在非线性系统辨识中 ,输入数据往往是非均匀分布的 .对于此类问题 ,正交小波网络处理起来较复杂 .本文采用分组中值法对非均匀分布样本进行均匀化处理 ,使正交小波网络的设计和应用能够在网格系上进行 ;然后 ,采用分频段加权技术以便和系统设计相配合 .最后用于辨识非线性动态系统 ,仿真结果验证了该方法的可行性和有效性 In the nonlinear system identification, the input data are often non-uniform distribution.For such problems, the orthogonal wavelet network is more complex to deal with.In this paper, the method of grouping the median to heterogeneous distribution of sample homogenization process, the orthogonal wavelet network The design and application of this method can be carried out in the grid system.Secondly, the technology of sub-band weighting is used to match the system design.Finally, the method is used to identify the nonlinear dynamic system and the simulation results verify the feasibility and effectiveness of the method