Search Results

‘Three kinds of new simultaneous optimum design methods of plural dynamic absorbers are proposed. These methods allow the optimum tuning in many natural modes of multiple degrees of freedom structures or a continuous bodies simultaneously to effectively suppress vibration. Changes of natural modes and natural frequencies of the main structure due to added mass effect of dynamic absorbers can be taken into account in the design. Validity and usefulness of the proposed methods are verified by both a computer simulation and by experiments.

State-space solutions of H∞ controller have been well developed. Hence to a real structure control design, the first step is to get a state space model of the structure. There are analytical and experimental dynamic modeling methods. As we know, it is hard to obtain an accurate model for a flexible and complex structure by FEM(Finite Element Method). Then the experimental modeling methods are used. In this paper, we use frequency domain modal analysis technique based on system FRF(Frequency Response Function) data and ERA(Eigensystem Realization Algorithm) time domain method based on system impulse response data to establish state-space model in order to design H∞ control law for the purpose of vibration suppression. The robust control implementation is exerted on a testbed (truck cab model device) with three degrees of freedom. The validity of experimental state-space modeling is testified and the obvious vibration control performances are achieved.

Two methods have been developed for real time identification and classification of the roughness pattern of road surfaces using the neural network. These methods are directly available both for semi-active and active vibration controls of cars. Accelerations of the rear wheel axis under the suspension are used as the input data for real time identification. The neural network which has acquired the informations of the seven typical roughness patterns is used for real time classification of actual road surfaces during driving. Validity and usefulness of these methods are verified by simulation.

Two kinds of active control systems, using neural networks (NN), are presented for realizing optimal driving motion of four wheel steer (4WS) cars. The first system is based on the assumption that the car is simplified as a linear two wheel bycycle model, and that the friction force between tire and road surface is represented by Fiala's nonlinear model. The nonlinear relation between the slip angle of tire and the cornering force is expressed with NN. A model-following type control strategy is adopted in the first system, with both the feedforward and feedback gains for the control of the rear wheel steering angle adaptively determined with NN according to change of front wheel steering angle. The second system is based on the assumption that both the dynamical characteristics of the car and the tire friction force are nonlinear. The nonlinear dynamical characteristics of the car and the friction force are identified with NN, using the measured data of an actual car.

This paper concerns the Direct System Identification Method (hereafter referred to as DSIM) which allows accurate and quick determination of two groups of properties which exercise dominant effects on low frequency vibration of a vehicle body. The first group is the rigid body properties of an engine. The second group is the properties of each engine mount. Under the assumption that the engine/mount system is a rigid body, this paper makes theoretical discussion for using the DSIM to induce the parameters of an engine/mount system, and makes improvements for better correlation with experiments. Also mentioned is a comparison of this study with the experimental results and verification of consistency on those parameters obtained from DSIM to predict the accurate vehicle characteristics, along with the role this method will play in upgrading the technology of prediction analysis.

In this paper, a new method which directly identifies characteristic matrices (the mass, damping and stiffness matrices) of the mechanical structure using measured forces input and responses data is proposed. This algorithm is based upon the Maximum Likelihood Estimation, so that the accuracy of identified matrices is stable to experimental errors (random errors). After a theoretical formulation is performed, two examples are provided to illustrate and validate this algorithm. One is analytical example which identifies analytically generated data with random noises, and the other experimentarly identified engine/mount system of automobiles.