基于组合预测模型的飞机刹车系统性能趋势预测分析

Analysis of Aircraft Braking System Performance Trend Based on Combined Forecasting Model

为了对飞机刹车系统进行性能趋势预测分析,提出一种灰色关联分析确定权重的组合预测方法。首先,利用BP神经网络(back propagation network,BP)对刹车片的累积磨损量进行预测,得到网络输出序列与向后预测序列。对于灰色预测(grey model,GM)模型利用粒子群(particle swarm optimization,PSO)对其优化;用粒子群优化灰色模型(particle swarm optimization-grey model,PSO-GM)进行预测得到拟合序列与向后预测序列。在此基础上对BP网络输出序列、PSO-GM(1,1)拟合序列与原始数据序列进行灰色关联分析,确定组合加权的权重。最后对各预测模型的向后预测序列用灰色关联分析法得到的权重进行组合加权,得到最终的刹车片累积磨损量趋势预测值。仿真结果表明,采用灰色关联分析确定权重的组合预测方法具有比单预测模型更好的趋势预测效果,具有对刹车系统性能趋势预测分析很好的实际应用价值。

In order to analyze the performance state of the aircraft brake system,a grey correlation forecasting method for determining the weights is proposed.First,the back propagation(BP)neural network is used to predict the cumulative wear of brake pads,and the network output sequence and backward prediction sequence are obtained.For the grey prediction grey model(GM)model,the particle swarm particle swarm optimization(PSO)was used to optimize it,and the PSO-GM model was used to predict the fit sequence and the backward prediction sequence.Finally,the backward prediction sequence of each prediction model is weighted by the combination of the weights obtained by the grey correlation analysis method to obtain the final predicted value of the cumulative wear of the brake pads.The simulation results show that the combined forecasting method using grey correlation to determine the weight has a better forecasting effect than the single forecasting model.It has a good practical application value to the braking system performance trend forecast analysis.