Fault prognostics using dynamic wavelet neural networks
Artificial Intelligence for Engineering Design, Analysis and Manufacturing
Wavelet support vector machine
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
De-noising by soft-thresholding
IEEE Transactions on Information Theory
Hi-index | 0.00 |
To improve the accuracy of the prediction in avionics prognostics and health management (PHM), a variety of theories and methods are studied. In this paper, a prediction algorithm based on multiwavelet support vector machine(WSVM)is proposed. Multiwavelet denoising is used for signal data preprocessing. Then multiwavelet is employed to decompose the data into several subsequences at different scales. These subsequences are predicted by different support vector machines respectively. Finally, the final predicted results reconstituted from the subsequences are obtained. To validate the model, experiment data from a set of certain avionics voltage data is used. Predicted results of the proposed algorithm are validated to be more accurate than that of traditional support vector machine prediction algorithm. The mean square error (MSE) is decreased to 0.1956.