基于改进灰色模型的仿真数据预测

Prediction of Simulation Data Based on Improved Grey Model

摘要:

为解决仿真中的实时性问题,提出利用灰色预测模型进行仿真数据预测的方式加快仿真速度,并针对灰色模型在仿真数据预测方面实际应用时的问题做出了相应的改进；首先利用三次样条插值对非等时距序列进行插值处理得到等时距的序列；然后改进了灰色模型建模过程中的背景值计算过程；最后在持续的预测过程中结合新陈代谢的思想,弱化旧数据、强化新数据对模型的影响；最后将灰色模型接入到飞机分布式虚拟仿真平台中进行验证,验证结果表明改进后的灰色预测模型性能更好,接入到仿真平台中能加速8.9%,效果明显。

Abstract:

In order to solve the real-time problem in simulation, a method of using grey prediction model to predict simulation data is proposed to speed up the simulation speed, and corresponding improvements have been made to the practical application of grey model in simulation data prediction. Firstly, cubic spline interpolation is used to interpolate non-equal time interval sequences to obtain equal time interval sequences; then, the background value calculation process in the grey model modeling process is improved; finally, the idea of metabolism is combined in the continuous forecasting process to weaken the old data and strengthen the influence of the new data on the model. Finally, the grey model is connected to the aircraft distributed virtual simulation platform for verification. The verification results show that the improved grey prediction model has better performance and can accelerate 8.9% when it is connected to the simulation platform.