Modified Expert Inference Method of Power Substation Monitoring System Based on Expansion of Multi-sensor Utilization for Fire Discrimination

Abstract

In this paper, we propose a monitoring system that applies a new fire detection method to quantify the possibility of fire by using the modified expert inference method based on fuzzy logic. This system improves the expert inference method through expansion of the utilization of the sensor data related to the fire in multiple sensors and improving the performance of fire detection in the power substation. At this time, the proportional sum operation is applied to the inference process to reflect the data from various sensors during the min - max operation on the fuzzy value of the sensor input. By applying the proportional sum operation, the data of all the sensors are reflected in the inference, and the utilization of the sensor is expanded, and consequently, the expert inference is improved. Thus, we can expect a more accurate and careful implementation of power substation fire-monitoring system by applying the expert inference improvement method to fire detection.