基于K-均值聚类与粒子群核极限学习机的推力估计器设计
|
摘要
鉴于航空发动机直接推力控制与健康管理需要高精度及高实时性的推力估计器,提出了一种基于K-均值聚类与粒子群优化的核极限学习机推力估计方法。采用K-均值聚类对全工况范围内的测量数据进行聚类,在每一个子类中,通过核极限学习机建立推力估计器,采用粒子群算法对核极限学习机的核参数和惩罚系数进行优化,利用了核极限学习机稳定性好、非线性拟合能力强的特点,实现了对发动机推力的估计。经涡扇发动机台架试车数据训练与测试表明,本推力估计方法平均预测时间为0.27ms,实时性满足机载在线状态评估和直接推力控制需求,且在估计精度上较现有方法存在一定优势。
In order to achieve direct thrust control and health management of aero-engines,the thrust estimator with high precision and high real-time is needed. A design method for thrust estimator was proposed,which was based on K-means clustering and particle swarm optimization(PSO)kernel extreme learning machine(KELM). Firstly,the K-means method was used to cluster the measured data in the whole behavior range. Then the thrust estimator model was established by KELM in each sub-class,and PSO was utilized to search the best kernel parameter and penalty coefficient. The stability and non-linear fitting ability of KELM were fully utilized to estimate engine thrust. Finally,the training and testing results of turbofan engine bench test data show that the average forecast time of thrust estimation method is 0.27 ms,which meets the requirements of direct thrust control and airborne on-line state assessment,and it is more accurate than the existing methods.
In order to achieve direct thrust control and health management of aero-engines,the thrust estimator with high precision and high real-time is needed. A design method for thrust estimator was proposed,which was based on K-means clustering and particle swarm optimization(PSO)kernel extreme learning machine(KELM). Firstly,the K-means method was used to cluster the measured data in the whole behavior range. Then the thrust estimator model was established by KELM in each sub-class,and PSO was utilized to search the best kernel parameter and penalty coefficient. The stability and non-linear fitting ability of KELM were fully utilized to estimate engine thrust. Finally,the training and testing results of turbofan engine bench test data show that the average forecast time of thrust estimation method is 0.27 ms,which meets the requirements of direct thrust control and airborne on-line state assessment,and it is more accurate than the existing methods.
引文
[1]李业波,李秋红,黄向华,等.航空发动机性能退化缓解控制技术[J].航空动力学报,2012,27(4):930-936.
[2]陈恬,孙健国.基于相关性分析和神经网络的直接推力控制[J].南京航空航天大学学报,2005,37(2):183-187.
[3]刘毅男,张胜修,张超.基于核方法的航空发动机推力估计器设计[J].推进技术,2013,34(6):829-835.(LIU Yi-nan,ZHANG Sheng-xiu,ZHANG Chao.Aero Engine Thrust Estimator Design Based on Kernel Method[J]. Journal of Propulsion Technology,2013,34(6):829-835.)
[4] Huang G B. An Insight into Extreme Learning Machines:Random Neurons,Random Features and Kernels[J]. Cognitive Computation,2014,6(3):376-390.
[5]姚彦龙,孙健国.基于神经网络逆控制的发动机直接推力控制[J].推进技术,2008,29(2):249-252.(YAO Yan-long,SUN Jian-guo. Aero Engine Direct Thrust Control Based on Neural Network Inverse Control[J]. Journal of Propulsion Technology,2008,29(2):249-252.)
[6]赵永平,孙健国.最小二乘支持向量回归机在发动机推力估计中的应用[J].航空动力学报,2009,24(6):1420-1425.
[7] Huang G B,Zhu Q Y,Siew C K. Extreme Learning Machine:Theory and Applications[J]. Neurocomputing,2006,70(1–3):489-501.
[8]宋汉强,李本威,张赟,等.基于聚类与粒子群极限学习机的航空发动机推力估计器设计[J].推进技术,2017,38(6):1379-1385.(SONG Han-qiang,LI Ben-wei,ZHANG Yun,et al. Aero-Engine Thrust Estimator Design Based on Clustering and Particle Swarm Optimization Extreme Learning Machine[J]. Journal of Propulsion Technology,2017,38(6):1379-1385.)
[9] Deng C W,Huang G B,Jia X U,et al. Extreme Learning Machines:New Trends and Applications[J]. Science China Information Sciences,2015,58(2):1-16.
[10]裴飞,陈雪振,朱永利,等.粒子群优化核极限学习机的变压器故障诊断[J].计算机工程与设计,2015,(5):1327-1331.
[11] Huang G B,Bai Z,Chi M V. Local Receptive Fields Based Extreme Learning Machine[J]. IEEE Computational Intelligence Magazine,2015,10(2):18-29.
[12] Huang G B, Chen L. Convex Incremental Extreme Learning Machine[J]. Neurocomputing,2007,70(16):3056-3062.
[13] Zhu Q Y,Qin A K,Suganthan P N,et al. Evolutionary Extreme Learning Machine[J]. Pattern Recognition,2005,38(10):1759-1763.
[14] Huang G B,Wang D H,Lan Y. Extreme Learning Machines:a Survey[J]. International Journal of Machine Learning&Cybernetics,2011,2(2):107-122.
[15] Huang G B,Zhu Q Y,Siew C K. Extreme Learning Machine:a New Learning Scheme of Feedforward Neural Networks[C]. Budapest:IEEE International Joint Conference on Neural Networks,2004.
[16] Kennedy J,Eberhart R. Particle Swarm Optimization[C]. Killarney:IEEE International Conference on Neural Networks,1995.
[17]杨锡运,关文渊,刘玉奇,等.基于粒子群优化的核极限学习机模型的风电功率区间预测方法[J].中国电机工程学报,2015,35(S1):146-153.
[18]陈小磊,郭迎清,张书刚.某涡扇发动机智能应急控制系统[J].航空动力学报,2013,28(8):1897-1904.
[19]李秋红,孙健国,王前宇.航空发动机推力估计新方法[J].控制理论与应用,2011,28(2):185-191.
[20]李本威,朱飞翔,宋汉强,等.基于逆跟踪控制的航空发动机气路健康参数估计修正方法研究[J].推进技术,2016,37(5):966-973.(LI Ben-wei,ZHU Feixiang,SONG Han-qiang,et al. Aero-Engine Gas Path Health Parameters Estimation and Correction Method Research Based on Inverse Track Control[J]. Journal of Propulsion Technology,2016,37(5):966-973.)
[21]赵永平,孙健国,王前宇,等.基于K-均值聚类和约简最小二乘支持向量回归机的推力估计器设计[J].航空动力学报,2010,25(5):1177-1183.
[2]陈恬,孙健国.基于相关性分析和神经网络的直接推力控制[J].南京航空航天大学学报,2005,37(2):183-187.
[3]刘毅男,张胜修,张超.基于核方法的航空发动机推力估计器设计[J].推进技术,2013,34(6):829-835.(LIU Yi-nan,ZHANG Sheng-xiu,ZHANG Chao.Aero Engine Thrust Estimator Design Based on Kernel Method[J]. Journal of Propulsion Technology,2013,34(6):829-835.)
[4] Huang G B. An Insight into Extreme Learning Machines:Random Neurons,Random Features and Kernels[J]. Cognitive Computation,2014,6(3):376-390.
[5]姚彦龙,孙健国.基于神经网络逆控制的发动机直接推力控制[J].推进技术,2008,29(2):249-252.(YAO Yan-long,SUN Jian-guo. Aero Engine Direct Thrust Control Based on Neural Network Inverse Control[J]. Journal of Propulsion Technology,2008,29(2):249-252.)
[6]赵永平,孙健国.最小二乘支持向量回归机在发动机推力估计中的应用[J].航空动力学报,2009,24(6):1420-1425.
[7] Huang G B,Zhu Q Y,Siew C K. Extreme Learning Machine:Theory and Applications[J]. Neurocomputing,2006,70(1–3):489-501.
[8]宋汉强,李本威,张赟,等.基于聚类与粒子群极限学习机的航空发动机推力估计器设计[J].推进技术,2017,38(6):1379-1385.(SONG Han-qiang,LI Ben-wei,ZHANG Yun,et al. Aero-Engine Thrust Estimator Design Based on Clustering and Particle Swarm Optimization Extreme Learning Machine[J]. Journal of Propulsion Technology,2017,38(6):1379-1385.)
[9] Deng C W,Huang G B,Jia X U,et al. Extreme Learning Machines:New Trends and Applications[J]. Science China Information Sciences,2015,58(2):1-16.
[10]裴飞,陈雪振,朱永利,等.粒子群优化核极限学习机的变压器故障诊断[J].计算机工程与设计,2015,(5):1327-1331.
[11] Huang G B,Bai Z,Chi M V. Local Receptive Fields Based Extreme Learning Machine[J]. IEEE Computational Intelligence Magazine,2015,10(2):18-29.
[12] Huang G B, Chen L. Convex Incremental Extreme Learning Machine[J]. Neurocomputing,2007,70(16):3056-3062.
[13] Zhu Q Y,Qin A K,Suganthan P N,et al. Evolutionary Extreme Learning Machine[J]. Pattern Recognition,2005,38(10):1759-1763.
[14] Huang G B,Wang D H,Lan Y. Extreme Learning Machines:a Survey[J]. International Journal of Machine Learning&Cybernetics,2011,2(2):107-122.
[15] Huang G B,Zhu Q Y,Siew C K. Extreme Learning Machine:a New Learning Scheme of Feedforward Neural Networks[C]. Budapest:IEEE International Joint Conference on Neural Networks,2004.
[16] Kennedy J,Eberhart R. Particle Swarm Optimization[C]. Killarney:IEEE International Conference on Neural Networks,1995.
[17]杨锡运,关文渊,刘玉奇,等.基于粒子群优化的核极限学习机模型的风电功率区间预测方法[J].中国电机工程学报,2015,35(S1):146-153.
[18]陈小磊,郭迎清,张书刚.某涡扇发动机智能应急控制系统[J].航空动力学报,2013,28(8):1897-1904.
[19]李秋红,孙健国,王前宇.航空发动机推力估计新方法[J].控制理论与应用,2011,28(2):185-191.
[20]李本威,朱飞翔,宋汉强,等.基于逆跟踪控制的航空发动机气路健康参数估计修正方法研究[J].推进技术,2016,37(5):966-973.(LI Ben-wei,ZHU Feixiang,SONG Han-qiang,et al. Aero-Engine Gas Path Health Parameters Estimation and Correction Method Research Based on Inverse Track Control[J]. Journal of Propulsion Technology,2016,37(5):966-973.)
[21]赵永平,孙健国,王前宇,等.基于K-均值聚类和约简最小二乘支持向量回归机的推力估计器设计[J].航空动力学报,2010,25(5):1177-1183.