Source location of cracks in a turbine blade based on kernel principal component analysis and support vector machines

Source location of acoustic emission signals of a crack based on kernel principal component analysis (KPCA) and support vector machines (SVM) was studied here.The results showed that the accuracy of location using the feature parameters extracted with KPCA technique is improved comparing with that using the original parameters,i.e.,the recognition rate of the crack region is 100 percent; the maximum error of the support vector regression analysis for the distance from the source of cracks to the welding seam is 20cm when the number of the input feature parameters is nine.As a result,it was a good method to combine KPCA with SVM for crack source location of complex big-size structures.It decreased the dimensions of input signals and improved the accuracy of location as well.