An adaptive PD type FLC with its real-time implementation on a servo position control system

摘要

A non-fuzzy adaptation scheme is proposed for a PD type fuzzy logic controller (FLC) by continuously modifying its input scaling factors (G, GΔℯ). Both Gℯ and GΔℯ are updated online by a nonlinear parameter computed through a mono-polar sigmoid function defined on the process error (ℯ) and change of error (Δℯ). Performance of the proposed adaptive fuzzy PD controller (AFPD) is compared with conventional as well as self-tuning fuzzy PD controller (FPD and STFPD). Simulation study and hard-ware based experimental results clearly establish the superiority of the proposed AFPD over FPD and STFPD during set point tracking and load rejection. Unlike STFPD which uses 49 control rules and additional 49 expert's defined gain rules, our proposed AFPD uses only 25 control rules along with an online non-fuzzy adaptation mechanism of the input scaling factors Gℯ and GΔℯ. In spite of almost 75% reduced rules compared to the well studied STFPD both the transient and steady state responses justify the effectiveness of the proposed AFPD.