Infrared image diagnosis method and infrared image diagnosis system for transformer station equipment

The invention discloses an infrared image diagnosis method for transformer station equipment. The infrared image diagnosis method is used for diagnosing operation state of the transformer station equipment and comprises the steps of acquiring k infrared thermal image pictures of the transformer station equipment and performing dividing, extracting the infrared thermal image picture characteristic parameter of the infrared thermal image pictures as an input vector sample, constructing an SOM nerve network and performing circulated training, and defining classification of an output neuron in a mapping relationship with the input vector sample by means of the transformer station equipment operation state which corresponds with the input vector sample; acquiring the infrared thermal image picture of to-be-diagnosed transformer station equipment and performing dividing, extracting the infrared thermal image picture characteristic parameter of the infrared thermal image picture, performing self-organizing mapping by the trained SOM nerve network in a manner that the infrared thermal image picture characteristic parameter is used as the input vector, wherein the classification which comprises the output neuron in the mapping relationship with the input vector sample is judged as a diagnosis result for the transformer station equipment operation state which corresponds with the infrared thermal image picture of the to-be-diagnosed transformer station equipment.

[1]  Bo-Suk Yang,et al.  Application of Self Organizing Map for Intelligent Machine Fault Diagnostics Based on Infrared Thermography Images , 2011, 2011 Sixth International Conference on Bio-Inspired Computing: Theories and Applications.