Array sensors online pattern recognition based on FCM and ANFIS
Chengbing Li, Mao Xi-hao · International Journal of Computers and Applications · 2018
The measurement errors of array sensors are generated by cross-sensitivity during measuring, the algorithm based on fuzzy C-means clustering algorithm (FCM) and adaptive neuro-fuzzy inference system (ANFIS) is proposed for pattern recognition of array sensors in this article. The fuzzy C-means clustering algorithm is used to reduce the number of experimental data, the number of training samples has been reduced from the 150–15, and use the center points that generated by fuzzy C-means clustering algorithm as the input of adaptive neuro-fuzzy inference system to complete the training of system. The speed of computation and convergence of adaptive neuro-fuzzy inference system performs better than traditional neural networks. The accuracy and speed of calculation will be influenced seriously. The simulation results show that the hybrid algorithm based on fuzzy C-means clustering and adaptive neuro-fuzzy inference system can effectively identify four kinds of gases, the performance of convergence speed and success rate of pattern recognition is excellent.