| With the continuous development of economy and social progress, industrial boiler plays a more and more important role in national economy and society. But the industrial boiler works in bad condition, frequent boiler pipe leakage accident, the serious influence production equipment safety, reliability, stability, economic operation. In order to early detection of boiler pipe leakage, to reduce the losses to a minimum, but also to reduce and has extremely important significance to avoid the pollution of the environment. Therefore, in the early detection of leakage in this paper boiler pipeline leak. But the traditional boiler leak detection methods need of boiler tube leak detection is very significant to further expand out, easy to cause the accident, so the traditional method has been increasingly unable to meet modern production requirements. In this paper, the acoustic emission pipeline leak detection method based on the study, and provide the evidence for the application of acoustic emission testing of boiler tubes.In this paper, leakage acoustic emission testing technology is used in a pipeline of boiler which based on support vector machine. The pipeline leakage feature extraction of acoustic emission signal, and then the characteristic parameters are extracted as feature vector is input to the artificial neural network and support vector machine classification, and training, and carries on the contrast analysis. The neural network and support vector machine can transmit signal to judge the leakage signal is measured by using the acoustic category, but the support vector machine is more accurate, this method can effectively identify the leakage signal. The test results show that the support vector machine method is feasible and effective in the diagnosis of leakage in the pipeline, to verify the application of acoustic emission technology in pipeline leak detection and the feasibility of detection, which provides a new method for waste heat boiler pipe leakage.
|
PDF Full Text Request