https://hdl.handle.net/20.500.12412/2539 2026-05-06T00:22:41Z 2026-05-06T00:22:41Z Revollo, Henrry Ferrada, Pablo Martin, Pablo Marzo, Aitor Del Campo, Valeria https://hdl.handle.net/20.500.12412/7198 2026-04-20T21:00:22Z 2023-08-17T00:00:00Z

HIT Solar Cell Modeling Using Graphene as a Transparent Conductive Layer Considering the Atacama Desert Solar Spectrum Revollo, Henrry; Ferrada, Pablo; Martin, Pablo; Marzo, Aitor; Del Campo, Valeria The optical and geometrical properties of transparent conductive oxide (TCO) are crucial factors influencing the efficiency of a-Si:H/c-Si heterojunction (HIT) solar cells. Graphene is a potential candidate to be used as TCO due to its optical and electrical properties. Here, the effect of graphene as TCO is numerically analyzed by varying the number of graphene layers from one to ten. First, the optical properties are calculated based on the transmittance data, and then the HJT cell’s performance is simulated under the AM1.5 standard spectrum and the mean Atacama Desert solar spectral irradiance in Chile. In the modeling, the most relevant properties are calculated with the spectrum of the Atacama Desert. The most relevant values were obtained as follows: open circuit voltage Voc = 721.4 mV, short circuit current Jsc = 39.6 mA/cm2, fill factor FF = 76.5%, and energy conversion efficiency Ef f = 21.6%. The maximum power of solar panels irradiated with the Atacama Desert spectrum exceeds the results obtained with the AM1.5 standard spectrum by 10%. When graphene is the transparent conducting oxide, quantum efficiency has a higher value in the ultraviolet range, which shows that it may be convenient to use graphene-based solar cells in places where ultraviolet intensity is high.

2023-08-17T00:00:00Z Aranda Brítez, Diego Tapia Córdoba, Alejandro Millán Gata, Pablo https://hdl.handle.net/20.500.12412/7155 2026-04-08T21:00:13Z 2025-01-01T00:00:00Z

A self-calibration algorithm for soil moisture sensors using deep learning Aranda Brítez, Diego; Tapia Córdoba, Alejandro; Millán Gata, Pablo In the current era of smart agriculture, accurately measuring soil moisture has become crucial for optimising irrigation systems, significantly improving water use efficiency and crop yields. However, existing soil moisture sensor technologies often suffer from accuracy issues, leading to inefficient irrigation practices. The calibration of these sensors is limited by conventional methods that rely on extensive ground reference data, making the process both costly and impractical. This study introduces an innovative self-calibration method for soil moisture sensors using deep learning. The proposed method focuses on a novel strategy requiring only two characteristic points for calibration: saturation and field capacity. Deep learning algorithms enable effective and accurate in-situ self-calibration of sensors. This method was tested using a large dataset of simulated erroneous sensor readings generated with simulation software. The results demonstrate that the method significantly improves soil moisture measurement accuracy, with 84.83% of sensors showing improvement, offering a more agile and cost-effective implementation compared to traditional approaches. This advance represents a significant step towards more efficient and sustainable agriculture, offering farmers a valuable tool for optimal water and crop management, while highlighting the potential of deep learning in solving complex engineering challenges.; Se trata de la verión preprint del artículo. Se puede consultar la versión final en https://doi.org/10.1007/s10489-024-05921-0

2025-01-01T00:00:00Z Aranda Brítez, Diego Tapia Córdoba, Alejandro Johnson, Princy Pacheco Viana, Erid Eulogio Millán Gata, Pablo https://hdl.handle.net/20.500.12412/7147 2026-03-06T22:00:19Z 2024-01-01T00:00:00Z

Improving the Calibration of Low-Cost Sensors Using Data Assimilation Aranda Brítez, Diego; Tapia Córdoba, Alejandro; Johnson, Princy; Pacheco Viana, Erid Eulogio; Millán Gata, Pablo In the context of smart agriculture, accurate soil moisture monitoring is crucial to optimise irrigation, improve water usage efficiency and increase crop yields. Although low-cost capacitive sensors are used to make monitoring affordable, these sensors face accuracy challenges that often result in inefficient irrigation practices. This paper presents a method for calibrating capacitive soil moisture sensors through data assimilation. The method was validated using data collected from a farm in Dos Hermanas, Seville, Spain, which utilises a drip irrigation system. The proposed solution integrates the Hydrus 1D model with particle filter (PF) and the Iterative Ensemble Smoother (IES) to continuously update and refine the model and sensor calibration parameters. The methodology includes the implementation of physical constraints, ensuring that the updated parameters remain within physically plausible ranges. Soil moisture was measured using low-cost SoilWatch 10 capacitive sensors and ThetaProbe ML3 high-precision sensors as a reference. Furthermore, a comparison was carried out between the PF and IES methods. The results demonstrate that the data assimilation approach markedly enhances the precision of sensor readings, aligning them closely with reference measurements and model simulations. The PF method demonstrated superior performance, achieving an 84.8% improvement in accuracy compared to the raw sensor readings. This substantial improvement was measured against high-precision reference sensors, confirming the effectiveness of the PF method in calibrating low-cost capacitive sensors. In contrast, the IES method showed a 68% improvement in accuracy, which, while still considerable, was outperformed by the PF. By effectively mitigating observation noise and sensor biases, this approach proves robust and practical for large-scale implementations in precision agriculture.

2024-01-01T00:00:00Z Martínez, Fátima Belén Romaine, James Brian Johnson, Princy Cardona Ruiz, Adrián Millán Gata, Pablo https://hdl.handle.net/20.500.12412/7146 2026-03-06T22:00:33Z 2025-01-01T00:00:00Z

Novel Fusion Technique for High-Performance Automated Crop Edge Detection in Smart Agriculture Martínez, Fátima Belén; Romaine, James Brian; Johnson, Princy; Cardona Ruiz, Adrián; Millán Gata, Pablo Optimising vegetable production systems is crucial for maintaining and enhancing agricultural productivity, particularly for crops like lettuce. Separating the crop from the background poses a significant challenge when using automated tools. To address this, a novel technique has been developed to automatically detect the vegetative area of lettuces, optimising time and eliminating subjectivity during crop inspections. The proposed deep learning model integrates the YOLOv10 object detector, the K-means classifier, and a segmentation method known as superpixel. This combination enables lettuce area identification using bounding box labels instead of contour labels during training, improving efficiency compared to other methods like YOLOv8 and Detectron2. Additionally, the combination of the YKMS method with YOLOv8 (YKMSV8) is evaluated, where YKMS serves as a label assistant. These methods are also used as benchmarks to compare the proposed approach. For the training of each methods, a custom database has been created using a low-cost, low-power custom IoT node deployed on a real farm to provide the most accurate data. Throughout the comparison, a custom metric is used to evaluate performance both in training and inference, balancing computational cost and area error, making it applicable in agriculture. Performance metric is associated with computational cost factor and accuracy factor whose value are respectively 65% and 35%, ensuring applicability for autonomous agricultural devices. Computational cost is prioritised to maintain battery life during extended campaigns. The results of the custom metric during inference indicated that the YKMSV8 method achieved the highest performance, followed by Detectron2, YOLOv8, and, lastly, YKMS. Regarding area error, YOLOv8 exhibited the lowest mean error, followed by Detectron2, while YKMSV8 and YKMS produced similar values. In terms of inference time, YKMSV8 was the most computationally efficient, followed by YOLOv8, YKMS, and, finally, Detectron2.

2025-01-01T00:00:00Z