Research on infrared diagnosis of circuit based on data fusion

Circuit diagnosis usually take voltage or current as its input. This contact method has deficiencies of variation of data input during testing and inaccessibility of inevitable nodes. On the other hand, non-contact diagnosis, such as temperature or magnetic field diagnosis, also has its defects of inaccuracy of data and insufficiency of algorithms. The paper acquires contact data of voltage and non-contact data of temperature of circuit under test. It diagnoses the circuit by contact and non-contact method separately, and then fuses the two results into a final one. It makes good use of the advantages of both contact and non-contact diagnosis but makes up for the deficiencies of the two. The experiment shows the combined method gets the best fault coverage and resolution among the three methods.