Sensor validation using hardware-based on-line learning neuralnetworks

The objective of this document Is to show the capabilities of parallel hardware-based on-line learning neural networks (NNs). This specific application is related to an on-line estimation problem for sensor validation purposes. Neural-network-based microprocessors are starting to be commercially available. However, most of them feature a learning performed with the classic back-propagation algorithm (BPA). To overcome this lack of flexibility a customized motherboard with transputers was implemented for this investigation, The extended BPA (EBPA), a modified and more effective BPA, was used for the on-line learning, These parallel hardware-based neural architectures were used to implement a sensor failure detection, identification, and accommodation scheme in the model of a night control system assumed to be without physical redundancy in the sensory capabilities. The results of this study demonstrate the potential for these neural schemes for implementation in actual flight control systems of modern high performance aircraft, taking advantage of the characteristics of the extended back-propagation along with the parallel computation capabilities of NN customized hardware     

某型涡扇发动机控制系统传感器故障诊断研究

研究利用卡尔曼滤波器及多重故障假设检验方法来检测某发动机控制系统传感器硬、软故障,并实现故障传感器的输出重构。给出了传感器故障诊断原理及算法,针对传感器常见典型故障进行了故障诊断过程仿真。仿真结果表明,所研究的方法能及时、有效的检测到故障传感器,并对其进行隔离和重构,没有发生误报和漏报现象。      

基于部件跟踪滤波器的解析余度技术

研究一种以发动机部件跟踪滤波器(CTF)为基础的解析余度技术, 它将CTF与故障检测、隔离和适应逻辑进行了有效的综合, 以改进发动机数控系统的可靠性。仿真表明, 本文所设计的解析余度技术, 在传感器无故障时, 机载模型能正确跟踪发动机的变化。当传感器发生故障时, 在不损坏机载模型的情况下, 又能及时、有效地进行硬、软故障的检测、隔离与适应。      

基于人工免疫网络模型的航空发动机传感器故障诊断

提出了一种用于传感器故障诊断的免疫网络,对其结构和特点进行了分析,给出了相应的诊断算法。对传感器典型故障进行了故障诊断仿真,分析了免疫网络能检测出的最小故障偏差水平以及在不同噪声水平下的故障诊断效果。仿真结果表明,所研究的方法能有效检测到故障传感器,并具有良好的灵敏性及抗噪声干扰能力。     

采用一组卡尔曼滤波器检测发动机传感器故障

在发动机全功能数字电子控制系统中,提高传感器工作的可靠性是十分重要的,除了不断对传感器本身的性能加以改进提高外,现在广泛地采用了余度技术。近二十年来对解析余度(Analyt ical Redundancy)进行了广泛的研究,解析余度(AR)方法是基于各状态变量之间存在的解析关系,在系统可观条件下,利用无故障的输出测量值去估计(构造)已故障传感器正常工作状态时的输出信息,从而实现对故障的检测、隔离与重构,保证控制系统具有预定的控制性能。      

采用序列概率比方法检测航空发动机传感器软故障

提出基于卡尔曼滤波和序列概率比方法进行某型涡扇发动机控制系统传感器软故障检测新方法.研究了采用修正的序列概率比方法处理滤波残差,检测传感器软故障;并将该方法与残差加权二乘算法WSSR(Weighted Sum of Squared Residual)检测传感器软故障过程进行了对比.仿真结果表明,序列概率比方法较WSSR法所需决策时间短,适合于航空发动机传感器软故障检测.      

神经元网络在液体火箭发动机健康监控中的应用

含有噪声的、正常和稳定的传感器数据训练 ART2神经元网络 ,用于液体火箭发动机( L RE)故障检测。每个传感器连续窗的功率谱输入 ART2神经元网络进行学习 ,试验学习好的神经网络 ,验证其能否有效地检测出发动机故障以及故障发生时间。传感器数据来自某变推力液体火箭发动机地面试车 RS61。试验结果表明 ,神经元网络显示的故障发生时间与试车后专家分析的故障开始时间相符     

故障诊断的神经网络多重模型自适应方法

将神经网络与故障诊断的多重模型自适应方法相结合,提出了故障诊断的神经网络多重模型自适应方法,并对某型航空发动机控制系统传感器故障进行诊断仿真。仿真表明,该方法能够用来解决具有模型不确定性系统的故障诊断问题,同时,对未知的故障模态具有自学习能力     

歼击机对传感器故障的分散鲁棒容错控制

 从提高控制系统的可靠性及安全性出发,提出了对关联大系统的传感器故障具有完整性的分散鲁棒控制器设计方法,应用鲁棒性指数特性,对歼击机纵向控制系统进行分散鲁棒性设计。这种完整性的设计方法,使得歼击机在传感器正常或故障情况下,都能稳定地飞行并具有参数鲁棒性。     

涡扇发动机气路传感器故障诊断

为了实现对某涡扇发动机传感器故障的在线诊断,提出并设计了1种基于在线贯序极端学习机的故障诊断算法。其核心思想是在定位某传感器故障后,在线建立针对该故障传感器"预学习"的信号重构算法,解决多故障混叠问题。在线信号重构算法以泛化能力指标为判定条件,利用选择策略对算法网络权值进行选择性更新,提高了故障诊断系统的实时性。以某型涡扇发动机为对象开展了传感器故障诊断与重构仿真,结果表明:该算法能够对发动机单、双传感器故障进行准确地诊断与信号重构,且具有良好的实时性。     

基于自联想神经网络的发动机控制系统传感器故障诊断与重构

研究自联想神经网络及其在发动机控制系统传感器故障诊断及重构中的应用。自联想神经网络关键在于特征提取和噪声滤波。综合自联想网络的最优估计与故障诊断 ,自动区分估计误差和传感器故障。仿真结果表明这种方法不需要模型 ,能诊断传感器硬、软故障 ,当发动机性能蜕化时也能提供很好的解析余度。     

基于自联想网络的发动机传感器解析余度技术

本文提出了一种基于自联想神经网络的传感器解析余度技术。在这种网络中,冗余传感器的信息被压缩、重组进入网络的第一部分,网络的第二部分将压缩信息恢复出来。基于数据融合原理,若一个传感器发生故障,其它传感器仍可提供足够的信息代替发生故障的传感器。理论分析和用于涡轴发动机的仿真结果表明,这种特殊结构的自联想网络具有良好的过滤噪声和故障信号的作用,特别适合于用作不易建模的复杂对象的传感器信号重构     

飞机着陆时的容错导航

研究了飞机在微波着陆环境中的传感器容错系统。该系统的主要目的是检测辅助导航设备和机载传感器的故障,并在这些传感器可能发生故障时提供可靠的飞机状态估计值。自动导航和控制系统利用这些状态估计控制和引导飞机按预定路线降落。     

推力矢量飞机自适应控制系统仿真平台研究

研究了具有自修复功能的推力矢量飞机自适应控制系统的结构功能特点,研究了RHO优化控制算法实现在线控制器设计,利用MSLS辨识算法实现在线飞行参数辨识和等价空间算法、传感器信息融合技术和概率统计理论实现FDI算法。并且根据系统各个部分的算法,采用面向对象技术语言VC 6.0和三维图形语言OpenGL开发了仿真平台,利用仿真平台实时演示了飞机存在舵面故障情况下的飞行控制系统运行仿真,解决了飞机飞行过程中存在舵面损伤和气动参数变化的问题,该仿真平台可以根据需求进行飞机故障加载,具备完备的推力矢量飞机自适应控制系统仿真功能。     

Failure Isolation for a Minimally Redundant Inertial Sensor System

The application of two-degree-of-freedom inertial sensors in a minimally redundant strapdown configuration is considered. The potential improvement in reliability which can be achieved by exploiting the failure isolation capability unique to this configuration is evaluated. A unified, statistical approach to the detection and isolation of both hard and soft sensor failures is presented. The effectiveness of this unified approach to FDI in terms of the mean time to detection, the mean time between false alarms, and the accumulated attitude error prior to detection is indicated by simulation results.     

Sensor/actuator failure detection in the Vista F-16 by multiplemodel adaptive estimation

Multiple model adaptive estimation (MMAE) is applied to the Variable-In-flight Stability Test Aircraft (VISTA) F-16 flight control system at a low dynamic pressure flight condition (0.4 M at 20000 ft). Single actuator and sensor failures are addressed first, followed by dual actuator and sensor failures. The system is evaluated for complete or “hard” failures, partial or, “soft” failures, and combinations of hard and soft actuator and sensor failures. Residual monitoring is discussed for single and dual failure scenarios. Performance is enhanced by the application of a modified Bayesian form of MMAE, scalar residual monitoring to reduce ambiguities, automatic dithering where advantageous, and purposeful commands     

THE DECENTRALIZED ROBUST FAULT TOLERANT CONTROL FOR THE SENSOR FAILURE OF THE FIGHTER

ＴＨＥＤＥＣＥＮＴＲＡＬＩＺＥＤＲＯＢＵＳＴＦＡＵＬＴＴＯＬＥＲＡＮＴＣＯＮＴＲＯＬＦＯＲＴＨＥＳＥＮＳＯＲＦＡＩＬＵＲＥＯＦＴＨＥＦＩＧＨＴＥＲＨｕＳｈｏｕｓｏｎｇ；ＪｉｎＬｏｎｇ（Ｄｅｐｔｏｆ．ＡｕｉｏＣｏｎｉｒｏｌ，ＮａｎｊｉｎｇＵｎｉｖｅｒｓ...     

航空发动机控制系统传感器FDIA系统仿真

基于卡尔曼滤波器组,建立了航空发动机控制系统传感器故障诊断系统,实现了对单个传感器故障的检测、隔离与重构(FDIA),确保了发动机控制系统即使在传感器故障发生的情况下,依然能安全可靠地工作.同时利用Simulink建立一个通用的发动机传感器故障诊断仿真平台,在此仿真平台上,对故障的诊断进行仿真验证,并分析了测量噪声对故障诊断系统性能的影响,为发动机在线传感器故障诊断系统的实现提供理论依据.      

A fault-tolerant neural aided controller for aircraft auto-landing

This paper presents a neural-aided controller that enhances the fault tolerant capabilities of a high performance fighter aircraft during the landing phase when subjected to severe winds and failures such as stuck control surfaces. The controller architecture uses a neural controller aiding an existing conventional controller. The neural controller uses a feedback error learning mechanism and employs a dynamic Radial Basis Function neural network called Extended Minimal Resource Allocating Network (EMRAN), which uses only on-line learning and does not need a priori training. The conventional controller is designed using a classical design approach to achieve the desired autonomous landing profile with tight touchdown dispersions called herein as the pillbox. This design is carried out for no failure conditions but with the aircraft being subjected to winds. The failure scenarios considered in this study are: (i) Single faults of either aileron or elevator stuck at certain deflections, and (ii) double fault cases where both the aileron and elevator are stuck at different deflections. Simulation studies indicate that the designed conventional controller has only a limited failure handling ability. However, neural controller augmentation considerably improves the ability to handle large faults and meet the strict touchdown dispersion requirements, thus enlarging the fault-tolerance envelope.     

贝叶斯假设理论检测发动机传感器故障

贝叶斯多重假设检验是将被检测传感器的M个可能状态,作相应M个假设H_i,其先验概率分别为P(H_i)(i=1,2,…,M),故障决策就是从给定观测量M,寻求H_j为真,由贝叶斯风险函数H_i(i=1,2,…,M,i≠j)个假设中的最小值确定最可能发生的假设H_l。