RESUMEN
This paper focuses on the use of smart manufacturing in lathe-cutting tool machines, which can experience thermal deformation during long-term processing, leading to displacement errors in the cutting head and damage to the final product. This study uses time-series thermal compensation to develop a predictive system for thermal displacement in machine tools, which is applicable in the industry using edge computing technology. Two experiments were carried out to optimize the temperature prediction models and predict the displacement of five axes at the temperature points. First, an examination is conducted to determine possible variances in time-series data. This analysis is based on the data obtained for the changes in time, speed, torque, and temperature at various locations of the machine tool. Using the viable machine-learning models determined, the study then examines various cutting settings, temperature points, and machine speeds to forecast the future five-axis displacement. Second, to verify the precision of the models created in the initial phase, other time-series models are examined and trained in the subsequent phase, and their effectiveness is compared to the models acquired in the first phase. This work also included training seven models of WNN, LSTNet, TPA-LSTM, XGBoost, BiLSTM, CNN, and GA-LSTM. The study found that the GA-LSTM model outperforms the other three best models of the LSTM, GRU, and XGBoost models with an average precision greater than 90%. Based on the analysis of training time and model precision, the study concluded that a system using LSTM, GRU, and XGBoost should be designed and applied for thermal compensation using edge devices such as the Raspberry Pi.
RESUMEN
Assessing mental workload is imperative for avoiding unintended negative consequences in critical situations such as driving and piloting. To evaluate mental workload, measures of eye movements have been adopted, but unequivocal results remain elusive, especially those related to fixation-related parameters. We aimed to resolve the discrepancy of previous results by differentiating two kinds of mental workload (perceptual load and cognitive load) and manipulated them independently using a modified video game. We found opposite effects of the two kinds of mental workload on fixation-related parameters: shorter fixation durations and more fixations when participants played an episode with high (vs. low) perceptual load, and longer fixation durations and fewer fixations when they played an episode with high (vs. low) cognitive load. Such opposite effects were in line with the load theory and demonstrated that fixation-related parameters can be used to index mental workload at different (perceptual and cognitive) stages of mental processing.
Asunto(s)
Conducción de Automóvil , Movimientos Oculares , Cognición , Humanos , Factores de Tiempo , Carga de TrabajoRESUMEN
Big Data analysis has become a key factor of being innovative and competitive. Along with population growth worldwide and the trend aging of population in developed countries, the rate of the national medical care usage has been increasing. Due to the fact that individual medical data are usually scattered in different institutions and their data formats are varied, to integrate those data that continue increasing is challenging. In order to have scalable load capacity for these data platforms, we must build them in good platform architecture. Some issues must be considered in order to use the cloud computing to quickly integrate big medical data into database for easy analyzing, searching, and filtering big data to obtain valuable information.This work builds a cloud storage system with HBase of Hadoop for storing and analyzing big data of medical records and improves the performance of importing data into database. The data of medical records are stored in HBase database platform for big data analysis. This system performs distributed computing on medical records data processing through Hadoop MapReduce programming, and to provide functions, including keyword search, data filtering, and basic statistics for HBase database. This system uses the Put with the single-threaded method and the CompleteBulkload mechanism to import medical data. From the experimental results, we find that when the file size is less than 300MB, the Put with single-threaded method is used and when the file size is larger than 300MB, the CompleteBulkload mechanism is used to improve the performance of data import into database. This system provides a web interface that allows users to search data, filter out meaningful information through the web, and analyze and convert data in suitable forms that will be helpful for medical staff and institutions.
Asunto(s)
Registros Médicos , Nube Computacional , Programas InformáticosRESUMEN
Indoor air quality monitoring in healthcare environment has become a critical part of hospital management and policy. Manual air sampling and analysis are cost-inhibitive and do not provide real-time air quality data and response measures. In this month-long study over 14 sampling locations in a public hospital in Taiwan, we observed a positive correlation between CO(2) concentration and population, total bacteria, and particulate matter concentrations, thus monitoring CO(2) concentration as a general indicator for air quality could be a viable option. Consequently, an intelligent environmental monitoring system consisting of a CO(2)/temperature/humidity sensor, a digital plug, and a ZigBee Router and Coordinator was developed and tested. The system also included a backend server that received and analyzed data, as well as activating ventilation and air purifiers when CO(2) concentration exceeded a pre-set value. Alert messages can also be delivered to offsite users through mobile devices.