基于经验模态分解和灰色关联度分析的仿真模型验证方法

Validation of simulation models based on empirical modal decomposition and grey relevance analysis

比较分析仿真输出与参考输出之间的差异是仿真模型验证的一种常用手段。为解决系统输出为非平稳快变数据的仿真模型验证问题,提出了基于经验模态分解和灰色关联分析的验证方法。首先采用经验模态分解方法将仿真输出和参考输出均分解为趋势项和平稳项两部分,然后从位置差异和外形差异两方面刻画趋势项之间的差异,用谱密度差异刻画平稳项之间的差异,再者采用熵权法确定各类差异的权重,最后基于灰色关联度分析综合三类差异得到验证结果。通过实例应用,验证了方法的有效性。

Comparing the simulation output with the reference output is one of the most popular methods to validate simulation models. For a simulation model whose output is the nonstationary fast data, a validation method based on empirical modal decomposition and grey relevance analysis is proposed. The simulation output and reference output are both decomposed into trend items and stationary items. The difference of trend items is depicted by the differences of position and shape, the difference of stationary items is depicted by the difference of spectral density, the weight of each difference is obtained by using the entropy weight, and the three kinds of differences are integrated based on grey relevance analysis to get the validation result. Finally, the effectiveness of the method is verified by an example.