Research on User Abnormal Behavior Recognition Model for Water Economy Operating System Based on Transformer

The water economic operation system is an important tool for water resources management and can greatly improve the operating efficiency of power stations. This article studies the commonalities of user abnormal behavior from system logs. The paper proposes a Transformer-based user abnormal behavior identification model (UABID- TF) for water economic operation systems to ensure stable operation. The logs of the water economy operation system are structured data composed of text, dates and numbers. Traditional data preprocessing methods normalize features with hashing, resulting in the loss of correlation between various features. This article proposes a feature fusion preprocessing method that can effectively reflect the interaction between features rather than treating them independently. Considering the imbalance of user behavior data, gaussian random noise is utilized in this paper to augment the anomaly sample dataset, addressing the issue of imbalanced data. Experimental results show that proposed method achieves a 6% improvement in prediction accuracy over traditional data preprocessing methods. Additionally, experiments compared the UABID- TF model with other deep learning models, including the bi-directional long short-term memory network (Bi-LSTM) and the ID-CNN model. The results demonstrated that UABID-TF achieved higher accuracy and a lower false alarm rate. The research results provide scientific decision-making support for the electric power industry. By identifying abnormal user behavior in real time, potential security threats can be discovered promptly, allowing for the implementation of corresponding preventative measures. This approach can better address network security incidents and ensure the stable operation of the power grid, ultimately contributing to national security and stability.