• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2000.06a
• /
• pp.286-290
• /
• 2000

Optimal neuro-control algorithm is extended to the control of a multi-degree-of-freedom structure. An active mass driver(AMD) system on the top roof is used as an exciter. The control signals are made by a multi-layer perceptron(MLP) which is trained by minimizing a sub-optimal performance index. The performance index is a function of both the output responses and the control signals. Structure having nonlinear hysteretic behavior is also trained and controlled by using proposed control algorithm. In training neuro-controller, emulator neural network is not used. Instead, sensitivity-test data are used. Therefore, only one neural network is used for the control system. Both the time delay effect and the dynamics of hydraulic actuator are included in the simulation. Example shows that optimal neuro-control algorithm can be applicable to the multi-degree of freedom structures.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1994.10a
• /
• pp.414-419
• /
• 1994

In recent years, there has been an increasing intest in control of active automotive suspension systems with a goal of improving the ride comfort and safety. Many approaches for these purposes have used linearized models of the suspension's dynamics, allowing the use of linear control theory. However, the linearized model does not well descriibe the actual system behavior which is inherently nonlinear. The object of this study is to develop a neuro controlled active suspension for the ride quality improvement. After obtaining active control law using optimal control theory, we use the artificial neural network to train the neuro controller to learn the relation of road input and control force. Form the numerical results, we found that back propagation learning does show good pattern matching and vertical acceleration of the driver's seat and sprung mass.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 1999.10a
• /
• pp.203-211
• /
• 1999

Threr are several methods in protecting the building structures from dynamic loads such as an earthquake and a wind. Among them applying a control force to the building structure is one of the methods to decrease the vibration. The most important and difficult problem in the active control is to obtain the mathematical model of the building structure with a controller. the effective active controller can be designed from the exact model of the system In this paper the three story test building with an active mass driver is identified experimentally. the system matrices corresponding to the experimental building are found and verified with the experimentally-obtained transfer functions and responses efficiently.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2000.06a
• /
• pp.978-983
• /
• 2000

This paper presents the system identification of a small-scale building model with an active mass driver and the controller design using Matlab program. As the AMD is a mechanical system which has a dynamic characteristic and whose mass can not be neglected compared to that of the building mass, the AMD-building interaction should be included in the controller design. The system identification is carried out for the AMD-building system with two acceleration inputs of the shaking table and the AMD and single acceleration output of the building. The mathematical model for the AMD-building is obtained and compared with the experimental result. The controller is designed based on the mathematical model using the optimal control algorithm of LQG strategy. The experimental results are compared with the numerical results. It is shown that both results are in good agreement in the system identification and the controlled responses.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 2000.10a
• /
• pp.501-508
• /
• 2000

It is demanded to find the dynamic model of a real structure to design a controller. However, as the structure has inherently infinite number of degree-of-freedom, it is impossible to obtain an exact dynamic model of the structure. Instead a reduction model with finite degree-of-freedom is used for the design of a controller. So there exists uncertainty between a real model and a reduction model which causes poor performance of control. All these uncertainties can degrade the control performance and even cause the control instability. Thus, robust control strategy considering the above uncertainties can be an alternative one to guarantee the performance and stability of the control. This study deals with the experimental verification of robust controller design for the active mass driver. $\mu$-synthesis technique is employed as a robust control strategy. Some weights are chosen based on the difference between the initial plant with which the controller is designed and the perturbed plant to be controlled having the actuator uncertainty. The robustness of $\mu$-synthesis technique is compared with the result of LQG strategy, which does not consider the uncertainty.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.2 no.4
• /
• pp.95-102
• /
• 1998

본 연구의 목적은 능동질량 장치를 이용하여 지진하중을 받는 건물모델의 응답을 제어하는 것으로서 실험에 사용된 능동질량 장치는 교류형 서보 모터에 의해 관성력이 건물모델의 응답에 반작용하여 제어를 하게 되는 원리를 이용한 것이다 소형 진동대에 의한 외부하중 묘사 신호처리와 제어력 발생을 위한 장비들이 구축된 실험 모델로써 능동 질량 추진기가 1층 전단형 건물모델 상부에 설치된 해석 모델을 실현하였으며 제어력 산정을 위한 선형 2차 제어 알고리듬은 LabVIEW 프로그램을 사용해서 구현하였다. 건물의 응답과 제어력을 고려해서 제어성능을 검증하였으며 능동 질량 장치를 설치함으러써 공진하중과 지진하중에 대한 건물의 응답이 감소하고 또한 속도피드백 알고리듬이 그 외의 피드백 알고리듬 보다 제어력이 가장 적게 소용되면서 건물의 응답을 감소시키는 것을 실험적으로 파악하였다.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2000.04b
• /
• pp.271-276
• /
• 2000

Optimal neuro-control algorithm is extended to the control of a multi-degree-of-freedom structure. An active mass driver(AMD) system on the top roof used as a controller. The control signals are made by a multi-layer perceptron(MLP) which is trained by minimizing a sub-optimal performance index. The performance index is a function of both the output responses and the control signals. Structure having nonlinear hysteretic behavior is also trained and controlled by using proposed control algorithm. Bothe the time delay effect and the dynamics of hydraulic actuator are included in the simulation. Example shows that optimal neuro-control algorithm can be applicable to the multi-degree of freedom structures.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2002.11b
• /
• pp.676-681
• /
• 2002

System Identification is carried out for a full scale five-story builing to design a vibration controller. Dynamic characteristics such as natural frequencies, damping ratios, and modes are obtained from the input/output information by both sine-sweep method and white noise method. The active mass driver installed on the five floor is applied as external loading to move the building and each floor acceleration is measured and processed for the system identification. The identified building will be experimentally investigated again with viscoelastic dampers installed at inter-stories to obtain the response behavior. Corresponding result will be presented soon.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2002.11a
• /
• pp.368.1-368
• /
• 2002

System Identification is carried out for a full scale five-story holing to design a vibration controller design. Dynamic characteristics such as natural frequencies, damping ratios, and modes are obtained from the input/output informal ion by both sine-sweet method and white noise method. The active mass driver installed on the third floor is applied as external loading to move the building and each floor acceleration is measured and processed for the system identification. (omitted)

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2003.04a
• /
• pp.461-470
• /
• 2003

This paper discusses a time domain controller, LQR, and a frequency domain controller, H₂, for optimal control of civil structures under seismic loads. Numerical simulations are performed on a three-story structure with Active Mass Driver (AMD), which is experimentally identified. Control effectiveness of each controller for the suppression of third floor acceleration responses is investigated when the similar maximum control force is used. Simulation results indicate that LQR is effective for acceleration response reduction while H₂ controller is efficient for utilizing control force.