基于神经网络的地震模拟振动台补偿算法研究

Research on shaking table compensation algorithm based on neural network

  • 摘要: 地震模拟振动台试验的时滞性会影响系统的试验精度、系统的稳定性和收敛性,甚至造成系统的试验结果发散。以信阳师范大学建筑与土木工程学院地震模拟振动台为例,采用BP神经网络,对地震模拟振动台试验进行时滞性补偿的仿真设计与分析,通过MATLAB软件进行仿真模拟验证其有效性。结果表明,神经网络控制算法对地震模拟振动台时滞性补偿具有显著效果。

     

    Abstract: The time delay of the earthquake simulation shaking table test is known to affect the test accuracy, system stability and convergence, which can even lead to divergence in the test results. Using the earthquake simulation shaking table from College of Architecture and Civil Engineering at Xinyang Normal University as an example, a simulation design and analysis for delay compensation in the earthquake simulation shaking table test were conducted using a BP neural network, with its effectiveness validated through MATLAB software simulation. The results indicated that the significant effect on time delay compensation for the earthquake simulation shaking table was provided by the neural network control algorithm.

     

/

返回文章
返回