Transformer Temperature Anomaly Detection Method Based on Adaptive Support Vector Machine
Due to the neglect of the influence of environmental temperature differences,the residual value of the surface temperature detection results of power transformers is relatively large.Therefore,a transformer temperature anomaly detection method based on adaptive support vector machine.Collect,enhance,and preprocess infrared images of the surface temperature of power transformers.Using an adaptive support vector machine with Gaussian kernel added,preliminarily classify the infrared images,construct a Gaussian kernel feature space,and classify linear and nonlinear features;Considering the influence of environmental temperature,the conventional surface temperature of power transformers is analyzed.After difference calculation,combined with adaptive calculation process,temperature anomaly detection results are obtained.The experimental results show that after the application of the proposed method,the temperature residual value of the surface temperature detection results of power transformers is small,the detection accuracy is high,and it meets the safe operation requirements of power transformers.
power transformertransformer abnormalityadaptive support vector machineinfrared imagessurface temperatureabnormal detection
NETL
NSTL
万方数据
