Weak signal detection based on stochastic resonance （SR） can hardly succeed when noise intensity exceeds the optimal value of SR. This paper explores a novel parallel bistable SR array mechanism by decomposed multi-...Weak signal detection based on stochastic resonance （SR） can hardly succeed when noise intensity exceeds the optimal value of SR. This paper explores a novel parallel bistable SR array mechanism by decomposed multi-scale noises from input signal. A smoother output with lower noise is obtained from the combination of colored noise SR ellect and parallel bistable SR array. The influence of noise intensity and array size on the SR effect and output noise intensity is analyzed through numerical simu- lation. A signal detection method based on the new SR mechanism and normalized scale transform is proposed for the case of heavy background noise. Simulation is conducted to confirm the effectiveness of parameter tuning and amplitude tuning of normalized scale transform on the proposed SR model. The proposed method has three advantages： the input noise intensity of each unit is reduced by wavelet decomposition; the output noise level decreases due to array ensemble average; the SR effect of each unit is optimized by normalized scale transform for high frequency signal. Experiment on bearing inner and outer race fault diagnosis has verified the effectiveness and advantages of the proposed SR model in comparison with traditional SR method and kurlogram.展开更多
The front dynamics driven by a convection field in a model of FitzHugh-Nagumo type is studied both analytieMly and numerically. Saddle-node bifurcation induced by the convection field is found by using a singular pert...The front dynamics driven by a convection field in a model of FitzHugh-Nagumo type is studied both analytieMly and numerically. Saddle-node bifurcation induced by the convection field is found by using a singular perturbation analysis of front solutions. Convection field accelerates the B1och front propagating opposite the direction of convection field, but inhibits the Bloch front propagating along the direction of convection field. In addition convection field drives Ising front to travel opposite the direction of convection field.展开更多
基金supported by the National Natural Science Foundation of China (Grant Nos. 5107539, 51105366 and 51205401)the Research Project of National University of Defense Technology (Grant No. JC12-03-02)
文摘Weak signal detection based on stochastic resonance （SR） can hardly succeed when noise intensity exceeds the optimal value of SR. This paper explores a novel parallel bistable SR array mechanism by decomposed multi-scale noises from input signal. A smoother output with lower noise is obtained from the combination of colored noise SR ellect and parallel bistable SR array. The influence of noise intensity and array size on the SR effect and output noise intensity is analyzed through numerical simu- lation. A signal detection method based on the new SR mechanism and normalized scale transform is proposed for the case of heavy background noise. Simulation is conducted to confirm the effectiveness of parameter tuning and amplitude tuning of normalized scale transform on the proposed SR model. The proposed method has three advantages： the input noise intensity of each unit is reduced by wavelet decomposition; the output noise level decreases due to array ensemble average; the SR effect of each unit is optimized by normalized scale transform for high frequency signal. Experiment on bearing inner and outer race fault diagnosis has verified the effectiveness and advantages of the proposed SR model in comparison with traditional SR method and kurlogram.
基金Supported by the National Natural Science Foundation of China Grant Nos.11205044,11375051the Natural Science Foundation of Hebei Province,China under Grant Nos.A2011201006,A2012201015+1 种基金the Research Foundation of Education Bureau of Hebei Province,China under Grant No.Y2012009the Science Foundation of Hebei University
文摘The front dynamics driven by a convection field in a model of FitzHugh-Nagumo type is studied both analytieMly and numerically. Saddle-node bifurcation induced by the convection field is found by using a singular perturbation analysis of front solutions. Convection field accelerates the B1och front propagating opposite the direction of convection field, but inhibits the Bloch front propagating along the direction of convection field. In addition convection field drives Ising front to travel opposite the direction of convection field.
