the control signals generation, producing the opposite effect. Controller complexity of the digital controller must be considered because it increases processing time and influences the efficiency of the closed-loop operation.

This article introduces an innovative design method for public lighting systems that surpasses the limitations of conventional approaches, which rely on predefined lamp characteristics and spatial arrangements. By employing a linear additive model to solve a multi-criteria decision model, our study proposes an optimal design methodology considering several key aspects, including the distance between lamps, their type, power, and light distribution. The goal is to achieve optimal illumination that enhances visibility on public roads for drivers and pedestrians while simultaneously minimizing glare and installation costs and maximizing energy efficiency. The proposed methodology is implemented through an algorithm developed in MATLAB R2023b, with results validated through simulations in DIALux evo 12.0. This information is used to construct a decision matrix, assessed using the CRITIC method across 180 different scenarios within a specific case study. The findings demonstrate the effectiveness of multi-criteria decision-making as a tool for significantly improving the planning and design of lighting in public illumination systems, allowing for selecting the optimal combination of parameters that ensure the best lighting conditions.

This article introduces an innovative design method for public lighting systems that surpasses the limitations of conventional approaches, which rely on predefined lamp characteristics and spatial arrangement. By employing a multi-criteria decision algorithm, our study proposes an optimal design methodology considering several key aspects, including the distance between lamps, their type, power, and light distribution. The goal is to achieve optimal illumination that enhances visibility on public roads for drivers and pedestrians while simultaneously minimizing glare, installation costs and maximizing energy efficiency. The proposed methodology is implemented through an algorithm developed in Matlab, with results validated through simulations in Dialux. This information is used to construct a decision matrix, assessed using the CRITIC method across 180 different scenarios within a specific case study. The findings demonstrate the effectiveness of multi-criteria decision-making as a tool for significantly improving the planning and design of lighting in public illumination systems, allowing for selecting the optimal combination of parameters that ensure the best lighting conditions.