Abstract Physical exercise is essential for the sustainable self-cultivation of college students and is an indispensable part of higher education. However, most college students lack systematic knowledge of physical exercise, making it challenging to exercise scientifically. Therefore, this paper introduces the ant colony algorithm into the model for extracting characteristics of college student's physical exercise behavior to improve the effectiveness of... Physical exercise is essential for the sustainable self-cultivation of college students and is an indispensable part of higher education. However, most college students lack systematic knowledge of physical exercise, making it challenging to exercise scientifically. Therefore, this paper introduces the ant colony algorithm into the model for extracting characteristics of college student's physical exercise behavior to improve the effectiveness of behavior recognition and clustering. Furthermore, the model's performance is enhanced through optimization of the ant colony algorithm. The experimental results demonstrate that the clustering model of college student's physical exercise behavior, based on the ant colony algorithm, effectively reduces the error rate and maintains good accuracy as the sample size increases, indicating good stability and reliability. Additionally, for different physical exercise behaviors, the clustering F-measure standard values of the model all exceed 0.8, indicating a better clustering effect. Read More Read Less

Abstract Physical exercise is essential for the sustainable self-cultivation of college students and is an indispensable part of higher education. However, most college students lack systematic knowledge of physical exercise, making it challenging to exercise scientifically. Therefore, this paper introduces the ant colony algorithm into the model for extracting characteristics of college student's physical exercise behavior to improve the effectiveness of... Physical exercise is essential for the sustainable self-cultivation of college students and is an indispensable part of higher education. However, most college students lack systematic knowledge of physical exercise, making it challenging to exercise scientifically. Therefore, this paper introduces the ant colony algorithm into the model for extracting characteristics of college student's physical exercise behavior to improve the effectiveness of behavior recognition and clustering. Furthermore, the model's performance is enhanced through optimization of the ant colony algorithm. The experimental results demonstrate that the clustering model of college student's physical exercise behavior, based on the ant colony algorithm, effectively reduces the error rate and maintains good accuracy as the sample size increases, indicating good stability and reliability. Additionally, for different physical exercise behaviors, the clustering F-measure standard values of the model all exceed 0.8, indicating a better clustering effect. Read More Read Less

Abstract This paper designs a distribution network external force damage early warning system based on audio classification technology, focusing on the potential damage that large construction machinery might cause to cables. By utilizing audio data, the system can accurately monitor and analyze the real-time status of construction sites to identify potential cable damage activities. Leveraging a cloud-edge collaboration architecture, the system employs cloud-based model training... This paper designs a distribution network external force damage early warning system based on audio classification technology, focusing on the potential damage that large construction machinery might cause to cables. By utilizing audio data, the system can accurately monitor and analyze the real-time status of construction sites to identify potential cable damage activities. Leveraging a cloud-edge collaboration architecture, the system employs cloud-based model training and edge computing technology to achieve realtime monitoring and data processing, thereby enhancing the accuracy and timeliness of warnings. Integrating machine learning algorithms and pattern recognition technology, the system can automatically analyze and predict the impact of construction machinery on cables. Through a visualization system, it provides real-time warnings and response measures, effectively reducing losses caused by external damage and ensuring the safe operation and reliability of the distribution network and power supply. Read More Read Less

Abstract Under the influence of geographical environment and seasonal differences, wind speed changes show strong characteristics of volatility, intermittency and high variability. To make better use of wind energy in the Kumtag Desert of Xinjiang, this paper proposes a Conv-Informer model and a loss function including a trend penalty factor, combined with a data set partitioning strategy. It researches ultra-short time surface wind prediction in... Under the influence of geographical environment and seasonal differences, wind speed changes show strong characteristics of volatility, intermittency and high variability. To make better use of wind energy in the Kumtag Desert of Xinjiang, this paper proposes a Conv-Informer model and a loss function including a trend penalty factor, combined with a data set partitioning strategy. It researches ultra-short time surface wind prediction in the Kumtag Desert based on deep learning. The results show that the data set partitioning strategy has a crucial impact on the effect of model training. In the results of Conv-Informer model training, the prediction accuracy of the wind speed above 10 m/s within 30 minutes is more than 93%, and the linear trend of wind speed change can be predicted more accurately. Read More Read Less