Font Size: a A A

The Research On Semi-active Vibration Control System Of High-rise Structure

Posted on:2004-10-26Degree:DoctorType:Dissertation
Country:ChinaCandidate:J LiFull Text:PDF
GTID:1102360092980665Subject:Structural engineering
Abstract/Summary:PDF Full Text Request
It has been proved by theory and experiment that structural control as a new technology is a valid method to reduce the seismic disaster. As one method in the structural control, semi-active control combine active control and passive control, and it can approach to the control effectiveness of active control with only little energy, this advantage makes it the developmental tendency of the structural control. But with the strong nonlinear, semi-active control is not perfect in theory and experiment. The research of smart material and control method bring vigor to the development of structural control, it solve the problem of the traditional control theory, so it break new ground for the structural control method.Following points are the achievement of my research: 1. Make clear the relationship between the ductile structure earthquake-resistant design and the vibration control technology. Put forward the semi-active design: procedure of high-rise structure. Through the complete system simulation analysis of a high-rise structure with AVS/D control system and with MR damper control respectively, above-mentioned procedure is proved to be practicable and correct. Moreover, it has referent value to the engineering application of the semi-active control system.2. Put forward an on-line adoption model tracing identification method based on Artificial Neural Network of the time-varying structure. The method presented divides the actual structure into a mechanism model part and a time-varying error model part. The former is created off line on the basis of a computing model, which is in accordance with the actual situation of the structure; the latter is identified on line by using a small-scale network, which employs the system identification ability of the dynamic recurrent network. The simulation results show that this method has such merits like clear concept, high computing precision and properties of a time-carrying structure correctly, and is applicable to the structural vibration control and health monitor.Put forward an on-line adoption parameter tracing identification method using GA aiming at the time-varying structure. This method is based on the adoption tracing identification idea and makes some improvements on the general GA, it uses the mechanism model which is created according to the existing knowledge, and the on-line computation is only to identify the parameter of the error model which in return makes some amendment to the mechanism model. The idea3. above-mentioned reduces highly the search scope and decreases the computation quantity; moreover it meets the real-time request of on-line identification. The method presented has such merits as clear concept, high computing precision and properties of a time-carrying structure correctly, and is applicable to the structural vibration control and health monitor.4. Put forward the multi-population GA, on the basis of this idea conducting the optimal design study of semi-active control actuator (AVS/D actuator and MR damper) in high-rise structure. This method solves the combination optimal problem with multi-goal functions successfully; it can determine the optimal locations and parameters of the actuators simultaneously under a premise of meeting some control design goal. The simulation analysis shows that by using this method we can proceed the optimal design of actuator to meet different vibration control objectives in active or semi-active control system, and the computation time is short and the reliability is high, thus, it is applicable to practical engineering.5. Put forward a self-adjust on-off control law based on GA. This algorithm is based on the self- adjust control idea, the model of the object to be controlled is identified on line, the control law is adjusted according to the object's current mathematical model which brings the system to the optimal control state. In this algorithm, the controller is designed by GA, which overcomes the shortcomings of large computing quantities existing in the instantaneous optim...
Keywords/Search Tags:high-rise structure, AVS/D intelligent control, Artificial neural network, multi-population GA, shaking table test, model identification, parameter identification, control law, location optimization, parameters optimization, time-delay compensation
PDF Full Text Request
Related items