Flaw identification of undercarriage based on particle swarm optimization algorithm and support vector machine

Aerial in-situ test is an important constituent of modern aerial maintenance and repairing technology, which can detect the performance of aircraft structure quickly. Aiming at the low efficiency of the present in-situ test methods, the paper proposed a novel method named PSO-SVM which combined the improved particle swarm optimization (PSO) with support vector machine (SVM). The method could classify different defects accurately. The relationship between key parameter C and σ of SVM and the classification accuracy were calculated, therefore the optimized results were got. Experiments of classification were made and the results illustrated that the method could classify typical aircraft defects effectively, which had its special advantages on the short training time cost. This would have a promising application in rapid in-situ aircraft test.