A new structural reliability index based on uncertainty theory

The classical probabilistic reliability theory and fuzzy reliability theory cannot directly measure the uncertainty of structural reliability with uncertain variables,i.e.,subjective random and fuzzy variables.In order to simultaneously satisfy the duality of randomness and subadditivity of fuzziness in the reliability problem,a new quantification method for the reliability of structures is presented based on uncertainty theory,and an uncertainty-theory-based perspective of classical Cornell reliability index is explored.In this paper,by introducing the uncertainty theory,we adopt the uncertain measure to quantify the reliability of structures for the subjective probability or fuzzy variables,instead of probabilistic and possibilistic measures.We utilize uncertain variables to uniformly represent the subjective random and fuzzy parameters,based on which we derive solutions to analyze the uncertainty reliability of structures with uncertainty distributions.Moreover,we propose the Cornell uncertainty reliability index based on the uncertain expected value and variance.Experimental results on three numerical applications demonstrate the validity of the proposed method.     

Real-time fault detection method based on belief rule base for aircraft navigation system

Real-time and accurate fault detection is essential to enhance the aircraft navigation system’s reliability and safety. The existent detection methods based on analytical model draws back at simultaneously detecting gradual and sudden faults. On account of this reason, we propose an online detection solution based on non-analytical model. In this article, the navigation system fault detection model is established based on belief rule base (BRB), where the system measuring residual and its changing rate are used as the inputs of BRB model and the fault detection function as the output. To overcome the drawbacks of current parameter optimization algorithms for BRB and achieve online update, a parameter recursive estimation algorithm is presented for online BRB detection model based on expectation maximization (EM) algorithm. Furthermore, the proposed method is verified by navigation experiment. Experimental results show that the proposed method is able to effectively realize online parameter evaluation in navigation system fault detection model. The output of the detection model can track the fault state very well, and the faults can be diagnosed in real time and accurately. In addition, the detection ability, especially in the probability of false detection, is superior to offline optimization method, and thus the system reliability has great improvement.     

Measuring reliability under epistemic uncertainty:Review on non-probabilistic reliability metrics

In this paper, a systematic review of non-probabilistic reliability metrics is conducted to assist the selection of appropriate reliability metrics to model the influence of epistemic uncertainty. Five frequently used non-probabilistic reliability metrics are critically reviewed, i.e., evidence-theory-based reliability metrics, interval-analysis-based reliability metrics, fuzzy-interval-analysis-based reliability metrics, possibility-theory-based reliability metrics (posbist reliability) and uncertainty-theory-based reliability metrics (belief reliability). It is pointed out that a qualified reli-ability metric that is able to consider the effect of epistemic uncertainty needs to (1) compensate the conservatism in the estimations of the component-level reliability metrics caused by epistemic uncertainty, and (2) satisfy the duality axiom, otherwise it might lead to paradoxical and confusing results in engineering applications. The five commonly used non-probabilistic reliability metrics are compared in terms of these two properties, and the comparison can serve as a basis for the selection of the appropriate reliability metrics.     

Loopy belief propagation based data association for extended target tracking

The data association problem of multiple extended target tracking is very challenging because each target may generate multiple measurements. Recently, the belief propagation based multiple target tracking algorithms with high efficiency have been a research focus. Different from the belief propagation based Extended Target tracking based on Belief Propagation (ET-BP) algorithm proposed in our previous work, a new graphical model formulation of data association for multiple extended target tracking is proposed in this paper. The proposed formulation can be solved by the Loopy Belief Propagation (LBP) algorithm. Furthermore, the simplified measurement set in the ET-BP algorithm is modified to improve tracking accuracy. Finally, experiment results show that the proposed algorithm has better performance than the ET-BP and joint probabilistic data association based on the simplified measurement set algorithms in terms of accuracy and efficiency. Additionally, the convergence of the proposed algorithm is verified in the simulations.     

Basic belief assignment approximations using degree of non-redundancy for focal element

Dempster-Shafer evidence theory, also called the theory of belief function, is widely used for uncertainty modeling and reasoning. However, when the size and number of focal elements are large, the evidence combination will bring a high computational complexity. To address this issue, various methods have been proposed including the implementation of more efficient combination rules and the simplifications or approximations of Basic Belief Assignments (BBAs). In this paper, a novel principle for approximating a BBA into a simpler one is proposed, which is based on the degree of non-redundancy for focal elements. More non-redundant focal elements are kept in the approximation while more redundant focal elements in the original BBA are removed first. Three types of degree of non-redundancy are defined based on three different definitions of focal element distance, respectively. Two different implementations of this principle for BBA approximations are proposed including a batch and an iterative type. Examples, experiments, comparisons and related analyses are provided to validate proposed approximation approaches.     

An improved α-cut approach to transforming fuzzy membership function into basic belief assignment

In practical applications, pieces of evidence originated from different sources might be modeled by different uncertainty theories. To implement the evidence combination under the Dempster–Shafer evidence theory(DST) framework, transformations from the other type of uncertainty representation into the basic belief assignment are needed. a-Cut is an important approach to transforming a fuzzy membership function into a basic belief assignment, which provides a bridge between the fuzzy set theory and the DST. Some drawbacks of the traditional a-cut approach caused by its normalization step are pointed out in this paper. An improved a-cut approach is proposed, which can counteract the drawbacks of the traditional a-cut approach and has good properties. Illustrative examples, experiments and related analyses are provided to show the rationality of the improved a-cut approach.     

Combination of classifiers with incomplete frames of discernment

The methods for combining multiple classifiers based on belief functions require to work with a common and complete(closed) Frame of Discernment(Fo D) on which the belief functions are defined before making their combination. This theoretical requirement is however difficult to satisfy in practice because some abnormal(or unknown) objects that do not belong to any predefined class of the Fo D can appear in real classification applications. The classifiers learnt using different attributes inform...     

一种结合证据理论和Vague集信息融合方法

针对强冲突下证据理论结果与直觉相悖及Vague集数值获取主观性强等问题,提出一种结合证据理论和Vague集的新方法.该方法将获取证据理论的信度函数转化为Vague集区间值形式,通过求解与理想点的距离进行决策.在进行信度函数到Vague集区间值的转化时,提出一种基于聚焦度递减顺次分配的思想,经分析得出该思想与人的主观思想是一致的,最后举例阐明整个算法过程.     

基于DBN效能拟合的舰艇编队作战效能敏感性分析

针对传统的舰艇编队作战效能分析方法中存在的对数据利用不充分、对数据完整性要求较高的问题,提出了基于深度学习的效能拟合方法。从最具有代表性的敏感性分析方法Sobol指数法入手,利用深度学习方法优越的特征学习能力,基于深度信念网络（DBN）构建了效能拟合网络,结合无监督预训练和有监督调优实现了网络训练和参数优化,构建出效能拟合模型。将产生的数据应用于效能分析模型并与完全数据条件下的效能分析结果进行对比,验证了所提出的效能拟合模型对于不完全数据下的作战系统敏感性分析的有效性。     

Epistemic uncertainty quantification in flutter analysis using evidence theory

Aimed at evaluating the structural stability and flutter risk of the system, this paper manages to quantify epistemic uncertainty in flutter analysis using evidence theory, including both parametric uncertainty and method selection uncertainty, on the basis of information from limited experimental data of uncertain parameters. Two uncertain variables of the actuator coupling system with unknown probability distributions, that is bending and torsional stiffness, which are both described with multiple intervals and the basic belief assignment(BBA) extricated from the modal test of actuator coupling systems, are taken into account. Considering the difference in dealing with experimental data by different persons and the reliability of various information sources, a new combination rule of evidence––the generalized lower triangular matrices method is formed to acquire the combined BBA. Finally the parametric uncertainty and the epistemic uncertainty of flutter analysis method selection are considered in the same system to realize quantification. A typical rudder of missile is selected to examine the present method, and the dangerous range of velocity as well as relevant belief and plausibility functions is obtained. The results suggest that the present method is effective in obtaining the lower and upper bounds of flutter probability and assessing flutter risk of structures with limited experimental data of uncertain parameters and the belief of different methods.