Search

Stochastic activity network (SAN) is a flexible formalism that permits performing reliability and prognostics analysis of complex degrading systems. However, their use for the analysis of high-reliability equipment including fast evolving and degrading mechanisms requires long simulation times and large hardware resources. This paper presents, a field programmable gate array (FPGA)-based architecture that permits performing online Monte Carlo simulations of SAN models of high reliability systems. The synthesis is automated and the result preserves the structure of the models, permitting model inspection, and validation. The architecture runs without the aid of a central processing unit (CPU) and the simulation results are directly recorded into CPU-accessible random access memory (RAM). The design is validated on reliability monitors for an overheating detector and for degrading underground cables. The resource usage and simulation time are compared to the performance of software solutions that are run on parallel processors and large computer resources. Obtained results confirm that the proposed FPGA architecture can be employed for the accelerated runtime reliability analysis of equipment.

A unified evaluation framework for stochastic tools is developed in this paper. Firstly, we provide a set of already existing quantitative and qualitative metrics that rate the relevant aspects of the performance of a stochastic prognosis algorithm. Secondly, we provide innovative guidelines to detect and minimize the effect of side aspects that interact on the algorithms’ performance. Those aspects are related with the input uncertainty (the uncertainty on the data and the prior knowledge), the parametrization method and the uncertainty propagation method. The proposed evaluation framework is contextualized on a Lithium-ion battery Remaining Useful Life prognosis problem. As an example, a Particle Filter is evaluated. On this example, two different data sets taken from NCA aged batteries and two semi-empirical aging models available in the literature fed up the Particle Filter under evaluation. The obtained results show that the proposed framework gives enough details to take decisions about the viability of the chosen algorithm.

The end-of-life event of the battery system of an electric vehicle is defined by a fixed end-of-life threshold value. However, this kind of end-of-life threshold does not capture the application and battery characteristics and, consequently, it has a low accuracy in describing the real end-of-life event. This paper proposes a systematic methodology to determine the end-of-life threshold that describes accurately the end-of-life event. The proposed methodology can be divided into three phases. In the first phase, the health indicators that represent the aging behavior of the battery are defined. In the second phase, the application specifications and battery characteristics are evaluated to generate the end-of-life criteria. Finally, in the third phase, the simulation environment used to calculate the end-of-life threshold is designed. In this third phase, the electric-thermal behavior of the battery at different aging conditions is simulated using an electro-thermal equivalent circuit model. The proposed methodology is applied to a high-energy electric vehicle application and to a high-power electric vehicle application. The stated hypotheses and the calculated end-of-life threshold of the high-energy application are empirically validated. The study shows that commonly assumed 80 or 70% EOL thresholds could lead to mayor under or over lifespan estimations. The iModBatt project has received funding from the European Union’s Horizon 2020 Programme for research and innovation under Grant Agreement No. 770054.

Power cables are critical assets for the reliable operation of the grid. The cable lifetime is generally estimated from the conductor temperature, and associated lifetime reduction. However, these tasks are intricate due to the complex physics-of-failure (PoF) degradation mechanism of the cable. This is further complicated with the different sources of uncertainty that affect the cable lifetime estimation. Generally, simplified or deterministic PoF models are adopted resulting in non-accurate decision-making under uncertainty. In contrast, the integration of uncertainties leads to a probabilistic decision-making process impacting directly on the flexibility to adopt decisions. Accordingly, this paper presents a novel cable lifetime estimation framework that connects data-driven probabilistic uncertainty models with PoF-based operation, and degradation models through Bayesian state-estimation techniques. The framework estimates the cable health state, and infers confidence intervals to aid decision-making under uncertainty. The proposed approach is validated with a case study with different configuration parameters, and the effect of measurement errors on cable lifetime are evaluated with a sensitivity analysis. Results demonstrate that ambient temperature measurement errors influence more than load measurement errors, and the greater the cable conductor temperature the greater the influence of uncertainties on the lifetime estimate.

Power transformers are critical assets for the correct and reliable operation of the power grid. However, the use of power transformers in the context of smart grids creates new challenges for efficient lifetime management and maintenance planning. The use of intermittent sources of energy and dynamic loads increases the sources of uncertainty and causes non-linear operation dynamics. In addition, the increased use of probabilistic forecasting models for the estimation of influential parameters such as temperature or load, influences the uncertainty associated with the transformer lifetime estimation. These variable operation mechanisms influence the operation and lifetime planning of power transformers. Accordingly, this paper presents a novel probabilistic health state estimation framework to improve the lifetime management of power transformers operated in smart grids through the integration of probabilistic forecasting models with Monte Carlo based Bayesian filtering methods.

This chapter presents an FPGA-based SDR platform which serves as a proof-of-concept for cognitive radio techniques. The platform is based on a fully reconfigurable hardware and operates as a 12.8 Mbps RF-to-Ethernet bridge in the Industrial, Scientific, and Medical (ISM) bands of 868 MHz and 2.45 GHz. The data-processing algorithms of the platform are implemented in an FPGA using Xilinx’s System Generator rapid prototyping tool. A MicroBlaze processor is also included to control the dynamic partial reconfiguration of the FPGA for small in-band frequency changes. In order to achieve full-band reconfiguration, a commercial RF front-end and a custom reconfigurable antenna are integrated. Design and implementation details are presented, along with measurement results.

The use of dedicated components as a means of achieving desirable levels of fault tolerance in a system may result in high costs. A cost effective way of restoring failed functions is to use heterogeneous redundancies: components that, besides performing their primary intended design function, can also restore compatible functions of other components. In this paper, we apply a novel design methodology called D3H2 (aDaptive Dependable Design for systems with Homogeneous and Heterogeneous redundancies) to assist in the systematic identification of heterogeneous redundancies, the design of hardware/software architectures including fault detection and reconfiguration, and the systematic dependability and cost assessments of the system. D3H2 integrates parameter uncertainty and criticality analyses to model inexact failure data in dependability assessment. The application to a railway case study is presented with a focus on analysing different reconfiguration strategies as well as types and levels of redundancies.

The design of health assessment applications for the electromechanical actuation system of the aircraft is a challenging task. Physics-of-failure models involve non-linear complex equations which are further complicated at the system-level. Data-driven techniques require run-to-failure tests to predict the remaining useful life. However, components are not allowed to run until failure in the aerospace engineering arena. Besides, when adding new monitoring elements for an improved health assessment, the airliner sets constraints due to the increased cost and weight. In this context, the health assessment of the electromechanical actuation system is a challenging task. In this paper we propose a data-driven approach which estimates the health state of the system without run-to-failure data and limited health information. The approach combines basic reliability theory with Bayesian concepts and obtained results show the feasibility of the technique for asset health assessment.