Nonlinear error correction method for temperature sensor of on-load tap changer based on neural network optimized by genetic algorithm

On-line temperature monitoring of on-load tap-changer is a conventional technology to monitor the running state of on-load tap-changer. In order to ensure the effectiveness of monitoring, it is necessary to test the sensitivity of the sensor. However, there are inevitable errors in the detection process of temperature sensor, which seriously affect the detection results. In order to eliminate the influence of errors on the detection results as much as possible, this paper establishes a nonlinear error correction method of on-load tap-changer temperature sensor based on neural network optimized by genetic algorithm. The algorithm flow is proposed. 140 of 158 data are randomly selected as neural network training, and the remaining 18 values are used as neural network correction. The results show that the corrected test results are basically consistent with the actual values.