Revistas
Revista:
APPLIED SOFT COMPUTING
ISSN:
1568-4946
Año:
2021
Vol.:
108
Págs.:
107465
In the automotive industry, despite the robotic systems on the production lines, factories continue employing workers in several custom tasks getting for semi-automatic assembly operations. Specifically, the assembly of electrical harnesses of engines comprises a set of connections between electrical components. Despite the task is easy to perform, employees tend not to notice that a few components are not being connected properly due to physical fatigue provoked by repetitive tasks. This yields a low quality of the assembly production line and possible hazards. In this work, we propose a sound detection system based on machine/deep learning (ML/DL) approaches to identify click sounds produced when electrical harnesses are connected. The purpose of this system is to count the number of connections properly made and to feedback to the employees. We collect and release a public dataset of 25,000 click sounds of 25 ms length at 22 kHz during three months of assembly operations in an automotive production line located in Mexico. Then, we design an ML/DL-based methodology for click sound detection of assembled harnesses under real conditions of a noisy environment (noise level ranging from ¿16.67 dB to ¿12.87 dB) including other machinery sounds. Our best ML/DL model (i.e., a combination between five acoustic features and an optimized convolutional neural network) is able to detect click sounds in a real assembly production line with an accuracy of 94.55±0.83 %.
Revista:
IEEE LATIN AMERICA TRANSACTIONS
ISSN:
1548-0992
Año:
2021
Vol.:
19
N°:
6
Págs.:
1002 - 1009
COVID-19 healthcare professionals recommend the general population staying at home and remote contacting the authorities, e.g. via SMS messages, if they show symptoms like cough, body pain, fever, and breathing difficulties. Although this approach considers the patient self-report, it is not supported by physiological data, i.e. medical personnel does not have a remote mechanism to validate such symptoms. This paper proposes a system, called Rinku, to address the abovementioned scenario. Rinku integrates an electronic system (ClinicalKit) comprising biomedical sensors for body temperature, pulse rate, and oxygen saturation, as well as a digital platform for storing and displaying the collected data. Rinku system aims to provide health professionals with relevant information to remotely validate COVID-19 symptoms. Rinku can handle simultaneous information from multiple patients and provide valuable data related to the severity of the reported symptoms, which in turn could help healthcare professionals to make management decisions to optimize their clinical resources. In this paper, the functionality of the ClinicalKit, communication between the IoT architecture and the cloud, and the monitoring of physiological parameters were tested. The results showed that the enclosure design is convenient, IoT architecture is functional and the tracking of temperature, heart rate, and blood oxygen levels from subjects is promising. We consider that the Rinku system has the potential to provide an accurate forecast regarding the demand for clinical resources and take prompt actions related to this pandemic.
Revista:
SOLAR ENERGY
ISSN:
0038-092X
Año:
2020
Vol.:
209
Págs.:
11 - 20
In this paper, the soiling impact on photovoltaic systems in Aguascalientes, in central Mexico, an area where 1.4GWp of new photovoltaic capacity is being installed, is characterised experimentally. A soiling rate of ¿0.16%/day in the dry season for optimally tilted crystalline silicon modules, and a stabilization of the soiling losses at 11.2% after 70 days of exposure were observed. With these data, a first of its kind novel method for determining optimum cleaning schedules is proposed based on minimising the levelised cost of energy. The method has the advantages compared to other existing methods of considering the system investment cost in the determination of the optimum cleaning schedule. Also, it does not depend on economic revenue data, which are often subject to uncertainty. The results show that residential and commercial systems should be cleaned once per year in Aguascalientes. On the other hand, cleaning intervals from 12 to 31 days in the dry season were estimated for utility-scale systems, due to the dramatic decrease of cleaning costs per unit photovoltaic capacity. We also present a comparative analysis of the existing criteria for optimising cleaning schedules applied to the same case study. The different methods give similar cleaning intervals for utility-scale systems and, thus, the choice of a suitable method depends on the availability of information.
Revista:
COMPUTERS IN BIOLOGY AND MEDICINE
ISSN:
0010-4825
Año:
2019
Vol.:
115
N°:
103520
The automatic recognition of human falls is currently an important topic of research for the computer vision and artificial intelligence communities. In image analysis, it is common to use a vision-based approach for fall detection and classification systems due to the recent exponential increase in the use of cameras. Moreover, deep learning techniques have revolutionized vision-based approaches. These techniques are considered robust and reliable solutions for detection and classification problems, mostly using convolutional neural networks (CNNs). Recently, our research group released a public multimodal dataset for fall detection called the UP-Fall Detection dataset, and studies on modality approaches for fall detection and classification are required. Focusing only on a vision-based approach, in this paper, we present a fall detection system based on a 2D CNN inference method and multiple cameras. This approach analyzes images in fixed time windows and extracts features using an optical flow method that obtains information on the relative motion between two consecutive images. We tested this approach on our public dataset, and the results showed that our proposed multi-vision-based approach detects human falls and achieves an accuracy of 95.64% compared to state-of-the-art methods with a simple CNN network architecture.
Autores:
Rodríguez, J.L.; Velázquez, R.; Del-Valle-soto, C; et al.
Revista:
ELECTRONICS
ISSN:
2079-9292
Año:
2019
Vol.:
8
N°:
3
Págs.:
355
Real-time haptic interactions occur under two exploration modes: active and passive. In this paper, we present a series of experiments that evaluate the main perceptual characteristics of both exploration modes. In particular, we focus on haptic shape recognition as it represents a fundamental task in many applications using haptic environments. The results of four experiments conducted with a group of 10 voluntary subjects show that the differences in motor activity between active and passive haptics ease the perception of surfaces for the first case and the perception of pathways for the latter. In addition, the guidance nature of passive haptics makes the pathway direction easy to recognize. This work shows that this last observation could find application in more challenging tasks such as navigation in space.
Revista:
MEASUREMENT
ISSN:
0263-2241
Año:
2019
Vol.:
135
Págs.:
170 - 179
The prediction and understanding of environmental conditions are of great importance to prevent and analyze changes in environment, supporting meteorological based sectors, such as agriculture or smart cities. In that sense, this paper presents an Internet of Things (IoT) system for predicting climate conditions inside enclosures, i.e. temperature, using artificial intelligence by means of a supervised learning method, the artificial hydrocarbon networks model. It allows predicting the temperature of remote locations using information from a web service comparing it with field temperature sensors. Experimental results of the supervised learning model are presented in two modes: offline training to detect the suitable parameters of the model and testing to validate the model with new data retrieval from the web service. Experimental results over ten days of data conclude that artificial hydrocarbon networks model helps to predict remote temperatures with root-mean square error of 2.7 °C in testing mode.
Revista:
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
ISSN:
0018-9545
Año:
2019
Vol.:
68
N°:
4
Págs.:
3294 - 3305
Electric motorcycles use a battery storage system and a controlled inverter (motor drive) to drive the wheels. Given that the sale of electric motorcycles is expected to grow significantly over the next decade, the aim of this research is to add a supercapacitor storage system to this simple network in order to extend the autonomy of the motorcycle. This paper analyzes and demonstrates the benefits of two different structures for adding a supercapacitor bank to a lithium-ion battery pack: a passive filter and a buffer for the battery. The advantages of each structure are presented and design guidelines are provided. In addition, we propose a synthetic inductor filter typology based on the two analyzed structures in order to increase the autonomy of the motorcycle. Simulations and experimental results using real data validate the analysis. This paper then uses real data extracted from an electric motorcycle to compare the four topologies: batteries only, the passive filter, the buffer, and the synthetic inductor semi-active hybrid storage system. The comparison shows that battery capacity is extended up to 7.8% with a smoother current profile. However, due to the losses from the added converter in cascade, the net capacity of the whole system is extended only up to 3.7%. Therefore, improvements in the dc/dc converter can further increase the entire system capacity.
Revista:
SOLAR ENERGY
ISSN:
0038-092X
Año:
2017
Vol.:
157
Págs.:
244 - 250
Sun trackers allow the energy yield of photovoltaic systems to be increased. Among the different types of sun tracking strategies, North-South horizontal single-axis sun trackers are of big interest nowadays, especially for their implementation in utility-scale photovoltaic plants. It is important to find concepts which simplify as much as possible these sun trackers. In this paper, an energetic analysis of two very simple tracking strategies, based on a 2-position or 3-position rotatory movement for the North-South horizontal single-axis tracker, is carried out for latitudes up to 50°. The tracker positions (morning/afternoon for 2-positions, or morning/noon/afternoon for 3-positions) change at pre-defined daily schedules and the tilt angle at the morning and afternoon periods is fixed. However, both the daily schedules and the tilt angles can be adjusted in the installation process, to maximize the annual irradiation capture in a given location. Such simple tracking concepts can be implemented without the need of illumination sensors, complex sun position calculations or continuous control of the motion. The energetic analysis was carried out over a set of 22 locations worldwide and the captured irradiation was compared to that of equator-pointed optimally tilted fixed systems and to that of the best energetically efficient North-South horizontal single-axis tracker, i.e. the continuous sun tracker. Results indicate that, depending on the analysed location, the 2-position tracker can provide between 41% and 74%, and the 3-position tracker can provide between 68% and 87% of the annual irradiation gains that the North-South horizontal single-axis continuous sun tracker achieves with respect to an optimally tilted fixed system. Therefore, these types of simplified tracking concepts, especially the 3-position tracker, can be seen as alternatives to existing North-South horizontal single-axis continuous sun trackers, providing significant energy gains with respect to fixed photovoltaic systems.
Revista:
ENERGY
ISSN:
0360-5442
Año:
2015
Vol.:
89
Págs.:
768 - 777
In this paper, a methodology for characterizing the I¿V curve of partially shaded high concentrator photovoltaic modules is presented. It takes into account atmospheric parameters (direct normal irradiance, ambient temperature, wind speed and air mass) as well as the amount of shading on each receiver lens and the module electrical configuration. Two high concentrator photovoltaic modules have been characterized in order to apply the methodology. Both modules have been measured under different shading conditions and a comparison between experimental and simulated behavior has been carried out. It was found root mean squared errors for the I¿V curves lower than 4% and absolute values of relative errors at the maximum power point lower than 3% in 80% of the analyzed measurements. The methodology can be regarded as a step for the development of a complete model for the electrical characterization of shaded high concentrator photovoltaic power plants.
Revista:
IEEE TRANSACTIONS ON DIELECTRICS AND ELECTRICAL INSULATION
ISSN:
1070-9878
Año:
2012
Vol.:
19
N°:
5
Págs.:
1774 - 1781
This paper presents an analysis to determine the effect of the stress enhancement factor due to the presence of protrusions in the semiconductor shields of HVDC cables. The theoretical analysis of the electric field containing protrusions is based on a harmonic solution for the electric field for spherical and spheroidal protrusions. HVDC analysis takes into account the additional space charge accumulation due to the presence of protrusions.
Capítulos de libros
Libro:
Further Advances in Internet of Things in Biomedical and Cyber Physical Systems
Editorial:
Ed. Springer
Año:
2021
Págs.:
227 - 238
The Internet of Things plays role in all arenas. In this manuscript, we have to consider the growing flowers in a garden, vegetable, fruit, and other farming. We are considering the greenhouse which aims to introduce the productions of yields. Of course, the growth of plants, and farms are vital and need of everyone, keeping in view of this manuscript is aimed to discuss and study in line of IoT and agriculture. In this work, we propose a greenhouse automation system based on Arduino for the monitoring of temperature, humidity, and moisture of the soil. Arduino can obtain data on the environmental conditions of the greenhouse from various sensors and transfer the data to the ESP8266 module. Consequently, it's possible to change the state of greenhouse control devices like fans, lamp heater, and water pump in obedience to the necessary conditions of the crops. These parameters are modified by the type of plant to maximize their growth, the Aloe Vera plant was used in this project. For the architecture of the Internet of Things was used Blynk coming from the embedded board and the communication link with the Blynk Server was through the Wi-Fi protocol. Results indicate that the system allows the control and monitoring in real-time of the greenhouse correctly. As a future improvement, it is intended with the data obtained, to search for the best optimal conditions for plant growth through artificial intelligence
