A Neuro-Fuzzy System Design Methodology for Vibration Control

Abstract

Fuzzy system has been known to provide a framework for handling uncertainties and imprecision by taking linguistic information from human experts. However, difficulties arise in determining effectively the fuzzy system configuration, i.e., the number of rules, input and output membership functions. A neuro-fuzzy system design methodology by combining neural network and fuzzy logic is developed in this paper to adaptively adjust the fuzzy membership functions and dynamically optimize the linguistic-fuzzy rules. The structure of a five-layer feedforward network is shown to determine systematically the correct fuzzy logic rules, tune optimally (in the sense of local region) the parameters of the membership functions, and perform accurately the fuzzy inference. It is shown both numerically and experimentally that engineering applications of the neuro-fuzzy system to vibration control have been very successful.

Development of a rule selection mechanism by using neuro-fuzzy methodology for structural vibration suppression

“…A neuro-fuzzy system with self-organized, optimal fuzzy rules and membership functions presented by Yang et al [3,12] and Chen et al [13,14] is cited. The structure of the model is a five-layer structure with a multi-input-multi-output (MIMO) feedforward network, as shown in Fig.…”

“…Two parts need to be tuned: the first part is the determination of membership functions and the second part is the selection of fuzzy rules. Based on the model of Lin and Lee [5,6], Farag et al [11] emphasized the learning algorithm and finding fuzzy rules, and Yang et al [3,12] and Chen et al [13,14] implemented a neuro-fuzzy system with three phases in the five-layer feedforward network using Mamdani's fuzzy model [15]. After selecting the rules, the parameters of the membership functions are then tuned optimally in error backpropagation learning.…”

“…In this paper, an experiment-verified five layer structure and three-phase (self-organized, backpropagation, and supervised learning) network presented by Yang et al [3,12] and Chen et al [13,14] are cited to show how the neuro-fuzzy system automatically selects the activation rules using both system identification and vibration control examples.…”

Adaptive Compensation of Contact‐Induced Vibration in High Density HDD Servo Systems Using Phase Lead Peak Filter Method

“…These vibrations are likely to result in off‐track vibration and delay the read/write time significantly. Compared with the abundant research results in other types of vibration control problems , the investigations on CIV compensation of HDD systems are few, mainly due to the difficulties in fundamental control theory and implementation technology. Three main obstacles are: (i) CIV modeling; (ii) actuator bandwidth limit; and (iii) the performance requirement for a very fast or deadbeat type vibration control.…”