State Detection of Switch Cabinet Based on Multi-sensor Information Fusion

The main function of switch cabinet is to open and protect the electrical equipment in the power system. It is an important hub in the power system, therefore, it is necessary to detect its operating state to ensure that it works well. Switch cabinet is composed of lots of electrical equipment. As a result, the state detection of the switch cabinet is equivalent to detect the state of the electrical equipment in the switch cabinet. However, because of the complex environment of the switch cabinet, the accuracy and adaptability of traditional detection methods are low. In this paper, a new method for the state detection is proposed, which combines fuzzy theory with D-S evidence reasoning. Firstly, temperature and humidity sensor is used to detect the ambient temperature and the relative humidity in the switch cabinet, infrared sensors are used to detect the temperature of the electrical equipment, ultraviolet sensors are used to detect the partial discharge of the electrical equipment. Secondly, these information is transmitted to workstation through ZigBee wireless network. Then, the fuzzy membership degrees of characteristic quantities are calculated according to the given fuzzy membership function. In order to enhance the anti-jamming ability of the system, the method of calculating the reliability among the characteristic quantities is introduced. According to the result of each measurement, the basic probability distribution function (mass function) is calculated. Finally, the information of multiple measurements in a detection cycle is fused by evidence theory. The experimental results show that the method enhances the accuracy and adaptability of the state detection of the electrical equipment and reduces the uncertainty and one-sidedness caused by detecting with only one sensor.