Abstract
The basic principle of wavelet neural network is analyzed and the learning algorithm is improved as well. Correspondingly, fault pattern recognition model is constructed based on improved wavelet neural network. Simulation experiment shows that improved wavelet neural network has good fault-tolerant capacity and global convergence. However, how to establish an objective criterion of parameter settings and enhance the real time capability of fault pattern recognition still need further research.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have