The OGY method can harness chaotic motion of plants to the periodic orbit which is arbitrarily chosen and has been already verified on comparatively simple and well-known chaotic attractors like Lorentz attractor and so on. Its method is expected because it doesn't need any arithmetic model like differential equations but time-series data of single variable which represents the substantial behavior of target plants. We have already studied on an application of its method to a forced pendulum, whose pivot point is sinusoidally vibrated horizontally, as a typical mechanical plant and shown some problems to be considered.
In this paper, trying to apply the OGY method to the number of the saddle points of the periodic orbits which occur chaotic motion on the forced pendulum, influence of noise given onto the parameters of the forced pendulum is studied using a numerical simulation. Consequently it is suggested that the angle between stable direction and unstable one and the relation between their directions around saddle points on the Poincare section and the directions of the axes of Poincare sections is substancial for the control performance.
