Search

Since the mass distribution of behind-armor debris (BAD) is an important measurement to evaluate the level of destruction caused by BAD, this work proposes a new approach for predicting the mass distribution of BAD generated during the normal penetration of variable cross-section explosive formed projectile (EFP) on rolled homogeneous armor (RHA) steel. Few papers considered the variable cross-section characteristic of EFP, and it is an important influencing factor on number of BAD from target and EFP. In this paper, the variable cross-section characteristic of EFP, the strength of EFP, the axial length of mushroom of EFP, the nose shape of the EFP are all considered, besides, the fit factor of the average strain rate and the fit factor of Mott formula are introduced. The number of BAD from target and EFP as well as the mass distribution of BAD from target and EFP, could both be predicted by modified formula on the base of theoretical analysis and numerical simulation, then it is verified by the experiment. The results indicate that: 1) compared with ignoring the variable cross-section characteristic of EFP, the accuracy of the number of BAD from target and EFP could be improved as the variable cross-section characteristic of EFP is considered; 2) the modified Mott formula can describe the mass distribution of more than 90% BAD from target and EFP.

Analyzing the mass of behind-armor debris (BAD) generated by Rolled Homogeneous Armor (RHA) subjected to normal penetration of variable cross-section Explosively Formed Projectile (EFP) is the purpose of this paper. So theoretical analysis, numerical simulation and experimental data are combined to analyze the influence of variable cross-section characteristic on the time history of crater radius. Moreover the relationships between time history of crater radius (as well as mass of BAD) and the thickness of RHA (from 30 mm to 70 mm) and the impact velocity of EFP (1650 m/s to 1860 m/s) are also investigated. The results indicate that: 1) being compared to the variable cross-section characteristic is ignored, the theoretical time history of crater radius is in better agreement with the simulation results when the variable cross-section characteristic is considered; 2) being compared to the other three conditions of plug, the theoretical mass of BAD is in the best agreement with the simulation results when the shape of plug is frustum of a cone and the angle between generatrix and bottom is 45° and the axial length of mushroom is considered. Keywords: Explosively formed projectile, Behind-armor debris, Variable cross-section characteristic, Shape of plug, Axial length of mushroom, Theoretical model

In this article, a method is proposed to conduct a global sensitivity analysis of epistemic uncertainty on both system input and system structure, which is very common in early stage of system development, using Dempster-Shafer theory (DST). In system reliability assessment, the input corresponds to component reliability and system structure is given by system reliability function, cut sets, or truth table. A method to propagate real-number mass function through set-valued mappings is introduced and applied on system reliability calculation. Secondly, we propose a method to model uncertain system with multiple possible structures and how to obtain the mass function of system level reliability. Finally, we propose an indicator for global sensibility analysis. Our method is illustrated, and its efficacy is proved by numerical application on two case studies.

This paper aims to develop a continuous-discrete finite memory observer (CD-FMO) for a class of nonlinear dynamical systems modeled by ordinary differential equations (ODEs) with discrete measurements. The nonlinear systems under consideration are at least locally Lipschitz, which guarantees the existence and uniqueness of solution at each time instant. The proposed nonlinear observer uses a finite number of collected measurements to estimate the system state in the presence of measurement noise. Besides, a one-step prediction algorithm incorporated with an iterative-update scheme is performed to solve the integral problem caused by system nonlinearity, and an analysis of the numerical integration approximation error is given. The properties of estimation performance have been further proved in deterministic case and been analyzed by Monte Carlo simulation in stochastic cases. It is worth noting that the presented method has a finite-time convergence, while most nonlinear observers are usually asymptotically convergent. Another advantage of CD-FMO is that it has no initial value problem. For the application purpose, residuals are generated to implement fault detection cooperated with Cumulative Sum (CUSUM) control charts, while a bank of CD-FMOs is adopted to realize fault isolation for different sensor and actuator faults of the considered nonlinear robotic arm. The robustness and effectiveness of the proposed approach are illustrated via the simulation results.

A probabilistic common-cause failure (PCCF) is any condition or event that causes multiple components fail or malfunction simultaneously with different occurrence probabilities. A system subject to PCCFs is sometimes affected by multiple common causes (CCs) whose relationship may be very complex. This paper proposes an implicit method to model systems subject to PCCFs using Bayesian Network (BN). Three kinds of relationships between CCs are considered: s-independence, s-dependence or mutually exclusion. The proposed method has no limitation on the type of failure distributions of system components. Finally, the proposed method is illustrated by an example computer system.

The purpose of this study is to analyse how the thickness of Rolled Homogeneous Armor (RHA) and impact velocity of an Explosively Formed Projectile (EFP) influence the middle mass behind-armor debris (BAD) when a variable cross-section EFP penetrates RHA normally. Numerical simulation is adopted, the thickness of RHA varies from 10mm to 70mm, and the impact velocity of the EFP varies from 1650m/s to 1860m/s. The results indicate that: (1) when the impact velocity of the EFP is 1650m/s and the thickness of RHA varies from 10mm to 70mm, p1g of the RHA and EFP decreases with increasing H0. The thin target could be used to produce a large proportion of the middle mass BAD from RHA (including BAD from the EFP and BAD from the RHA and EFP). (2) When the impact velocity of the EFP varies from 1650m/s to 1860m/s and the thickness of the RHA is 40mm, p1g of the RHA is less than 50%, p1g of the EFP is more than 70%, and p1g of the RHA and EFP is more than 50%.

Continuous mobile vision is limited by the inability to efficiently capture image frames and process vision features. This is largely due to the energy burden of analog readout circuitry, data traffic, and intensive computation. To promote efficiency, we shift early vision processing into the analog domain. This results in RedEye, an analog convolutional image sensor that performs layers of a convolutional neural network in the analog domain before quantization. We design RedEye to mitigate analog design complexity, using a modular column-parallel design to promote physical design reuse and algorithmic cyclic reuse. RedEye uses programmable mechanisms to admit noise for tunable energy reduction. Compared to conventional systems, RedEye reports an 85% reduction in sensor energy, 73% reduction in cloudlet-based system energy, and a 45% reduction in computation-based system energy.

International audience

International audience

This paper is divided into three parts. First, it introduces the use of random sets for reliability assessment of components with rare failure events. The proposed approach is based on the use of operations defined in the random set framework (expectations, confidence intervals, etc.) to obtain upper and lower bounds and confidence intervals of components reliability without assuming any prior distribution about their lifetimes. Then, instead of using failure probabilities calculated directly from each component’s observation in order to obtain system reliability, we propose to construct pseudo-system observations directly from components observations in order to obtain the interval system reliability. Finally, the proposed approach is applied on the evaluation of reliability of large-scale systems with very large fault trees and censored reliability data by using Monte Carlo resampling procedure. A comparison with classical probabilistic approaches is also done.

In this paper, we introduce the use of random set theory for availability assessment of systems with rare failure events. Instead of using failure probabilities calculated directly from each component's observation, we propose to construct pseudo-system observations directly from components observations. We then applied operations defined in the random set framework (expectations, confidence intervals, etc.) to obtain upper and lower bounds and confidence intervals of system availability without assuming any prior distribution about failures of the system components. Finally, a case study is used to prove the efficiency of the proposed method and a comparison with imprecise probability approach based on Monte-Carlo simulations is also done.

In this paper, we study the existence and uniqueness of the solution for Markov renewal equation (MRE) of a semi-Markov process with countable state space. This method and its proof are based on an iterative scheme. A numerical solution is also given as well as a case study on system reliability assessment.

Current mobile imaging pipelines, provisioned for high quality photography, are ill-suited for wearable vision analytics, due to their high power consumption and privacy concerns, as exemplified by the slow adoption of wearables, such as Google Glass. Rather than constructing incremental improvements, we believe it is necessary to completely redesign a dedicated imaging pipeline for vision analytics. Toward this goal, we study a novel imaging pipeline, revolving around an in-imager analog vision processor that exports a low bandwidth irreversibly encoded signal, generating vision features before analog-to-digital conversion. To produce this signal at low power, we introduce energy-scaling mechanisms into the imager's analog frontend to produce the encoded signal. We use these mechanisms to generate a low-power signal that cannot be used to reconstruct the image, yet suffices as input for vision analytics. This imaging pipeline design will simultaneously achieve privacy and efficiency for continuous mobile vision tasks.

International audience

While many systems are coherent, some systems have non-coherent structure due to their negative feedback loops and redundant loops. A system is non-coherent for a component if it is in a failed state when the component is in a working state, and when the component fails, the system is restored to the non-failed state. In this paper, we propose a straightforward method to evaluate the imprecise reliability of non-coherent systems using binary decision diagrams (BDD) and imprecise probabilities. The system components are categorized into three types according to their coherency inside the studied system: positively coherent, negatively coherent and non-coherent. The reliability of components is represented by intervals. The imprecise system reliability is obtained using operations performed on the different paths of the BDD. A numerical example is also given in order to illustrate the proposed method.