A fast and compact fuzzy neural network for online extraction of fuzzy rules
Inst. of Autom., Dalian Maritime Univ., Dalian, ChinaDOI: 10.1109/CCDC.2009.5192394 Conference: Control and Decision Conference, 2009. CCDC '09. Chinese
Source: IEEE Xplore
A novel paradigm termed fast and compact fuzzy neural network (FCFNN), which incorporates a pruning strategy into some growing criteria, is proposed for online extraction of fuzzy rules. The proposed growing criteria not only speed up the online learning process but also result in a parsimonious fuzzy neural network while achieving comparable performance and accuracy by virtue of the growing and pruning mechanism. The FCFNN starts with no hidden neurons and parsimoniously generates new hidden units according to the proposed growing criteria as learning proceeds. In the second learning phase, all free parameters of the hidden units are updated by the extended Kalman filter (EKF) method. The performance of the FCFNN algorithm is compared with other popular algorithms like ANFIS, GDFNN and SOFNN, etc., for nonlinear function approximation. Simulation results demonstrate that the learning speed of the proposed FCFNN algorithm is faster and the network structure is more compact while comparable generalization performance and accuracy are achieved, moreover, it is capable of extracting fuzzy rules online.
- [Show abstract] [Hide abstract]
ABSTRACT: This thesis presents a fault diagnosis method based on the low, middle and high level fuzzy neural networks for the breakdown asynchronous motor according to the complex corresponding relations between the motor's fault symptoms and the fault causes. This can implement the fuzzy diagnosis for the motor fault. The thesis puts emphasis on the structure models of the new type hierarchical fuzzy neural network and the relative learning algorithms. And it also introduces the simulation training of the hierarchical fuzzy neural network based on the models and algorithms. At last, the experimental results show that this diagnosis method can effectively classify the single fault samples and the multi fault samples of the motor and this not only can raise the accurateness rate of the diagnosis, but it also possesses a good applicable value in engineering.Sixth International Conference on Natural Computation, ICNC 2010, Yantai, Shandong, China, 10-12 August 2010; 08/2010
- [Show abstract] [Hide abstract]
ABSTRACT: A multi-input/multi-output (MIMO) furnace control system based on the PID control and the fuzzy radical basis functions neural network(FRBFNN) control is presented, in this paper. In addition, the structure models and relative learning algorithms of the FRBFNN controller are gived. At last, simulation and experiment results based on an application of the MIMO furnace control system are provided to show that FRBFNN-control system has more robustness than the PID-control system especially in reducing noise and overcoming non linearities.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.