| Fuzzy interpolative control based on sparse rules is very interesting. Some works have been made to break through the shortcoming of the traditional fuzzy reasoning method of CRI, and some steps of the conventional fuzzy interpolative control algorithm have been deleted in this paper. Charged with object model unknown circumstances, a linear interpolative reasoning approach under multidimensional fuzzy sparse rules and cubic spline function were applied to design the fuzzy controller for sparse rules. According to the original rules and the linear interpolative reasoning approach, to build new rules at the same monotone interval, and received complete's fuzzy rules, which is satisfied the requirement of the control input for every fuzzy control algorithm.The linear interpolative reasoning approach is an effective way to receive complete's fuzzy rules. Then an approximate control function was established from control input to control output under the complete's fuzzy rules. The control function was used to calculate the output in response to the every control input. Because the piecewise Lagrange Interpolation and the piecewise Newton Interpolation have the nonsmooth phenomenon, and the high order interpolations have the range phenomenon, so the cubic spline function was put forward to structure the approximate control function, which has low degree and good smoothness. That is an important step in high precision to receive a higher accuracy of fuzzy systems.At last, in order to verify the feasibility of the above thought and the operability of the above methods, to select the automotive Semi-active Suspension system, of which has sparse rules, and used the Matlab programming to calculate the result. Under the complete's fuzzy rales, the simulation of fuzzy interpolative controller based on cubic spline function for the special road excitation is better than the conventional fuzzy control methods. That shows the above thought and method have a good application prospect.
|
PDF Full Text Request