基于组合预测模型的飞机刹车系统性能趋势预测分析
|
摘要
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.
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.
引文
1陶梅贞.现代飞机结构设计.西安:西北工业大学出版社,1997Tao Meizhen.Modern aircraft structure design.Xi'an:Northwestern Polytechnical University Press,1997
2左洪福,蔡景,吴昊.航空维修工程学.北京:科学出版社,2011Zuo Hongfu,Cai Jing,Wu Hao.Aviation maintenance engineering.Beijing:Science Press,2011
3朱红,孔德群,钱旭.基于ATD-BP神经网络的页岩气产量预测方法.科学技术与工程,2017;17(31):128-132Zhu Hong,Kong Dequn,Qian Xu.Shale gas production prediction method based on ATD-BP neural network.Science Technology and Engineering,2017;17(31):128-132
4陶文华,袁正波.焦炭质量的DE-BP神经网络预测模型研究.系统仿真学报.2018;30(5):1650-1656Tao Wenhua,Yuan Zhengbo.Study on DE-BP neural network prediction model of coke quality.Journal of System Simulation,2018;30(5):1650-1656
5张宇飞,沈瑶,杨威.利用BP神经网络分析检测网络时序型隐信道.小型微型计算机系统,2018;39(5):871-876Zhang Yufei,Shen Yao,Yang Wei.Using BP neural network analysis to detect network time-series hidden channels.Small Microcomputer System,2018;39(5):871-876
6李萌,包腾飞.基于PSO-GM-MC的某混凝土双曲拱坝监控模型.水电能源科学,2016;34(4):47-50Li Meng,Bao Tengfei.PSO-GM-MC based monitoring model for a concrete double-curvature arch dam.Hydroelectric Energy Science,2016;34(4):47-50
7尚军亮.一种优化的高精度灰色GM(1,1)预测模型.电子与信息学报,2010;32(6):1301-1305Shang Junliang.An optimized high-precision gray GM(1,1)prediction model.Journal of Electronics and Information,2010;32(6):1301-1305
8 Zheng B H,Luo L.Non-equidistance optimum grey model GM(1,1)of requirement analysis of urban construction land and its empirical study.Electronic Journal of Geotechnical Engineering,2015;20(22):1227-1232
9文传军,詹永照.粒子群高斯诱导核模糊C均值聚类算法.科学技术与工程,2018;18(8):78-84Wen Chuanjun,Zhan Yongzhao.Particle swarm optimization gaussinduced kernel fuzzy C-means clustering algorithm.Science Technology and Engineering,2018;18(8):78-84
10吴慰春,陈龙,丁仁凯.基于粒子群算法的主动悬架混合控制策略研究.机械设计与制造,2018;9(1):29-35Wu Weichun,Chen Long,Ding Renkai.Research on active suspension hybrid control strategy based on particle swarm optimization.Mechanical Design and Manufacturing,2018;9(1):29-35
11邓聚龙.灰理论基础.武汉:华中科技出版社,2002Deng Julong.The foundation of grey theory.Wuhan:Huazhong Science and Technology Press,2002
12陈新民,罗国煜.基于经验的边坡稳定性灰色系统分析与评价,岩土工程学报,1999;21(5):638-641Chen Xinmin,Luo Guoyu.Grey system analysis and evaluation of slope stability based on experience.Chinese Journal of Geotechnical Engineering,1999;21(5):638-641
2左洪福,蔡景,吴昊.航空维修工程学.北京:科学出版社,2011Zuo Hongfu,Cai Jing,Wu Hao.Aviation maintenance engineering.Beijing:Science Press,2011
3朱红,孔德群,钱旭.基于ATD-BP神经网络的页岩气产量预测方法.科学技术与工程,2017;17(31):128-132Zhu Hong,Kong Dequn,Qian Xu.Shale gas production prediction method based on ATD-BP neural network.Science Technology and Engineering,2017;17(31):128-132
4陶文华,袁正波.焦炭质量的DE-BP神经网络预测模型研究.系统仿真学报.2018;30(5):1650-1656Tao Wenhua,Yuan Zhengbo.Study on DE-BP neural network prediction model of coke quality.Journal of System Simulation,2018;30(5):1650-1656
5张宇飞,沈瑶,杨威.利用BP神经网络分析检测网络时序型隐信道.小型微型计算机系统,2018;39(5):871-876Zhang Yufei,Shen Yao,Yang Wei.Using BP neural network analysis to detect network time-series hidden channels.Small Microcomputer System,2018;39(5):871-876
6李萌,包腾飞.基于PSO-GM-MC的某混凝土双曲拱坝监控模型.水电能源科学,2016;34(4):47-50Li Meng,Bao Tengfei.PSO-GM-MC based monitoring model for a concrete double-curvature arch dam.Hydroelectric Energy Science,2016;34(4):47-50
7尚军亮.一种优化的高精度灰色GM(1,1)预测模型.电子与信息学报,2010;32(6):1301-1305Shang Junliang.An optimized high-precision gray GM(1,1)prediction model.Journal of Electronics and Information,2010;32(6):1301-1305
8 Zheng B H,Luo L.Non-equidistance optimum grey model GM(1,1)of requirement analysis of urban construction land and its empirical study.Electronic Journal of Geotechnical Engineering,2015;20(22):1227-1232
9文传军,詹永照.粒子群高斯诱导核模糊C均值聚类算法.科学技术与工程,2018;18(8):78-84Wen Chuanjun,Zhan Yongzhao.Particle swarm optimization gaussinduced kernel fuzzy C-means clustering algorithm.Science Technology and Engineering,2018;18(8):78-84
10吴慰春,陈龙,丁仁凯.基于粒子群算法的主动悬架混合控制策略研究.机械设计与制造,2018;9(1):29-35Wu Weichun,Chen Long,Ding Renkai.Research on active suspension hybrid control strategy based on particle swarm optimization.Mechanical Design and Manufacturing,2018;9(1):29-35
11邓聚龙.灰理论基础.武汉:华中科技出版社,2002Deng Julong.The foundation of grey theory.Wuhan:Huazhong Science and Technology Press,2002
12陈新民,罗国煜.基于经验的边坡稳定性灰色系统分析与评价,岩土工程学报,1999;21(5):638-641Chen Xinmin,Luo Guoyu.Grey system analysis and evaluation of slope stability based on experience.Chinese Journal of Geotechnical Engineering,1999;21(5):638-641