The gearbox, a key transmission device in industrial applications, can cause severe vibrations or failures when anomalies occur. With increasing industrial automation complexity, precise anomaly detection is crucial. This paper introduces DCCopGAN, a novel unsupervised framework that uses a deep convolutional copula-generative adversarial network for unsupervised multi-sensor anomaly detection in industrial gearboxes. Firstly, a Deep Convolutional Generative Adversarial Network (DCGAN) generator is trained on high-dimensional normal operational data from multi-sensors to learn its underlying distribution, enabling the calculation of reconstruction errors for input samples. Then, these reconstruction errors are analyzed by Copula-Based Outlier Detection (CopOD), an efficient non-parametric technique, to identify anomalies. In the testing phase, reconstruction errors for test samples are similarly computed, normalized, and then evaluated by the CopOD mechanism to assign anomaly scores and detect deviations from normal behavior. The proposed DCCopGAN framework has been validated on a real gearbox dataset, where experimental results demonstrate its superior anomaly detection performance over other methods.
5 June 2025 in Applied Mathematics and Nonlinear Sciences
The student information network of colleges and universities undertakes various tasks such as student management, digital campus construction, etc. Improving the network attack and defense capabilities is crucial for safeguarding campus information security. The research idea of this paper is to extract the characteristics of student network through network topology, select the improved small-world network as the attack and defense simulation model, and design the network information security protection system with the architecture of attack and defense game. The selected NW small-world network model is able to accurately simulate the process of network attacks and analyze the importance of the position of different sub-networks in the student network. In the face of a total of 1,100 attacks such as account stealing, Trojan planting, rebound attack, webpage attack, and data eavesdropping, the new system reduces a total of 780 information attacks compared with the traditional method, demonstrating efficient attack and defense capabilities. Therefore, it can be considered that the path of improving the network attack and defense ability of college students explored in this paper has achieved the established research purpose.