Skip to Main Content
To precisely measure temperature in high-voltage electrical power equipment subject to intense electromagnetic interference (EMI), we present an artificial neural network (ANN) inverse compensating method, based on which we also construct a compounded fiber-optic temperature measurement system (CFOTMS). The CFOTMS consists of a fiber-optic semiconductor absorption temperature sensor (FOSATS) and an ANN inverse compensator. On one hand, optical-fiber-based light transmission is insusceptible to EMI; on the other hand, due to the utilization of the ANN inverse compensator, the dynamic performance and measuring precision of the CFOTMS are significantly enhanced compared with the sole use of the sensor. Therefore, it is very suitable to be used in high-voltage electrical power equipment. Furthermore, as the intensity-dependent FOSATS is a cheap device, and the ANN inverse compensator is realized in software, the cost of the CFOTMS will be low, which makes it suitable to be used in economical equipment.