Abstract We have outlined the software-based evaluation of electrical power quality parameters in the Wireless Sensor Network settings. The proposed components include the reference waveforms, a software application for the measurement of standard PQ parameters and two microcontroller-based wireless sensor modules for transmission. The quality parameters are used to reference functions with the control software application in the Lab environment. We... We have outlined the software-based evaluation of electrical power quality parameters in the Wireless Sensor Network settings. The proposed components include the reference waveforms, a software application for the measurement of standard PQ parameters and two microcontroller-based wireless sensor modules for transmission. The quality parameters are used to reference functions with the control software application in the Lab environment. We have used the IEEE standard 802.15.4 protocol for transmission. According to the Guide to the Expression of Uncertainty in Measurement, the procedure for calculating Type A, Type B, combined and expanded measurement uncertainty, is supported using professional instrumentation with high accuracy levels. Metrological evaluation of a supported experimental system based on wireless sensor network (WSN), applicable in the measurement of electrical power quality (PQ) parameters and disturbances, is presented in this paper. The system includes a based generator of reference waveforms, a software application for measuring standard PQ parameters and two microcontroller-based wireless sensor modules for transmitting and receiving measurement results. Measures of basic quality parameters for reference waveforms are performed using the control software application in the LabVIEW environment. Wireless sensor networks are based on the communication standard IEEE 802.15.4 (Zigbee). Metrological assessment includes calculating measurement uncertainty components and the final presentation of measurement uncertainty budget. The procedure for calculation of Type A, Type B, combined and expanded measurement uncertainty, according to the Guide to the Expression of Uncertainty in Measurement, is supported using professional instrumentation with high accuracy levels. Read More Read Less

Abstract With the increase of social pressure and mental health problems, evaluating and diagnosing college students’ mental health has become particularly important. To improve the accuracy and efficiency of assessment and diagnosis, we propose a neural network-based mental health model for college students. This model adopts deep learning technology to assess and diagnose students’ mental health by analyzing their personal information,... With the increase of social pressure and mental health problems, evaluating and diagnosing college students’ mental health has become particularly important. To improve the accuracy and efficiency of assessment and diagnosis, we propose a neural network-based mental health model for college students. This model adopts deep learning technology to assess and diagnose students’ mental health by analyzing their personal information, behavioral data, and academic performance. Specifically, we used models such as Convolutional Neural Networks (CNN) and Long Short-Term Memory Networks (LSTM) to conduct a multidimensional analysis of students’ learning behavior, emotional changes, and social situations. At the same time, we have also utilized advanced technologies such as self-attention and attention mechanisms to better capture students’ information from different dimensions. Through extensive experimental data validation, we found that the neural network-based mental health model for college students has high accuracy and efficiency. This model has higher diagnostic accuracy and a lower misdiagnosis rate than traditional evaluation and diagnostic methods. In addition, the model also has good generalization performance and can adapt to the application needs of different types of students and different scenarios. Read More Read Less

Abstract With the development of the economy and the improvement of people’s living standards, the field of art and design has gained widespread application and attention. Evaluating art and design talents is a vital link in art and design. How to objectively and accurately evaluate the comprehensive quality of art and design talents is an urgent problem to be solved. This... With the development of the economy and the improvement of people’s living standards, the field of art and design has gained widespread application and attention. Evaluating art and design talents is a vital link in art and design. How to objectively and accurately evaluate the comprehensive quality of art and design talents is an urgent problem to be solved. This article proposes a talent evaluation model for art and design, combining the entropy method and BP neural network. This model aims to objectively and accurately evaluate the comprehensive quality of art and design talents and provide a scientific basis for talent selection. This article first introduces the evaluation model’s basic principle and construction process and then elaborates on the specific implementation method of combining the entropy method with the BP neural network. Finally, the feasibility and effectiveness of the evaluation model are verified through experiments. Read More Read Less