基于参数识别的截割头纵向随机振动响应优化

为减小纵轴式掘进机截割头的振动，利用Lagrange方法并结合虚拟激励法推导了该型掘进机的纵向运动微分方程，根据Daubechies小波函数多尺度逼近方法和最小二乘法原理，将掘进机在随机激励作用下系统的时变阻尼和时变刚度用多尺度函数的线性组合表示，从而将时变参数识别问题转化为时不变问题。将识别出的刚度和阻尼值作为初始条件，借助鱼群算法以截割头纵向振动的位移响应为目标函数对参数进行了优化。结果表明：参数优化后截割头纵向振动的位移响应的均值下降了18.3%。为验证优化结果的可靠性，利用ADAMS建立了整机的多刚体动力学模型。仿真结果表明，截割头的位移响应变化与理论计算的结果较接近，从而验证了优化结果的有效性。

Vertical Random Vibration Response and Optimization of Cutting Head Based on Parameter Identification

In order to reduce the vibration of a longitudinal roadheader's cutting head, differential equations of motion of the roadheader were derived by using Lagrange equation and pseudo-excitation method. By expanding the time-varying stiffness and damping of the system into linear combination of a series scaling functions based on Daubechies wavelet's compactness and regularization and least-square method, the time-varying parameter identification problem was turned into the problem of identifying invariant parameters. Taking the identifying parameters as the initial conditions, choosing displacement response of the cutting head as an optimal objective and parameters were optimized by means of fish-swarm algorithm. The results show that the displacement response of the cutting head is decreased by 18.3% after optimized. A multi-body dynamic model of the roadheader was simulated in ADAMS for the sake of identifying the conclusion, the model simulation results and the theoretical value are relatively close,which implies validity of the optimized conclusion.