基于小波神经网络辨识的PID神经MRAC研究

提出了一种基于小波神经网络辨识的PID神经网络模型参考自适应控制方法。该方法采用小波神经网络作为辨识器,PID神经网络作为控制器在线调节。由于小波变换具有良好的时频局部特性,神经网络具有强大的非线性映射能力,自学习、自适应等优势,采用规范正交的小波函数作为神经网络的基函数构成小波神经网络,该网络兼有小波函数的紧支性、波动性以及神经网络的非线性映射能力,自学习、自适应能力等优点,仿真结果表明用该方法构成的控制系统收敛速度快,逼近精度高,鲁棒性好,优于一般的BP网络控制。     

BP神经网络在三次有理B样条曲面近似等距面中的应用研究

利用改进的BP神经网络构造三次有理B样条曲面等距面。训练成功后的神经网络可作为描述自由曲面的模型,利用它能够产生任何需要的点。实际证明,该方法简单、有效。     

基于功率反馈的涡轴发动机神经网络PID控制研究

通过对功率平衡关系进行分析,提出了利用功率反馈设计智能神经网络PID控制器的方法。基于BP（Back Propagation）神经网络,将功率信号作为神经网络的输入层信号,并改进了网络权值的学习规则。通过在线整定PID参数,控制器能够根据功率误差信号的变化实时调整控制参数,从而使系统自主寻找到功率平衡点,具有良好的稳态和动态响应特性。仿真结果表明：该方法可以使涡轴发动机在全包线范围内具有理想的控制性能。     

基于神经网络的机械系统可靠性评价模型

机械系统相关失效的存在,严重地削弱了冗余结构的安全作用。从系统层的应力强度—干涉模型出发,通过Monte-Carlo仿真不同冗余系统的各阶失效概率,获得失效数据。利用BP神经网络算法的函数逼近功能、非线性映射功能和容错能力对系统相关失效概率模型进行离散化处理,建立系统失效概率与零件失效数据之间的非线性关系,进而构建基于神经网络的机械系统可靠性评价的参数化模型。利用该模型可以预测系统中的任意阶相关失效概率。实例给出其应用方法并与实验结果进行对比,表明该方法准确可靠。     

离港航空器滑出时间的BP神经网络预测模型

准确地预测离港航空器滑出时间可有效提升机场场面运行效率,降低运行成本。构建基于BP 神经网络的离港航空器滑出时间预测模型,分析影响离港航空器滑出时间的可量化因素,并对其相关性进行检验;通过我国中南某机场2 周实际运行数据对模型进行验证,并以均方根误差、平均绝对误差和平均绝对误差百分比检验预测结果的准确性。结果表明：同时...     

弹道导弹飞行时间的BP神经网络控制方法

时间协同攻击是提高弹道导弹突防能力和作战效能的一种重要手段。由于发动机推力偏差等干扰因素造成导弹的飞行时间偏差较大,为精确控制弹道导弹的飞行时间,运用BP神经网络算法,建立了一种适用于较大飞行时间偏差修正的飞行时间调控模型。该模型用视加速度偏差表征推力偏差等干扰的影响,根据视加速度偏差调整导弹飞行中某一时段的程序角来实现对飞行时间的精确控制。由BP神经网络建立满足飞行时间偏差要求下视加速度与飞行程序角速率、转弯结束时间之间的映射关系。导弹实际飞行时,根据实际视加速度,由BP神经网络在线计算出飞行程序角速率及转弯结束时间,使导弹在干扰条件下仍以预定飞行时间到达目标。最后,通过仿真验证了该模型的有效性和可行性。     

利用BP神经网络对卫星无热敏设备温度的估测

卫星上测温资源有限,只有部分设备有测温点,难以准确获得其他无测温点设备的温度.基于反向传播(BP)神经网络对复杂非线性系统优秀的拟合能力,建立了估测卫星上无测温点设备温度的神经网络,以在轨有测温点设备温度为输入层,以在轨无测温点设备为输出层,并使用卫星热试验获得的星上温度遥测数据和在轨无测温点设备的热电偶温度数据进行训...     

飞行器总体设计中主要参数的神经网络估算

运用近年来迅速发展的神经网络技术,成功构造了飞行器总体设计过程中具有重要作用的总体质量估算网,升力系数估算网,以及阻力系数估算网,结果表明训练后的参数估算网比通常使用的近似估计公式具有更高的精度.同时,根据所涉及问题的数据特点,为了提高对网络的训练精度,对现有MBP算法作了进一步改进,仿真结果证明改进的MBP算法具有更高的训练效率.这一思路和方法可适用于机械及航空航天其它产品的总体概念设计过程.      

Development cost prediction of general aviation aircraft projects with parametric modeling

The study of the development cost of general aviation aircraft is limited by small samples with many cost-driven factors. This paper investigates a parametric modeling method for prediction of the development cost of general aviation aircraft. The proposed technique depends on some principal components, acquired by utilizing P value analysis and gray correlation analysis. According to these principal components, the corresponding linear regression and BP neural network models are established respectively. The feasibility and accuracy of the P value analysis are verified by comparing results of model fitting and prediction. A sensitivity analysis related to model precision and suitability is discussed in detail. Results obtained in this study show that the proposed method not only has a certain degree of versatility, but also provides a preliminary prediction of the development cost of general aviation aircraft.     

Periodic attitude motions along planar orbits in the elliptic restricted three-body problem

A way to improve the accuracy of the three-body problem model is taking into account the eccentricity of primary attractors. Elliptic Restricted Three-Body Problem (ER3BP) is a model for studying spacecraft trajectory within the three-body problem such that the orbital eccentricity of primaries is reflected in it. As the principal cause of perturbation in the employed dynamical model, the primaries eccentricity changes the structure of orbits compared to the ideal Circular Restricted Three-Body Problem (CR3BP). It also changes the attitude behavior of a spacecraft revolving along periodic orbits in this regime. In this paper, the coupled orbit-attitude dynamics of a spacecraft in the ER3BP are exploited to find precise periodic solutions as the spacecraft is considered to be in planar orbits around Lagrangian points and Distant Retrograde Orbits (DRO). Periodic solutions are repetitious behaviors in which spacecraft whole dynamics are repeated periodically, these periodic behaviors are the main interest of this study because they are beneficial for future mission designs and allow delineation of the system’s governing dynamics. Previous studies laid the foundation for spacecraft stability analysis or studying pitch motion of spacecraft in the ER3BP regime. While in this paper, at first, initial guesses for correction algorithms were derived through verified search methods, then correction algorithms were used to refine calculated orbit-attitude periodic behaviors. Periodic orbits and full periodic solutions are portrayed and compared to previous studies and simpler models. Natural periodic solutions are valuable information eventuate in the longer functional lifetime of spacecraft. Since the problem assumption considered in this paper is much closer to real mission conditions, these results may be the means to use natural bounded motions in the actual operational environment.