考虑位移差矢量性的仿真模型灰色关联验证

Study on Validation of Simulation Models Based on Gray Relational Analysis Considering Displacement Difference Vector

针对将常规灰色关联应用于仿真模型验证时存在一定的风险,提出了一种考虑位移差矢量性的灰色关联度模型。综合考虑了时间序列几何形状的相似性和数值距离的接近性两个属性,通过判断位移差的正负性,引入了惩罚因子,提高了相似性关联的精度。通过指数函数将惩罚因子同时引入灰色关联系数计算中,提高了灰色关联相近性识别的准确性。最后,通过实例仿真验证了考虑位移差矢量性的仿真模型灰色关联验证方法的有效性和可行性。

In order to reduce the risk in application of ordinary gray relational analysis （GRA） into validation of simulation models, a new gray relational analysis model was proposed with consideration of displacement difference vector, which stresses both similarity and nearness between simulation and flight-test time series. The penalty factor was introduced into the new model by judging positive and negative of the displacement difference to improve accuracy of similarity association, and at the same time, the displacement difference by exponential function was also introduced into the grey relational coefficient calculation to improve the accuracy of grey correlation similar identification, so it has higher accuracy and reliability. Finally, a case study was given to show reasonability and validity of the improved model.