A Method for Identifying Distribution Automation Alert Signals Based on Decision Tree Modeling

In power systems, distribution automation terminals play a pivotal role in monitoring, controlling, and safeguarding the power distribution network. Among the numerous alarm signals generated, some may be redundant information, while others indicate critical defects, demanding precise analysis. This paper introduces a method that utilizes a decision tree classifier. Initially, we conduct a screening process to eliminate alarm signals stemming from equipment jitter, false alarms, and other non-critical causes. Subsequently, we identify the remaining alarm signals from multiple perspectives, taking into account factors such as battery usage, line fault types, and equipment status. Leveraging these characteristics, we categorize the alarm signals using the decision tree classifier. Each category corresponds to a specific fault type or event, enabling the identification of critical alarm signals. Ultimately, through application in the power distribution network, we have successfully validated the method's accuracy in identifying key alarm signals, thereby providing support for the stable operation of the power distribution system.