Building Vibration Control Force Optimization for Semi-Active Dampers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing semi-active control methods for building structures lack computational efficiency in determining optimal control forces, leading to suboptimal vibration control effects.
Innovation Solution
A method involving constructing a vibration control equation, transforming it into a mixed integer optimization problem using a segmented McCormick inequality, predicting integer variable values, and solving for an optimal control force to adjust damping coefficients of dampers in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing semi-active control methods are used, then the building structure can achieve vibration control without injecting energy, but the computational efficiency is insufficient and the control effect is suboptimal
Solution Approach 1:
The patent segments the control force determination problem into a mixed-integer optimization framework, where the control forces are divided into discrete selection variables and continuous optimization variables. This segmentation enables efficient computational solving while achieving optimal vibration control effects.
Solution Approach 2:
The patent implements real-time dynamic optimization by formulating the control force determination as a time-varying optimization problem that adapts to changing structural responses. The mixed-integer optimization approach dynamically adjusts control forces based on current structural state, achieving both computational efficiency and optimal control performance.
2Reliability
If active control is used, then high flexibility and good control effects are achieved, but a large amount of electrical energy is required resulting in extremely high costs
Solution Approach 1:
The patent employs semi-active control devices that utilize the structure's own vibration energy and relative deformations to generate control forces. The control system modulates these self-generated forces through small amounts of electrical energy, achieving effective vibration control without requiring large energy inputs like active control systems.
Solution Approach 2:
The patent changes the operational parameters of control devices (such as damping coefficients or stiffness) in response to structural vibrations. By dynamically adjusting these parameters based on real-time structural state, the system achieves adaptive vibration control with minimal energy consumption, contrasting with the constant high energy consumption of active control.
3Use of energy by moving object
If passive control is used, then no external energy is required and construction is simple, but the effect on vibration control is limited and flexibility is poor
Solution Approach 1:
The patent transforms passive control devices into semi-active devices by enabling dynamic parameter adjustment. The control devices can actively modulate their characteristics (such as damping or stiffness) in response to structural vibrations, providing adaptive control capability while maintaining the energy efficiency and construction simplicity of passive systems.
Solution Approach 2:
The patent implements parameter changes in the control devices by varying damping coefficients or stiffness values based on structural response. This allows the system to adapt to different vibration conditions and achieve superior control effectiveness compared to fixed-parameter passive control, while still requiring minimal external energy input.
Data Source
AI summary
A method, an application, and a device for determining an optimal control force of a building structure are provided. The method includes constructing a vibration control equation of the building structure based on an external excitation and a controller network for the building structure; constructing a target function and a constraint condition of the controller network at a current moment based on the vibration control equation for the building structure; transforming the target function and the constraint condition of the controller network at the current moment into a mixed integer optimization problem using a segmented McCormick inequality; predicting integer variable values in the mixed integer optimization problem at the current moment using a trained prediction module; and obtaining an optimal control force of the controller network at the current moment by solving the mixed integer optimization problem at the current moment based on the predicted integer variable values.