Intelligent Fault Diagnosis of Electronic Information Systems Using Lightweight Deep Networks with Attention and Multi-Representation Domain Adaptation

Abstract

With the continuous progress of information technology, the important role of electronic information system in modern society has become increasingly prominent, and its stability and reliability have become the focus of people’s attention. However, in the long-term operation of electronic information systems, various failures are inevitable, which poses great challenges to the normal operation of the system. Therefore, based on the urgent demand for fault diagnosis in electronic information systems and the development trend of AI technology, this study proposes a deep learning fault diagnosis model that integrates P-HetConv and CBAM, and introduces a federated learning mechanism to optimize data processing. The research collects fault data of electronic information systems in different fields and types, and constructs a dataset containing various fault types such as hardware and software, with a total of 1,000 samples. Experimental results show that the diagnostic accuracy of the model is as high as 96.81%, which is 15% higher than that of the traditional rule-based diagnosis method, and is significantly better than the traditional method in terms of accuracy, recall, F1 score and other indicators, and shows good adaptability and generalization ability in complex fault scenarios. This study verifies the application value of AI technology in the field of fault diagnosis of electronic information systems, and provides a strong guarantee for the efficient and stable operation of the system.