This paper investigates an algorithm for IT fault prediction and diagnosis, leveraging a generative intelligent model, with the aim of enhancing the stability and reliability of information technology systems. Amidst the proliferation of information technology, system faults have become more frequent, rendering traditional fault detection methods insufficient in terms of real-time capability and accuracy. Consequently, a novel model that integrates Generative Adversarial Networks (GANs) with Long Short-Term Memory (LSTM) networks is introduced. Initially, GANs are employed to generate a diverse array of historical fault samples, thereby augmenting the training dataset. GAN is composed of generator and discriminator. Through adversarial training, the generator optimizes the output and makes the generated data more close to the actual fault distribution. Secondly, LSTM conducts time series analysis on the generated data to capture the time series characteristics of the occurrence of faults. LSTM uses memory units and gating mechanisms to effectively deal with long time dependencies, thereby learning system behavior patterns and potential fault signals. In addition, an ensemble learning framework is designed to combine LSTM model and GAN to improve the prediction accuracy. The results show that this model can significantly improve the accuracy of fault prediction and reduce the false positive rate. Compared with traditional methods, it has improved many performance indicators.
Research on IT fault prediction and diagnosis algorithm based on generative intelligent model
23.10.2024
557862 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Intelligent charging pile health diagnosis and fault prediction system
Europäisches Patentamt | 2025
|Supervised SVM Based Intelligent Fault Diagnosis Methods
Springer Verlag | 2023
|