Based on the Reynolds equation with Reynolds boundary conditions, the Castelli method was employed to solve the Reynolds equation for oil lubrication upon bearings. By doing so, a profile of nonlinear oil film force of single-pad journal bearings is established. According to the structure of combination journal bearings, nonlinear oil film force of combination journal bearing is obtained by retrieval, interpolation and assembly techniques. As for symmetrical flexible Jeffcott rotor systems supported by combination journal bearings, the nonlinear motions of the center of the rotor are calculated by the self-adaptive Runge-Kutta method and Poincar6 mapping with different rotational speeds. The numerical results show that the system performance is slightly better when the pivot ratio changes from 0.5 to 0.6, and reveals nonlinear phenomena of periodic, period-doubing, quasi-periodic motion, etc.
The probability of long-range connection among neurons could be changeable in biological neuronal networks. In this paper, the probability of long-range connection between neurons is not fixed at a constant but varies in a numerical region (≤p0 ), and then the collective behaviors of neurons are detected. A statistical factor in the two-dimensional space is used to detect the phase transition and robustness of spiral wave in the active network of neurons. It is found that the development of spatiotem-poral pattern depends on the numerical region (≤p0 ) for the probability of long-range connection. Coherence resonance-like behavior is observed due to the fluctuation in the long-range probability. Spiral waves emerge to occupy the network of neurons under an optimized probability of long-range connection, and it shows certain robustness in weak channel noise.
