Adaptive Detection Method for Dynamic Network Attacks in New Power Systems

Under the requirements of intelligent power system, traditional power systems are gradually transforming into new power systems that are deeply integrated with the network. During this process, the high integration of power systems and information technology has made the new power system face the threat of network intrusion. Currently, there are insufficient detection methods for network attacks in the power field, and there are the following problems. Due to the privacy characteristics of the power system, it is difficult to obtain high-quality network security data in the power field. The power network environment presents dynamic changes, and traditional detection methods are difficult to adapt to network changes, resulting in a decrease in detection performance. This article proposes an IPCA-ATG intrusion detection method. Specifically, a feature weighted IPCA algorithm is proposed to deeply extract potential features from the data, calculate the degree of deviation of abnormal data in the overall data, and obtain the anomaly score of the detection data. Propose the ATG algorithm to generate adaptive thresholds that can dynamically adjust with changes in the network environment based on the overall data. Combining IPCA and ATG algorithms, calculate the anomaly score of detection data and perform threshold judgment to achieve effective detection of network intrusion. The experimental results show that this method has higher detection accuracy and lower false alarm rate compared to traditional intrusion detection methods, and does not rely on annotated data. At the same time, the experimental results demonstrate the effectiveness and robustness of this method in dynamic network environments.