基于SOM算法和免疫神经网络的飞机燃油系统故障诊断

针对传统的飞机燃油系统故障诊断方法如硬件冗余方法和系统模型检测方法存在的飞机重量限制和难以建立精确数学模型的问题,设计了一种基于SOM算法和BP神经网络的故障诊断模型;首先,建立了系统故障诊断模型并对诊断原理进行了描述,然后,对故障征兆数据进行预处理,即先采用SOM算法进行连续属性离散化处理,再通过粗糙集互信息方法进行属性降维,以减少数据量和提高诊断效率;最后,建立了基于BP神经网络的故障诊断模型,为了进一步提高故障诊断精度,在采用免疫优化算法对BP神经网络故障诊断模型中的各参数即权值和阈值等进行优化的基础上,进一步采用BP反向传播算法进行参数调整,从而得到最终的故障诊断模型。通过飞机燃油系统故障诊断实例仿真实验证明了文中方法能较为精确地实现故障诊断,且与其它方法相比,具有较高的诊断精度和诊断效率,具有较大的优越性。     

基于模型的飞机燃油系统故障诊断系统的设计与实现

燃油系统作为飞机不可或缺的功能和保障系统,对飞机的飞行安全有着重大影响.传统故障诊断方法已经不能满足日益复杂的燃油系统诊断和维护需求.基于模型诊断是为了克服传统诊断方法的缺点而兴起的一项新型的智能推理诊断技术.研究了基于模型的飞机燃油系统故障诊断方法,建立了故障诊断模型,给出了基于模型的燃油系统故障诊断推理策略.设计并...     

基于粒子群优化的可拓神经网络故障诊断方法研究

基于可拓学的故障诊断方法是智能故障诊断领域中较为新颖的研究方向;首先介绍了可拓故障诊断领域的研究现状;然后重点分析了可拓神经网络模型,包括它的结构和故障诊断原理,由于该模型存在参数设置主观、易早熟等问题,进而提出了基于粒子群优化的可拓神经网络模型,该模型以关联度作为测度工具物理意义明确,通过粒子群算法进行参数优化避免算法早熟;最后采用汽轮发电机组振动信号频谱数据进行算法验证,结果表明该算法能够正确诊断出全部故障,且诊断精度高。     

基于RS的提升机故障诊断规则获取方法

在提升机故障诊断中,从包含冗余和不一致信息的数据中获取简单有效的诊断决策规则是一个难题.本文中提出了一种基于粗糙集理论的提升机故障诊断规则获取模型.该模型使用基于粗糙集理论的启发式算法对诊断规则进行简约,从而生成诊断规则集,建立了用于故障诊断的规则库.通过对提升机故障诊断的仿真实例表明,该方法有效地简化了特征参数和诊断规则,减低了诊断成本,提高了故障诊断的准确率.     

基于ATML的D矩阵诊断模型实现

故障诊断是自动测试系统/设备的重要功能,能否快速、准确的隔离故障是影响装备维修效率的重要因素。当今先进自动测试系统采用ATML系列标准,实现信息交换,其核心在于利用XML语言,通过规定的语法和结构描述测试系统、被测设备、测试流程和测试诊断结果等信息。针对故障诊断,标准定义了贝叶斯网络、D矩阵推理、诊断逻辑和故障树等故障诊断模型。其中D矩阵推理模型建立较容易,易于工程实现,被广泛应用。文章采用图形化建模方法建立了测试-故障依赖模型,描述了D矩阵模型建立方法、建立过程、推理规则和推理算法,并以某电台为例介绍了XML语言相关描述方法。最后基于D矩阵对电台测试性进行分析,根据评估结果完善D矩阵内容,优化推理算法,有效提高了电台故障隔离率,降低诊断模糊度。     

涡扇发动机气路故障的信息融合诊断方法

为了提高包线内气路部件故障诊断能力,提出一种加权D-S(Dempster-Shafer)合成理论的故障诊断方法.基于传感器测量值,分别利用扩展卡尔曼滤波算法(EKF)和受限玻尔兹曼-极限学习机(RB-ELM)对主要气路部件性能进行估计,将计算值与基准值的偏差作为基本概率赋值函数.使用飞行状态参数结合混淆矩阵求解子证据体加权系数,最后进行决策级加权融合诊断.通过某型涡扇发动机仿真验证,结果表明与单独使用基于模型和数据驱动的诊断方法相比,融合诊断方法有效地提高了部件故障诊断精度.     

基于非齐次线性方程组的系统级故障最优诊断算法

提出了基于非齐次线性方程组的系统级故障诊断，给出了PMC、BGM、Chwa＆Hakimi、Maleks模型的方程组，并进一步给出了求解最优诊断算法及C＋＋语言算法描述。     

基于极限学习机的航空发动机传感器故障诊断

针对当前应用于航空发动机传感器故障诊断中的基于梯度的传统学习算法多存在参数选择困难、容易陷入局部最小化、过拟合等问题,提出了基于极限学习机(ELM)的航空发动机传感器故障诊断方法。算法只需设置隐含层神经元的个数,能够较好地避免上述问题,缩短故障诊断时间、提升诊断精度。通过仿真试验表明:基于ELM算法所建的航空发动机传感器故障诊断模型要比基于BP神经网络算法所建的模型耗时短且精度高。     

基于专家知识和模糊推理的大科学装置故障诊断方法

针对大科学装置技术综合、结构复杂、系统庞大,在故障诊断方面面临的故障机理不清楚,难以建立精确的数学模型;诊断信息不完整、不精确,难以进行确定性推理;诊断数据受限,无法实现数据驱动等诸多问题。提出了基于专家知识和模糊推理相结合的故障诊断方法和模式匹配算法,通过模糊因子的引入和基于数据库的模糊诊断知识可视化建模方法的使用,解决了故障诊断环节的诸多不确定性问题,形成了面向用户的模糊专家系统故障诊断基础平台,并在某大型激光驱动装置测试验证平台中得到初步应用,实现了电气驱动及控制系统故障的智能诊断。     

基于模型的机载燃油系统故障诊断研究

针对某型号飞机的燃油系统进行故障诊断,应用基于模型的故障诊断算法,采用离散的状态变量描述系统行为和功能,基于一致性算法判定系统当前行为是否正常以及冲突搜索算法对异常进行诊断.诊断过程引入故障可能性概率的函数Rank,避免了传统穷举方法,有效提高了算法的效率.诊断结果能够很好地涵盖单故障以及多故障组合,同时也能满足系统进行实时故障诊断的要求.     

Dimension reduction and its application to model-based exploration in continuous spaces

The sample complexity of a reinforcement-learning algorithm is highly coupled to how proficiently it explores, which in turn depends critically on the effective size of its state space. This paper proposes a new exploration mechanism for model-based algorithms in continuous state spaces that automatically discovers the relevant dimensions of the environment. We show that this information can be used to dramatically decrease the sample complexity of the algorithm over conventional exploration techniques. This improvement is achieved by maintaining a low-dimensional representation of the transition function. Empirical evaluations in several environments, including simulation benchmarks and a real robotics domain, suggest that the new method outperforms state-of-the-art algorithms and that the behavior is robust and stable.     

Monitoring and fault diagnosis of hybrid systems.

Many networked embedded sensing and control systems can be modeled as hybrid systems with interacting continuous and discrete dynamics. These systems present significant challenges for monitoring and diagnosis. Many existing model-based approaches focus on diagnostic reasoning assuming appropriate fault signatures have been generated. However, an important missing piece is the integration of model-based techniques with the acquisition and processing of sensor signals and the modeling of faults to support diagnostic reasoning. This paper addresses key modeling and computational problems at the interface between model-based diagnosis techniques and signature analysis to enable the efficient detection and isolation of incipient and abrupt faults in hybrid systems. A hybrid automata model that parameterizes abrupt and incipient faults is introduced. Based on this model, an approach for diagnoser design is presented. The paper also develops a novel mode estimation algorithm that uses model-based prediction to focus distributed processing signal algorithms. Finally, the paper describes a diagnostic system architecture that integrates the modeling, prediction, and diagnosis components. The implemented architecture is applied to fault diagnosis of a complex electro-mechanical machine, the Xerox DC265 printer, and the experimental results presented validate the approach. A number of design trade-offs that were made to support implementation of the algorithms for online applications are also described.     

IMPROVEMENT OF THE ROBUSTNESS OF MODEL-BASED MEASURING METHODS USING FUZZY LOGIC

In order to provide monitoring and diagnosis of the actual complete process state during both normal operation and accidental conditions, model-based measuring methods are applied as analytical redundancy in addition to or instead of existing hardware redundancies. Regarding the improvement of robustness of classical model-based measuring methods, the combination of model- and It Knowledge-based algorithms in the form of hybrid methods is proposed. This paper presents different kinds of developed hybrid methods which support the classical model-based measuring methods (observer) by fuzzy algorithms. The fuzzy controllers adapt or describe several parameters of the observer algorithm. Certain advantages of hybrid methods in comparison to classical model-based measuring methods are demonstrated. Subject of investigation is the determination of the collapsed and the mixture level within pressure vessels under two-phase flow conditions during accidental depressurizations.     

An Efficient Algorithm for Finding Minimal Overconstrained Subsystems for Model-Based Diagnosis

In model-based diagnosis, diagnostic system construction is based on a model of the technical system to be diagnosed. To handle large differential algebraic imemodels and to achieve fault isolation, a common strategy is to pick out small overconstrained parts of the model and to test these separately against measured signals. In this paper, a new algorithm for computing all minimal overconstrained subsystems in a model is proposed. For complexity comparison, previous algorithms are recalled. It is shown that the time complexity under certain conditions is much better for the new algorithm. This is illustrated using a truck engine model.     

An SE-tree-based prime implicant generation algorithm

Prime implicants/implicates (PIs) have been shown to be a useful tool in several problem domains. In model-based diagnosis (MBD), de Kleer et al. (Proc. AAAI-90) have used PIs to characterize diagnoses. We present a PI generation algorithm which, although based on thegeneral SE-tree-based search framework, is effectively an improvement of aparticular PI generation algorithm proposed by Slagle et al. (IEEE Trans. Comput. 19(4) (1970)). The improvement is achieved via adecomposition tactic which is boosted by the SE-tree-based framework. The new algorithm is also more flexible in a number of ways. We present empirical results comparing the new algorithm to the old one, as well as to current PI generation algorithms.This research was supported in part by a graduate fellowship ARO Grant DAAL03-89-C0031PRI.     

The computation of hitting sets: Review and new algorithms

In model-based diagnosis or other research fields, the hitting sets of a set cluster are usually used. In this paper we introduce some algorithms, including the new BHS-tree and Boolean algebraic algorithms. In the BHS-tree algorithm, a binary-tree is used for the computation of hitting sets, and in the Boolean algebraic algorithm, components are represented by Boolean variables. It runs just for one time to catch the minimal hitting sets. We implemented the algorithms and present empirical results in order to show their superiority over other algorithms for computing hitting sets.     

An algorithm for model-based diagnosis that considers time

This paper presents an algorithm for model-based diagnosis based on the GDE approach introduced by de Kleer and Williams. The algorithm subsumes the current state-of-the-art of this approach such as focusing on the most probable diagnoses and integrating fault models. This paper shows how to make the GDE approach applicable for observations at different time points. This enables an integrated diagnosis of systems with different test vectors as well as the diagnosis of systems containing components with a time-dependent behavior. As an example, it is shown how to model simple electrical circuits which contain fuses with time-dependent behavior. Unlike most of the other diagnostic engines following the GDE approach, the algorithm of this paper does not use an ATMS as a black box module, but rather integrates the necessary tasks directly into the top level. This paper is self-contained and provides precise interfaces for the use of heuristics which can be used to speed up the performance.     

Discrete-event models of quantized systems for diagnosis

The paper concerns the model-based diagnosis of continuous-variable systems whose state can only be measured through a quantizer. The diagnosis is based on the investigation whether the observed discrete input and output sequences are consistent with a discrete-event model of the quantized system. The paper describes a necessary and sufficient condition for the discrete-event model to be suitable for diagnosis. This condition is independent of the specific model form and of the diagnostic algorithm applied to the model. Within a hierarchy of models with increasing accuracy all of which satisfy this modelling requirement the quality of the diagnostic result increases. The results are illustrated by an example.     

刻画基于模型的中心诊断*

虽然对基于模型的诊断存在一系列不同的逻辑定义,但所幸的是存在一个统一的抽象定义,它概括了以往的不同定义.在该定义基础上提出了基于模型的中心诊断的概念.通过刻画基于模型的中心诊断过程,论证了基于模型的中心诊断与本原蕴含/蕴含式的直接关系,从而将其理论结果与ATMS(assumption-based truth maintenance system)这类算法联系起来.进一步指出,对基于一致性中心诊断的刻画仅仅是文中所给出的刻画的一个特殊情形.     

Model-Based Diagnosis and Conditional Logic

The model-based diagnostic approach was first introduced to overcome the limitations of heuristic systems. However, research on model-based systems showed that the model-based diagnosis approaches resort to assumptions that can be viewed as the return, though controlled, of heuristics into diagnostic reasoning. In this paper we focus on diagnosis with component-oriented device models. We argue for the need to represent and reason with these assumptions. We present a conditional logic, DL, That is suitable for diagnostic reasoning and allows us to represent and reason with assumptions.