identification error

The TSK fuzzy model is identified based on the SVM and the recursive least square method, and membership function parameters are optimized with the genetic algorithm so that identification error is minimized.

基于支持向量机和递推最小二乘法，辨识出TSK模糊模型，并且通过遗传算法优化隶属度函数参数，最小化辨识误差。

The analysis followed indicates that the identification error decreases with the increasing of stiffness and becomes smaller when the damp is in an appropriate extent ( about 0.1~ 0.3 Nm/ ( rad/ s)).

计算分析表明，随着系统刚度的增大，辨识误差逐渐减小；当系统阻尼在一个比较合理的范围内时（0.1～0.3Nm/（rad/s）左右），该辨识机制具有较小的辨识误差。

The worst case identification error upper bound is analyzed, and an explict error upper bound is obtained when the Lidstone interpolation spline and triangular window are used.

分析了相应的最差情况的辨识误差，并给出了采用Lidstone插值样条算子和三角窗函数时显式最差情况辨识误差上界。