Modeling of Fault Alarm Monitoring System on IoT-Based Control Panel Alarm Annunciator

Abstract

The existence of substation operators is essential for monitoring substation equipment and operating control rooms. In addition, substation operators also have other tasks, such as inspecting substation equipment in the switchyard or in-service inspection (Level 1 inspection). This condition is less than ideal because the operator cannot monitor the control room continuously while in the switchyard. The control panel at the substation is equipped with an alarm annunciator, which is auxiliary equipment that serves to provide warning signs to the operator regarding which protection functions are working. Therefore, it is necessary to have a monitoring system that can provide real-time information about the condition of the control panel alarm annunciator. This research models a fault alarm monitoring system that can provide notification messages according to the type of fault automatically through the Telegram application. The system test results show the average value of 100 test data on the five types of fault, resulting in an accuracy rate of 100%, a precision rate of 100%, a recall rate of 100%, an error rate of 0%, and an F1-score rate of 100%. This is because all test data on the five types of fault were detected correctly, and no test data from other types of fault was detected as the five types of fault. Based on these averages, it can be concluded that using a confusion matrix to measure the performance of the fault alarm monitoring system on the IoT-based control panel annunciator alarm shows excellent system performance results.

Keywords

• alarm annunciator
• confusing matrix
• iot
• monitoring
• substation operator