No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
A Model-based Prognostics Methodology for Electrolytic CapacitorsBased on Electrical Overstress Accelerated Aging
Abstract
A remaining useful life prediction methodology for elec-trolytic capacitors is presented. This methodology is based on the Kalman filter framework and an empirical degradation model. Electrolytic capacitors are used in several applications ranging from power supplies on critical avionics equipment to power drivers for electro-mechanical actuators. These devices are known for their comparatively low reliability and given their criticality in electronics subsystems they are a good can-didate for component level prognostics and health manage-ment. Prognostics provides a way to assess remaining use-ful life of a capacitor based on its current state of health and its anticipated future usage and operational conditions. We present here also, experimental results of an accelerated ag-ing test under electrical stresses. The data obtained in this test form the basis for a remaining life prediction algorithm where a model of the degradation process is suggested. This prelim-inary remaining life prediction algorithm serves as a demon-stration of how prognostics methodologies could be used for electrolytic capacitors. In addition, the use degradation pro-gression data from accelerated aging, provides an avenue for validation of applications of the Kalman filter based prognos-tics methods typically used for remaining useful life predic-tions in other applications.
... We studied accelerated degradation under electrical stress (Celaya, Kulkarni, Biswas, & Goebel, 2011;Celaya, Kulkarni, Goebel, & Biswas, 2012) as well as thermal stress (Kulkarni, 2012) in electrolytic capacitors . A preliminary approach to remaining useful life prediction of electrolytic capacitors was presented in (Celaya et al., 2011). ...
... We studied accelerated degradation under electrical stress (Celaya, Kulkarni, Biswas, & Goebel, 2011;Celaya, Kulkarni, Goebel, & Biswas, 2012) as well as thermal stress (Kulkarni, 2012) in electrolytic capacitors . A preliminary approach to remaining useful life prediction of electrolytic capacitors was presented in (Celaya et al., 2011). This paper here builds upon the work presented in the preliminary remaining useful life prediction in (Celaya et al., 2012). ...
... Unscented Kalman Filter (UKF) algorithm is used to track the state of health and the degradation model is used to make predictions of remaining useful life once no further measurements are available. A discussion and physical interpretation of the degradation model is presented (Celaya et al., 2011(Celaya et al., , 2012. ...
The implementation of prognostics methodologies to electrical and electronics components and systems has become essential and critical as these systems find more prominence recently as they replace traditional systems in several critical applications. There are several challenges due to the great variety of components used in a system, the continuous development of new electronics technologies, and a general lack of understanding of how electronics fail. Traditional reliability techniques in electronics tend to focus on understanding the time to failure for a batch of components of the same type. In this work, we discuss degradation in electrolytic capacitors which are part of power supplies and are very crucial in their operation. We discuss our experimental setup and further present our findings related to the degradation observed in these capacitors under accelerated electrical aging under different operating conditions. The understanding of the time-dependent degradation process is critical for the development of model-based prognostics algorithms which provide a continuous condition-based estimation of the remaining useful life of the device under test
... We have implemented this approach on empirical degradation models in our earlier work (J. Celaya et al., 2011a;Kulkarni et al., 2012). Experimental studies are being conducted and first principles based degradation models are being derived using the descriptions mentioned in J. Celaya et al., 2011a;Fife, 2006;MIL-C-62F, 2008). ...
... n time. The implemented prognostics architecture is as shown in Figure 1. We have implemented this approach on empirical degradation models in our earlier work (J. Celaya et al., 2011a;Kulkarni et al., 2012). Experimental studies are being conducted and first principles based degradation models are being derived using the descriptions mentioned in J. Celaya et al., 2011a;Fife, 2006;MIL-C-62F, 2008). Identifying the failure precursors and developing accurate models of degradation/ failure has been the most difficult problem of our research goal. These models depend on known as well as unknown and possibly time-varying wear parameters. Early detection and analysis may lead to better prediction and end of l ...
... Early detection and analysis may lead to better prediction and end of life estimates of the capacitor by tracking and modeling the degradation process. Faults and degra-dations appear as parameter value changes in the model, and this provides the mechanisms for tracking system behavior under degraded conditions (J. Celaya et al., 2011aCelaya et al., , 2011b In the next section we discuss in brief the basics of electrolytic capacitors, their detailed structure and different mechanisms under which the devices degrade. ...
Electrolytic capacitors are used in several applications ranging from power supplies for safety critical avionics equipment to power drivers for electro-mechanical actuators. Past experiences show that capacitors tend to degrade and fail faster when subjected to high electrical or thermal stress conditions during operations. This makes them good candidates for prognostics and health management. Model-based prognostics captures system knowledge in the form of physics-based models of components in order to obtain accurate predictions of end of life based on their current state of health and their anticipated future use and operational conditions. The focus of this paper is on deriving first principles degradation models for thermal stress conditions and implementing Bayesian framework for making remaining useful life predictions. Data collected from simultaneous experiments are used to validate the models. Our overall goal is to derive accurate models of capacitor degradation, and use them to remaining useful life in DC-DC converters.
... Several mechanisms have been reported for describing the early ageing phenomenon in AECs [7]. In general, degradation of the dielectric and the oxide layers performance due to thermal and electrical stresses is the main cause of early ageing [8]. In practice, early ageing is experienced as an increase in the equivalent series resistance (ESR) of the capacitor as well as a decrease in its capacitance [9]. ...
... Some studies have been presented in the literature to monitor the ageing of AECs under stress conditions. In [8] the ageing of electrolytic capacitors used in the DC-DC converters has been investigated under electrical overstress operation conditions. The ageing of these capacitors was monitored over a total of 180 h for an accelerated ageing time through measuring ageing data for ESR every 8-10 h. ...
... In this paper, periodic electrical stress is applied to the capacitors for implementation of accelerating ageing. Here, the proposed method in [8] is utilised in which periodic high-voltage stress is imposed on several capacitors. After applying the stress for a specified duration, they are checked by measuring their ESR to select the capacitors which have reasonable variations of the parameters regarding the accelerated ageing. ...
This study deals with a new approach for real‐time detection of early ageing in DC‐link electrolyte capacitors of DC–DC converters. The method is based on the comparison between the slope of the inductor current derived from a reference dynamic model and that of measurement. All of the normal deviations of the system parameters including gradual ageing of the capacitor and impact of the operating temperature are taken into account in the reference model. Since parameters of the reference model are set in accordance with the datasheets of the components, the reference slope represents normal gradual ageing of the capacitor. Therefore, any early ageing emerges in the measured slope. An accelerated ageing test bench was implemented to assess the proposed method, by which early ageing is emulated by experiments. The results confirm the performance of the method for the early ageing detection.
... The failures in power electronics systems and their consequences have been widely discussed in [7][8][9]. To be more specific, in the literature there exists component-level approaches for prognostics as in [4,5,10] for ECaps and in [11][12][13]. Their proposed methods normally require the failure modes derived from statistical studies, designers, reliability engineers, measurements and Accelerated aGing Tests (AGE). ...
... [9,21]. Similarly, the ECaps have been examined to outline their features [4,5,10]. Henceforth, only the slow parameter variations will be focused on as a type of causes of electronic-systems failure. ...
... Moreover, the degradation effects of the ECaps could be detected by observation of the electrolyte, leakage current and/or increase in the internal pressure [1,10]. ...
This paper proposes a system-level prognostic approach for power electronic systems with slow degradation profiles. Although a model-based approach has been adopted to deal with such multivariable dynamical systems with degradation properties, the forecasting of the Remaining Useful Life (RUL) is independent of prior knowledge of degradation profiles. Thus, this proposition is mainly based on the estimation of the degraded parameters. A robust and well-known technique, the Adaptive Joint Extended Kalman Filter (AJEKF), has been used in previous publications for degradation estimation. Consequently, a deep comprehension of the fault mechanisms of the critical electronic components such as Electrolytic Capacitors (ECaps) and power switching devices such as MOSFETs is needed to define their fault precursors and their degradation behaviors for analytical modeling. The developed forecasting methodology highlights the importance of the historical degradation data in the modeling and estimation stages. The main goal is to increase the reliability of the Prognostics and Health Management (PHM). Thus, this technique has been fully applied to a DC-DC converter used in electric vehicles to forecast its RUL on system-level from component-level basis and the simulation results are then presented.
... Considering the mathematical analysis of the experiments done in [6], [19], the empirical models of capacitance loss and ESR increase are as follows: ...
... where the parameters are numerically calculated from the real experiments [6], [19], and taken as references for the rest of the study. a 1 = 0.016, b 1 = −0.84, a 2 = 21.91, ...
... The aforementioned estimation techniques were simulated until the capacitance value reaches zero based on the degradation parameters of [19]. The following three figures show the states estimation compared to the empirical degradation model: The only affected state is the output capacitor voltage. ...
This paper aims to develop an integrated method-
ology that correlates the slow degradation of the output filter
capacitor with the state of a DC-DC Boost converter used
in electric vehicles (EVs). Thus, an intelligent Model-Based
prognostics technique is applied to predict the Remaining Useful Life (RUL) of this dynamic model with time-variant parameters. In order to capture the smooth degradation as well as reducing the innovation error which might not be detected by the sensors, we proposed an Adaptive Joint Extended Kalman Filter (AJEKF) and an Adaptive Dual Kalman Filter (ADKF) techniques to overcome this issue. The feasibility and effectiveness of the utilized techniques are demonstrated in simulation.
... This case study utilizes the model proposed in [20] to predict the end of life of electrolytic capacitors. The degradation model describes the capacitance loss in percentage terms. ...
... The degradation model describes the capacitance loss in percentage terms. Its differential form with additive Gaussian model noise ω follows Eq. (9), [20]. ...
... The model is shown in Eq. (20). The typical Paris' law parameters utilized in mechanical engineering have been slightly reformulated. ...
This work proposes a fast Monte Carlo method to solve differential equations utilized in model-based prognostics. The methodology is derived from the theory of stochastic calcu- lus, and the goal of such a method is to speed up the computation of the probability density functions describing the evolution of the quantity of interest over time. In the prediction case studies proposed in this paper, the stochastic differential equations describe quantities of in- terest directly or indirectly related to the degradation of a monitored system. The method allows the estimation of the probability density functions by solving the deterministic equation and approximating the stochastic integrals using samples of the model noise. By so doing, the prognostic problem is solved without the Monte Carlo simulation based on Euler’s forward method, which is typically the most time consuming task of the prediction stage. Three differ- ent prognostic scenarios are presented as a proof of concept: (i) life prediction of electrolytic capacitors, (ii) remaining time to discharge of Lithium-ion batteries, and (iii) prognostic of cracked structures under fatigue loading. The paper shows how the method produces proba- bility density functions that are statistically indistinguishable from the distributions estimated with Euler’s forward Monte Carlo simulation. However, the proposed solution is orders of magnitude faster when computing the time-to-failure distribution of the monitored system. The approach may enable complex real-time prognostics and health management solutions with limited computing power.
... Typically, physics-of-failure models of the prognostic candidate component derived from first principles of physics, have been used extensively under this category. There exists an extensive literature employing fatigue models for modelling the initiation and propagation of cracks in structural components (Cadini et al., 2009;Swanson et al., 2000;Zio et al., 2011), model for electrolytic overstress related ageing (Celaya et al., 2011) , usage of Arrhenius equation for prediction of resistance drift in thin film resistors (Kuehl, 2010),usage of physics inspired power model (Maricau et al., 2009) or log-linear model (Lu et al., 1997) for degradation of current drain in CMOS (complementary metal-oxide semi-conductor) and usage of physics-inspired exponential degradation model for aluminum electrolytic capacitors in (Kulkarni et al., 2012). Vachtsevanos et al. (Vachtsevanos, George et al., 2007) have included time-series models such as ARMA, ARIMA, ARMAX etc as model based approaches. ...
... Filter for estimation and prediction process is chosen depending upon the assumptions that may be made about the system and the desired performance (Daigle, M. et al., 2012). Wellknown Kalman filter, an optimal estimator for linear systems has been used for prognostics in (Celaya et al., 2011). Extended Kalman filter (EKF) or unscented Kalman filter may also be used for joint state-parameter estimation as presented in (Plett, 2004) and (Daigle, M. J. et al., 2014) respectively. ...
... and a and b are the model parameters. For example, in (Celaya et al., 2011) percentage capacitance loss data from an observed accelerated test is used as the DM with the associated model parameters being determined through non-linear least square regression and noise variance given by associated regression residuals. In (Saha et al., 2009b), relevance vector machine regression is performed over parametric data collected during ageing tests of batteries to find the representative ageing curves and exponential growth models are fit over them to find suitable decay parameters which in turn, are estimated online for prognostication. ...
This thesis develops the approaches for diagnostics and prognostics of uncertain dynamic systems in Bond Graph (BG) modeling framework. Firstly, properties of Interval Arithmetic (IA) and BG in Linear Fractional Transformation, are integrated for representation of parametric and measurement uncertainties on an uncertain BG model. Robust fault detection methodology is developed by utilizing the rules of IA for the generation of adaptive interval valued thresholds over the nominal residuals. The method is validated in real time on an uncertain and highly complex steam generator system.Secondly, a novel hybrid prognostic methodology is developed using BG derived Analytical Redundancy Relationships and Particle Filtering algorithms. Estimations of the current state of health of a system parameter and the associated hidden parameters are achieved in probabilistic terms. Prediction of the Remaining Useful Life (RUL) of the system parameter is also achieved in probabilistic terms. The associated uncertainties arising out of noisy measurements, environmental conditions etc. are effectively managed to produce a reliable prediction of RUL with suitable confidence bounds. The method is validated in real time on an uncertain mechatronic system.Thirdly, the prognostic methodology is validated and implemented on the electrical electro-chemical subsystem of an industrial Proton Exchange Membrane Fuel Cell. A BG of the latter is utilized which is suited for diagnostics and prognostics. The hybrid prognostic methodology is validated, involving real degradation data sets.
... There is a clear understanding of the underlying degradation process. There exists vast literature such as: fatigue models for modelling initiation and propagation of cracks in structural components [25], electrolytic overstress ageing [26], Arrhenius equation for prediction of resistance drift [27], physics inspired power model [28] or log-linear model for degradation of current drain [29], physics-inspired exponential degradation model for aluminum electrolytic capacitors [30] etc. Given the behavioral model of damage progression, the current SOH is popularly obtained in probabilistic domain with the help of Bayesian estimation techniques. ...
... Filter for estimation and prediction process is chosen based upon the modelling hypothesis and desired performances [35]. Well-known Kalman filter, an optimal estimator for linear systems has been used for prognostics in [26]. Extended Kalman filter (EKF) or unscented Kalman filter may also be used for joint state-parameter estimation as presented in [36] and [37] respectively. ...
... () Dt as an index representing the degradation (change, percentage change etc.) 6 and a and b as the model parameters. In this context, significant works are: obtaining capacitance loss DM using non-linear least square regression [26], relevance vector machine regression performed over ageing tests data [38], DM approximated by a linear part and logarithmic/exponential part [46] and residual based statistical DM [53]. Once the DM has been obtained with acceptable accuracy, recursive Bayesian techniques as discussed previously can be employed to estimate SOH and obtain subsequent RUL predictions. ...
This chapter presents a holistic method to address the issue of health monitoring of system parameters in Bond Graph (BG). The advantages of BGs are integrated with Bayesian estimation techniques for efficient diagnostics and prognostics of faults. In particular, BG in linear fractional transformations (LFT) are used for modelling the global uncertain system and sequential Monte Carlo method based particle filters (PF) are used for estimation of the state of health (SOH) and subsequent prediction of the remaining useful life (RUL). In this work, the method is described with respect to a single system parameter which is chosen as prognostic candidate. The prognostic candidate undergoes progressive degradation and its degradation model is assumed to be known a priori. The system operates in control feedback loop. The detection of degradation initiation is achieved using BG-LFT based robust fault detection technique. The latter forms an efficient diagnostic module. PFs are exploited for efficient Bayesian inference of SOH of the prognostic candidate. Moreover, prognostics is achieved by assessment of RUL in probabilistic domain. The issue of prognostics is formulated as joint state-parameter estimation problem, a hybrid prognostic approach, wherein the fault model is constructed by considering the statistical degradation model of the prognostic candidate. The observation equation is constructed from nominal part of the BG-LFT derived Analytical Redundancy Relations (ARR). Various uncertainties which arise because of noise on ARR based measurements, degradation process, environmental conditions, etc., are effectively managed by PF. This allows the production of effective predictions of the RUL of the prognostic candidate with suitable confidence bounds. The method is applied over a mechatronic system in real time and performance is assessed using suitable metrics.
... Model-based prognostics approaches perform these tasks employing first principles physics models that capture knowledge about the system, its components, and their degradation mechanisms [13,14]. Faults and degradations appear as parameter value changes in the model, and this provides the mechanism for tracking system behavior under degraded conditions [15,16]. We implement the prognostics modeling process illustrated in i.e., electrolytic capacitors that include descriptions of how fault parameters evolve in time, governed by their operating conditions. ...
... Our primary focus in this work is to study and model degradation under thermal overstress conditions. Exposure of the capacitors to temperatures that are greater than the rated operating temperatures (T rated ) results in accelerated aging of the devices [15,28]. Higher ambient storage temperature accelerates the rate of electrolyte evaporation leading to degradation of the capacitance parameter [29] prominently than the ESR. ...
... The deterioration of control surfaces and electromechanical components in aircraft powertrains as a function mechanical loading forces has been a topic of study for some time; examples include: electromechanical actuators (Balaban et al., 2010) and composite wing structures (Gobbato et al., 2012), to name a few. The degradation and failure of electrical components as a function of electrical power loading has also been examined for aircraft components such as batteries (Saha et al., 2009) and power electronics (Celaya et al., 2011). ...
Software-in-the-loop and hardware-in-the-loop testing of failure prognostics and decision making tools for aircraft systems will facilitate more comprehensive and cost-effective testing than what is practical to conduct with flight tests. A framework is described for the offline recreation of dynamic loads on simulated or physical aircraft powertrain components based on a real-time simulation of airframe dynamics running on a flight simulator, an inner-loop flight control policy executed by either an autopilot routine or a human pilot, and a supervisory fault management control policy. The offline testing framework is described for the example of battery charge depletion failure scenarios onboard a prototype electric unmanned aerial vehicle.
... A third-degree regression model, determined to be the best fit by least squares, is proposed for the average capacitance degradation with time. In [54], Here, Co and ESRo are the initial capacitance and ESR, respectively, A and B describe temperature-dependent degradation rates, A0 and B0 are base degradation rates, Ea1 and Ea2 are the activation energies, and κ is the Boltzmann constant. These models are then combined and integrated to obtain variable-temperature capacitance and ESR degradation models. ...
... The wastage of electrolytes increases over time. The performance-degradation model [65], i.e., ∆C value (t c ) = C value (0)−C value (t c ) C value (0) % of C value (capacity of capacitor), is expressed as: ...
A data-driven prediction method is proposed to predict the soft fault and estimate the service life of a DC–DC-converter circuit. First, based on adaptive online non-bias least-square support-vector machine (AONBLSSVM) and the double-population particle-swarm optimization (DP-PSO), the prediction model of the soft fault is established. After analyzing the degradation-failure mechanisms of multiple key components and considering the influence of the co-degradation of these components over time on the performance of the circuit, the output ripple voltage is chosen as the fault-characteristic parameter. Finally, relying on historical output ripple voltages, the prediction model is utilized to gradually deduce the predicted values of the fault-characteristic parameter; further, in conjunction with the circuit-failure threshold, the soft fault and the service life of the circuit can be predicted. In the simulation experiment, (1) a time-series prediction is made for the output ripple voltage using the model proposed herein and the online least-square support-vector machine (OLS-SVM). Comparative analyses of fitting-assessment indicators of the predicted and experimental curves confirm that our model is superior to OLS-SVM in both modeling efficiency and prediction accuracy. (2) The effectiveness of the service life prediction method of the circuit is verified.
... The percentage loss in capacitance is used as a precursor of failure variable and it could be used to build a model of the degradation process. The model relates aging time to the percentage loss in capacitance is described as follows [17], ∆ ...
A remaining useful life prediction model for electrolytic capacitor in DC-DC converter is presented in this paper based on Neural Network method. First, the degradation data of electrolytic capacitor is acquired from mathematical models. Next, the neural network prediction model is established by processing the degradation data of the electrolytic capacitor. Finally, the effectiveness of the developed life prediction model is verified using Matlab/Simulink. This paper provides the basis for the overall life prediction of dc/dc converter.
... A methodology used by Celaya et al. [64] for electrolytic capacitor prognosis is adapted for PEMFC lifespan prediction. The principal steps of this methodology are presented in Fig. 3. ...
This study presents a review of prognostic methods applied to automotive proton exchange membrane fuel cell (PEMFC). PEMFC durability is strongly affected when it is subjected to automotive load cycling (ALC). ALC is normally composed of four operation modes such as start‐up, idle, transient high‐current demand and shutdown. All of these operation modes drastically change the internal variables of the system like temperature, pressure, relative humidity etc. causing degradation of the fuel cell components in a short time. Prognostic methods could be a possible solution to tackle the PEMFC's low durability issue because they allow predicting the remaining useful life of the system in order to apply preventive maintenance plans. Therefore, the objective of this study is to review the prognostic techniques applied to PEMFC under ALC. In the first part of this study, a summary of PEMFC degradation mechanisms caused by ALC is realised based on literature review. In the second part, the prognostic methods review for automotive PEMFCs is carried out and a general synthesis and future challenges are given in the third part of the study.
... To the best of our knowledge, this is the first study that calculates online RUL for CMOS designs. Previous studies have used RC electric components [18] and gate bipolar transistors (IGBT) [19] performing RUL calculation with neural networks that was shown to suffer from local minima [20]. ...
Aging is known to impact electronic systems affecting performance and reliability. However, it has been shown that it also brings benefits for power saving and area optimization. This paper presents highlights of those benefits and further shows how aging effects can be leveraged by novel methods to contribute towards improving hardware oriented security and reliability of electronic circuits. We have demonstrated static power reduction in complex circuits from IWLS05 benchmark suite, reaching a noticeable 7S% of reduction in ten years of operation. In hardware oriented security, a novel aging sensor has been proposed for detection of recycled ICs, measuring discharge time Tdv of the virtual power (VV dd ) network in power-gated designs. This sensor utilizes discharge time of VV dd network through leakage current that is much more sensitive to aging than path delay, exhibiting up to 15.7X increment in 10 years. Furthermore, we show how frequency degradation caused by aging is used for online prediction of remaining useful lifetime (RUL) of electronic circuits. Results show an average RUL prediction deviation of less than 0.1 years. This methodology provides node calculations rather than a mean time to failure (MTTF) of the population. The set of techniques that are presented in this paper takes advantage of aging effects, having a positive impact in various aspects of microelectronic systems.
... Dans un premier temps (chapitre 2 et 3), nous validerons les différents résultats expérimentaux et de vieillissement des travaux de recherche antérieurs réalisés [11], [13]- [15], [19], [20], [22], [24], [25], [58], [70]- [76] sur différents condensateurs électrolytiques afin de déterminer si les différentes lois d'évolution des paramètres électriques ESR et C en fonction de la température et du temps de vieillissement sont généralisables pour tous les condensateurs électrolytiques. Nos résultats permettront de déterminer les différentes lois d'évolution des indicateurs de vieillissement en fonction du temps de vieillissement et l'influence de la température sur ces derniers. ...
Avec l’émergence de nouvelles technologies, nous assistons au développement de techniques permettant d’améliorer la sûreté de fonctionnement et notamment la maintenabilité des convertisseurs statiques d’énergie et de leurs composants. Dans ces systèmes de conversion, les condensateurs électrolytiques, assurant un réseau DC stable, représentent un élément important de la chaine de conversion de l’énergie électrique AC/DC et/ou DC/AC. En fonctionnement, ils sont sujets à des contraintes électriques et environnementales (température ambiante, ondulation de courant, tension appliquée, humidité, vibrations, etc.). Ces condensateurs subissent des réactions d’oxydo-réductions, qui consomment et évaporent de l’électrolyte. La durée de vie du condensateur en est alors affectée. Dès lors, il est intéressant d’estimer l’état de santé de ces composants afin de pouvoir programmer des opérations de maintenance. Il est donc utile d’élaborer des outils permettant d’appliquer une maintenance préventive conditionnelle. C’est dans ce cadre que nous avons effectué ces travaux dont l’objectif est de proposer un système de surveillance de l’état de santé des condensateurs électrolytiques. Nous avons donc développé des modèles d’évolution de leurs indicateurs de vieillissement qui sont les variations de la résistance équivalente série ESR et de la capacité équivalente C. L’algorithme de prédiction se base sur l’évolution de ces indicateurs pour estimer l’état de santé et la durée de vie restante du module de condensateurs. Le système de surveillance en temps réel développé ne comporte pas de capteurs supplémentaires à ceux déjà existants dans les convertisseurs d’énergie considérés et ne nécessite pas d’essais de vieillissement accéléré préalables. Dans ce manuscrit, nous détaillons d’abord la procédure expérimentale de vieillissement accéléré, les différentes étapes de la caractérisation, le processus de vieillissement et les résultats associés. Nous proposons une méthode simple et efficace pour identifier les indicateurs de vieillissement ESR et C en temps réel. L’algorithme proposé est basé sur une mise à l’échelle temporelle et sur un référentiel de contraintes en température et en tension. Par la suite, les résultats de la simulation du système de surveillance et de la prévision de l’état de santé retenu sont présentés. Des essais expérimentaux ont été menés sur des condensateurs intégrés au sein d’un variateur de vitesse industriel de puissance 15 kW. Les algorithmes mis en œuvre et leurs contraintes d'implémentation respectives, pour une application temps réel, sont détaillés
... The key task in model based approaches employ is to derive first principles physics models that capture knowledge about the system, components, and their underlying degradation mechanism [11,32]. Faults and degradations appear as parameter value changes in the model which provide the mechanism for tracking system behavior under degraded conditions [7,25]. The implement prognostics modeling process is illustrated in Figure 3. ...
This book chapter proposes the use of a model-based prognostics approach for electronics components. Components such as electrolytic capacitors, MOSFETs, and IGBTs are critical components in electronics systems in safety critical domains such as aeronautics, medical, etc. These devices are known to have lower reliability than other electronic components that are used in power supplies of avionics equipment and electrical drivers of electromechanical actuators of control surfaces. The field of prognostics for these is concerned with the prediction of Remaining Useful Life (RUL) of the components and systems. This notion of condition-based health assessment leverages the knowledge of the device physics to model the degradation process, which is then used to estimate remaining useful life as a function of current state of health and future operational and environmental conditions.
... A third-degree regression model, determined to be the best fit by least squares, is proposed for the average capacitance degradation with time. In [54], capacitance loss with time under overvoltage conditions is captured by an exponential degradation model and the model parameters are estimated by a nonlinear least-squares regression algorithm. ...
This paper firstly reviews the failure causes, modes and mechanisms of two major types of capacitors used in power electronic systems-metallized film capacitors and electrolytic capacitors. The degradation modeling related to these capacitors is then presented. Both physics-of-failure and data-driven degradation models for reliability and lifetime estimation are discussed. Based on the exhaustive literature review on degradation modeling of capacitors, we provide a critical assessment and future research directions.
... In this chapter, the online estimation of power sources' degradation belongs to joint state-parameter estimation problem, which is typically solved through the usage of a state observer or filter to estimate the changeable state parameters based on the degradation model [192]. Kalman filter as the most well-known filter is used for the prognostics problems in [193] and [194]. But Kalman filter can only be used in linear systems, which is not the case of most prognostic problems. ...
Global warming, environment pollution and exhaustion of petroleum energies have risen their attention of the humanity over the world. Fuel cell hybrid electric vehicle (FCHEV) taking hydrogen as fuel and have zero emission, is thought by public and private organisms as one of the best ways to solve these problems. This PhD dissertation consider a FCHEV with three power sources: fuel cell, battery and supercapacitor, which increases the difficult to design an energy management strategy (EMS) to split the power between the different power sources.Among the EMS available in the current literature, the Equivalent consumption minimization strategy (ECMS) was selected because it allows a local optimization without rely on prior knowledge of driving condition while giving optimal results.Due to low energy density of supercapacitor, its equivalent hydrogen consumption is neglected in most bibliographic references, which not only counter to the aim of minimizing whole hydrogen consumption but also increase the complication of EMS due to the need of an additional EMS to calculate supercapacitor power demand. Thus, a sequential quadratic programming ECMS (SECMS) strategy is proposed to consider energy cost of all three power sources into the objective function. A rule based control strategy (RBCS) and hybrid strategy (HEOS) are also designed in order to to be compared with SECMS. Degradation of energy sources represents a major challenge for the stability of the developed SECMS system. So, based on online estimating state of heath of fuel cell and battery, an adaptive ECMS (AECMS) has been designed through adjusting the equivalent factor and dynamical change rate of fuel cell. The simulation results show that the AECMS can ensure the charge sustenance of battery and the increase of fuel cell durability.To validate the proposed energy management algorithms and the numerical models an exerimental test bench has been built around the real time interface DSPACE. The comparison of the simulation and experimental results showed that the proposed SECMS is operated at around maximum efficiency, supercapacitor supplies peak power, battery works as the energy buffer. It has been proved that the neglect of supercapacitor equivalent hydrogen consumption in ECMS leads to not optimal operation. Compared with RBCS and HEOS, SECMS has least hydrogen consumption and most stable fuel cell current.
... A third-degree regression model, determined to be the best fit by least squares, is proposed for the average capacitance degradation with time. In [54], capacitance loss with time under overvoltage conditions is captured by an exponential degradation model and the model parameters are estimated by a nonlinear least-squares regression algorithm. Here, Co and ESRo are the initial capacitance and ESR, respectively, A and B describe temperature-dependent degradation rates, A0 and B0 are base degradation rates, Ea1 and Ea2 are the activation energies, and κ is the Boltzmann constant. ...
This paper firstly reviews the failure causes, modes and mechanisms of two major types of capacitors used in power electronic systems-metallized film capacitors and electrolytic capacitors. The degradation modeling related to these capacitors is then presented. Both physics-of-failure and data-driven degradation models for reliability and lifetime estimation are discussed. Based on the exhaustive literature review on degradation modeling of capacitors, we provide a critical assessment and future research directions.
... Thus their estimation accuracy could be easily influenced when the installed current sensor couldn't satisfy the specific requirements, such as the measurement bandwidth, frequency response speed, etc. b) ESR and C are both monitored, which can improve the estimation accuracy. According to the degradation analysis of AEC [41][42][43], we can observe that the capacitance always reach the failure limit before ESR. And since many reported methods are based on the ESR estimation, which could lead to erroneous judgments of the capacitor health status [40]. ...
The DC-Link capacitors are important components that tend to be one of the weakest part of power converters. Thus it is necessary to monitor the aging parameters of DC-Link capacitors in valuable power electronic systems. This paper proposes a simple yet effective variable electrical network (VEN) condition monitoring method for DC-Link capacitors in three-phase PWM AC/DC/AC power converters. The capacitance(C) and equivalent series resistance (ESR) of DC-Link capacitors are estimated through the designed VEN unit during the discharging process. This VEN method does not need the current sensor or injecting signals into the control loop. The above property is favorable for the hardware and controller design. Besides, the designed monitoring circuit can be an external unit or a built-in unit which offers a flexible solution for electrical devices. When the designed VEN is used as an external unit, neither the hardware design nor the programs in the controller needs to change, which is practical and economic for electrical devices in use. Experiments are carried out in a 55kW AC/DC/AC inverter with different capacitors. The experiment results prove the method to be effective and accurate.
... Les approches dites "physics-based" ou "model-based" [20,29,108], soit basées sur la physique. Ce sont des approches qui décrivent physiquement les phénomènes qui se produisent [97]. ...
Les préoccupations environnementales actuelles nous amènent à envisager des solutions alternatives, telles que la pile à combustible. Cette dernière malgré ses avantages présente des faiblesses qui ralentissent sa diffusion au sein de l'industrie, entre autres, sa trop courte durée de vie. Face à cette considération, nous proposons d'appliquer le PHM à la PEMFC. Il faut donc développer le pronostic puis considérer son insertion au sein d'un système industriel. Nous choisissons de baser l'approche proposée sur un modèle de comportement, tout en proposant de combler le manque de connaissance concernant le vieillissement de la pile par les données, ce qui nous permet amène à développer une approche hybride. Dans ces travaux, le modèle comportemental est étudié sur des durées de plus en plus grandes pour enfin proposer une prédiction de l'ordre du millier d'heure. Afin de prendre en compte une implantation au sein d'un système réel, une étude sur la généricité et applicabilité de l'approche est réalisée. Ainsi, ces travaux proposent une approche de pronostic hybride basée sur un modèle de comportement et étudie son insertion au sein d'un système réel.
... Numerous attempts have been made in the past to predict the RUL of solid state lighting systems based on reliability analysis and thermal behavior [12][13][14][15][16]. Physics based approaches assume a physical model describing the behavior of the system whereas, data driven approaches completely rely on the measured data to predict future state of a system. Hybrid approaches are basically fusion between the above mentioned two methods and improves the prediction accuracy in the prognostic process by complementing the advantages in each other's method [9][10] [17][18][19][20]. Since the LED lumen degradation is a non-linear process, KF based approaches were not practically feasible to implement. ...
LED based lights used in safety critical applications need continuous monitoring of their light output since lumen degradation is one of many failure modes. Airport Ground Lighting (AGL) is one such application where LED lights are replacing the traditional halogen lamp based lights in approach, runways and taxiways. The question that still remains to be addressed is the prediction of the Remaining Useful Life (RUL) of LED light any time during the service life cycle of the light. Life prediction methods based on reliability analysis and filtering algorithms proposed in the past were predominantly based on Accelerated Life Tests (ALT) and cannot be readily implemented for on-board diagnosis. Here we present a model based prognostic approach that uses a Particle Filtering (PF) to predict the remaining useful life (RUL) for high power white LEDs (HPWLED) that can be readily implemented on-board the lighting system. Lumen maintenance data at different test conditions are used in the PF method to calculate the L70 i.e. time at which lumen output reaches 70% of the initial value. A junction temperature model is developed by mapping the lumen degradation data to the junction temperature. L70 calculated from PF is compared with a standard method recommended by Illumination Engineering Society (IES). Finally the RUL obtained using both the methods were analyzed.
... Normally speaking, such reference value neither exists nor is available to estimate the length of equipment lifetime. Therefore, arbitrary definitions of degradation are produced [3]. ...
Prognostic and health management (PHM) plays a vital role in ensuring the safety and reliability of aircraft systems. The process entails the proactive surveillance and evaluation of the state and functional effectiveness of crucial subsystems. The principal aim of PHM is to predict the remaining useful life (RUL) of subsystems and proactively mitigate future breakdowns in order to minimize consequences. The achievement of this objective is helped by employing predictive modeling techniques and doing real-time data analysis. The incorporation of prognostic methodologies is of utmost importance in the execution of condition-based maintenance (CBM), a strategic approach that emphasizes the prioritization of repairing components that have experienced quantifiable damage. Multiple methodologies are employed to support the advancement of prognostics for aviation systems, encompassing physics-based modeling, data-driven techniques, and hybrid prognosis. These methodologies enable the prediction and mitigation of failures by identifying relevant health indicators. Despite the promising outcomes in the aviation sector pertaining to the implementation of PHM, there exists a deficiency in the research concerning the efficient integration of hybrid PHM applications. The primary aim of this paper is to provide a thorough analysis of the current state of research advancements in prognostics for aircraft systems, with a specific focus on prominent algorithms and their practical applications and challenges. The paper concludes by providing a detailed analysis of prospective directions for future research within the field.
In the modern age of digitalization, electronics are fundamental to any engineering system. With the current strong focus on the Internet of Things (IoT), autonomous vehicles and Industry 4.0, reliable electronics are gaining crucial importance. Predicting the health of complex systems is able to avoid catastrophic failures. Prognostic and Health Monitoring (PHM) approaches are an important step toward trustable and reliable electronics. Nowadays, Artificial Intelligence (AI) and machine learning (ML) algorithms are integrated into PHM approaches, enabling complex fault diagnosis. In this contribution, we provide an overview of the application of intelligent algorithms in PHM of electronics in a systematic manner. The challenges of prognostics in electronics are provided and a detailed overview of the available PHM precursors for various electronic components and the associated selection process is given. Based on the literature review conducted, the main research challenges with ML algorithms in PHM are discussed along with performances of each model.
DC-DC power converters are ubiquitously employed to produce an efficiently regulated voltage to a load that may be either constant or varying, from a source that may or may not be well controlled. DC-DC converters are power conversion circuits that use high-frequency switches and inductors, transformers, and capacitors to filter switching noise into regulated DC voltages. It is necessary to estimate the remaining useful life (RUL) of a power converter during operation to ensure the reliable and safe operation in aerospace, automotive, space and other mission critical applications and to provide early warning of failure for taking a pro-active action(s). This paper considers the effect of multiple components degradation on performance parameters of power converter. This study proposes a RUL prediction model by utilizing a multivariate-LSTM model to relate deviations in several performance parameters to the RUL. The superbuck power converter is used as a case study. This study follows the k-fold cross technique to validate the proposed RUL prediction model. The findings and comparison show that the multivariate-LSTM model is a better RUL predictive model with high prediction accuracy than other similar deep learning models.
Due to their high specific volumetric capacitance, electrolytic capacitors are used in many fields of power electronics, mainly for filtering and energy storage functions. Their characteristics change strongly with frequency, temperature and aging time. Electrolytic capacitors are among the components whose lifetime has the greatest influence on the reliability of electrical systems. Over the past three decades, many efforts in academic research have been devoted to improving reliability capacitor. Industrial applications require more reliable power electronic products. It is in this context that the different electrolytic capacitors and their characteristics are discussed. The aging process of aluminum electrolytic capacitors is explained. Finally, this paper reviews existing methods of failure prognosis of electrolytic capacitors.
The aging detection of dc‐link capacitors has great significance in enhancing the reliability of the power electronic converters in photovoltaic applications. This chapter summarizes the wear‐out failure characteristics of capacitors, including degradation models and end‐of‐life criteria. Generally, two categories of methods are applied to define the end‐of‐life criteria of capacitors. One category is to construct the relationship between electrical and nonelectrical parameters. Another is to identify the failure status of capacitors using the structure change of capacitors. Some electrical parameters and nonelectrical parameters will change with the degradation of capacitors. Based on this, the chapter presents the condition monitoring procedure for capacitors. According to the dependence on the physical model of capacitors, two main categories of principles are generally used to estimate the electrical parameters of dc‐link capacitors. One is the physical model‐based method, and another is the data‐driven‐based method.
A robust recursive zonotopic set-membership approach for remaining useful life forecasting with application to linear parameter-varying systems is proposed in this paper. The proposed approach addresses systems with degraded components formulated as a system-level prognostics problem. Thus, the critical degraded components of the system are augmented to the states resulting a nonlinear system that is reformulated as a linear parameter-varying model. Hence, joint estimation of states and parameters is adopted in a zonotopic set-membership scheme with an optimal linear matrix inequality-based tuning and assuming unknown-but-bounded noises and uncertainties. As a result, a recursive zonotopic set-membership approach is proposed for remaining useful life forecasting based on the prediction of the failure precursors of degraded systems. Finally, this approach is tested on a DC-DC converter case study with unknown degradation behaviors, and the obtained results show the estimation and the forecasting accuracy of this methodology.
Unpredicted failure of barrels is a common problem for cannons, artilleries, and other ballistic missiles. The major cause of the barrels’ failure is high pressure, which generates large residual stresses in the inner layers of the barrel. To reduce the stresses, autofrettage of the barrels is carried out in which the core material is allowed to deform plastically and the upper layers produce a continuous elastic force over the inner layers. Autofrettage is a common engineering practice which is carried out while manufacturing of barrels to increase the operational life significantly. The other reason for barrel failure is material degradation which occurs due to thermal fluctuations, fatigue, wear, erosion, corrosion, etc.In order to correlate them with the life of the gun barrels, various tools such as borescope, variety of gauges, optical bore-mapping, ultrasonic sensors, and diamond indenter-based instruments are used. The measured data is then used to formulate the empirical mathematical relation which is further used to calculate the remaining useful life of the barrels. The other mathematical models are the prognostics models which are generally Probabilistic models or Data-driven models (such as machine learning) or Hybrid models. By using these models, we estimate the remaining life of barrels by using any of the techniques such as heat emission method, dimension increment method, muzzle velocity method, and strain-based methods. The paper concisely summarize various techniques used for diagnostic and prognostic of barrels to estimate the degradation profile and to calculate the remaining useful life (RUL).
A revolution in the power grid driven by distributed energy resources and digitalization is taking place in transmission and distribution networks. Instrument transformers (ITs) are key elements to enable applications for metering, protection and control of modern power grids. Moreover, with the growing complexity of the power system, automation features are becoming more important in coordinating and optimizing grid operation. This increase in complexity will require larger deployment of higher accuracy ITs in the field in the future. Many industry standards exist for various types of IT; however, long-term performance is not covered. Based on available literature in the industry, there is limited knowledge of the long-term performance of capacitive voltage dividers (CVDs). Therefore, an assessment of accuracy over long term aging is critical. The long-term performance of CVDs is mainly linked to the stability of the dielectric material of the primary capacitance. During the service time, there is possibility of systematic drift of the voltage ratio due to aging of the dielectric. The importance of long term accuracy testing is outlined in this paper, identifying testing methodology gaps and reviewing existing methods of aging for other capacitor devices such as metallized film and multilayer ceramic capacitors. The existing aging protocols for capacitor devices and dielectrics are not applicable for high accuracy CVDs, and a new test method is proposed to characterize the long-term accuracy performance. The test method was applied to three high accuracy (0.5 accuracy class) CVD prototypes with different types of epoxy formulations and process methods further indicated as A, B, C in this paper. These prototypes were subjected to an elevated temperature to assess the long-term accuracy performance. Their accuracy has been evaluated in terms of ratio and phase error recorded on regular intervals from the beginning to the end of the test and the results show that a prototype with less variation of thermal coefficient produces a more stable result. In order to understand the impact of the long-term test, Fourier-transform infrared spectroscopy (FTIR) analysis was carried out on aged and fresh samples from the prototypes. It was found that during the aging process the dielectric material undergoes degradation that is caused by an extra hydroxyl group in the epoxy network which leads to an increase in dielectric permittivity, resulting in a significant ratio error. The outcome of the study will provide recommendations for aging tests that should be carried out to get a full understanding of the accuracy behavior of CVDs.
An accurate prediction of remaining useful lifetime (RUL) in high reliability and safety electronic systems is required due to its wide use in industrial applications. In this paper, we propose a novel methodology for online RUL prediction, using support vector regression (SVR) model. Through Cadence simulations with 22nm CMOS technology library, we demonstrate that frequency degradation follows a trackable path and depends on temperature, voltage and aging. This characteristic is exploited for training the SVR model, validated over 20 years of aging degradation. Our methodology is capable of highly accurate RUL estimation, requiring a ring oscillator (RO), temperature sensor and trained SVR software model. Using a supply voltage of 0.9 V and variation in temperature from 0C to 100C, 13 and 21 stage RO show 90% cases with a RUL prediction deviation of 0.2 years, and the remaining between 0.75 and 0.8 years, respectively. Furthermore, with voltage variation from 0.7 to 0.9V, with steps of 0.05V and four representative temperatures (25, 50, 75 and 100 C), the 13-RO shows 52% cases between 0.2 years, 21-RO has 80.5% cases concentrated between 0.2 years of RUL prediction deviation and remaining cases for both ROs are located between 0.8 years.
A review on reliability engineering applications in 4 industrial domains namely electronic, software, nuclear and aerospace from the 2000′s to the present day is compiled. The progress in industrial maintenance activities and Human Reliability Analysis (HRA) linked to these domains are explored. Then, the mathematical aspect of reliability evaluation, in particular multi-state system (MSS), which characterizes the system complexity in these domains are presented. Trends progression are obtained through literature of respective areas as well as available technical information and industrial circulations. Through this review, trend similarities between the mentioned domains and the challenges in reliability science implementation are uncovered. The methods employed in respective industry are explained, with each strength and weakness analysed, together with application examples. This work reveals the role of Prognostic and Health Management (PHM), HRA and analytical methodologies such as MSS and Probabilistic Risk Assessment (PRA) in managing reliability of complex industrial systems. Finally, it stresses on the importance of synergy between these frameworks to ensure a complete reliability assessment in the industry.
Beyond fault diagnosis, addressed in the previous chapter, failure prognostic is a constitutive part of Prognostic and Health Management (PHM) and is of imperative importance for safety critical engineering systems and processes, for the supervision, for automation and condition based maintenance (CBM) of industrial processes, for predictive maintenance, and for all kinds of emerging intelligent autonomously operating mobile systems. Failure prognostic builds on fault diagnosis and needs some knowledge about the degradation process or at least degradation data extracted from measurements at sampling points that can be projected into the future until a predefined failure alarm threshold is reached. This chapter provides a survey of data-driven and model-based approaches to failure prognosis and proposes a new hybrid bond graph model-based, data-driven approach to failure prognosis that does not need to know a mathematical model of the damage process nor assumes a priori that the degradation behaviour follows a certain mathematical function for which the coefficients are to be fitted. Instead, discrete numerical values of the unknown degradation model at sampling points are generated either by parameter estimation on a bicausal bond graph, or are obtained by evaluating ARRs derived from a stage one diagnostic bond graph and from a stage two diagnostic bond graph. The needed equations are set up offline. Both options are illustrated on a small example. The online evaluation of constitutive element equations, or of Analytic Redundancy Relations provides discrete values of a faulty parameter that increasingly deviates from its nominal value with time. The computed values are stored in a buffer for a sliding time window and can be used for simultaneous training of various degradation models. A criterion decides on the best fitting degradation model. Extrapolation of the latter one can be used to compute the intersection with a predefined failure alarm threshold. The learning of a best fitting degradation model and its extrapolation is repeated while the window is moving forward in time so that a sequence of estimates of the remaining useful life (RUL) is obtained. The chapter briefly addresses some uncertainties such as random parameters in degradation models and probability density functions in the prediction of the remaining useful life and concludes with a brief summary of the advantages of the presented hybrid bond graph model-based, data-driven method for failure prognostic.
LED luminaires are common lighting source used in general lighting application for it is energy efficient and long lasting. However, reliability of LED-Luminaire is not well established and no standard method for estimating the performance of LED-luminaire is developed. The paper presents an investigation to understand the performance of LED-luminaire at accelerated degradation test (ADT) condition and analyzes the reliability of LED-luminaire. LED luminaire has two subsystems, LED light engine and the LED constant current driver which are subjected to ADT and the light engine factors - lumen output and Duv changes are analyzed. Similarly, the driver output stage electrolytic capacitor- capacitance and ESR are estimated. The study of continuous operation in time and study when subjected to switching cycles to quantify reliability in terms of operating cycles is performed. All four key parameters - lumen maintenance, Duv, Capacitance and ESR are estimated upto their specified threshold and reliability is analyzed. The LED light engine lumen output is found to degrade very quickly compared to other factors. LED-luminaire failure at ADT is mainly due to lumen depreciation followed by colour-shift and, after significant intervals of time and cycles, the driver capacitor ESR follows as reason for failure of the LED-luminaire. SEM-EDS analysis is performed on the LED package and results show that LED silver (Ag) mirror tarnish is the main reason for significant reduction in light output.
This chapter develops a kernel‐based learning technique to estimate the health degradation of an electronic circuit due to parametric deviation in the circuit components. A model‐based filtering method is developed for predicting the remaining useful life (RUL) of electronic circuit‐comprising components exhibiting parametric faults. The existing approaches for predicting failures resulting from electronic component parametric faults emphasize identifying monotonically deviating parameters and modeling their progression over time. The existing literature is classified and reviewed based on the approach employed for health estimation and failure prediction ‐ either the component‐centric approach or the circuit‐centric approach. The chapter presents the developed first‐principles‐based model to capture the degradation in circuit performance. It discusses the stochastic algorithm used for joint state‐parameter estimation and RUL prediction. The chapter describes the validation results using data obtained from simulation‐based experiments on the critical circuits of a direct‐current (DC)‐DC converter system.
Prognostics and Health Management (PHM) is a method used for predicting reliability of a component or system by assessing its current health and future operating conditions. A physics-of-failure (PoF)-based program on PHM for reliability prediction has been initiated at our institute. As part of the program, we aim at developing PoF-based models for degradation of electronic components and their experimental validation. In this direction, a database on existing PoF models for different electronic components has been prepared. We plan to experimentally determine the model constants and propose suitable methodology for PHM. Electrolytic capacitors are one of the most common passive components which find their applications in devices such as power supplies in aircrafts and printed circuit boards (PCBs) for regulation and protection of a nuclear reactor. Experimental studies have established that electrolytic capacitors degrade under electrical and thermal stress and tend to fail before their anticipated useful life at normal operating conditions. Equivalent series resistance (ESR) and capacitance (C) are the two main parameters used for monitoring health of such capacitors. In this paper, we present an experimental program for thermal and electrical overstress studies towards degradation models for electrolytic capacitors.
Condition monitoring of Aluminum Electrolytic Capacitors (AEC) is essential for predictive maintenance of power electronic converters. AEC is considered at the end of its life when its capacitance or equivalent series resistance (ESR) reaches corresponding critical values. In literature, it is found that either of these parameters may reach its critical limit depending on the operating conditions and applications. However, most of the existing health monitoring techniques of AEC in dcdc converters are based on the estimation of ESR. To address the aforementioned issue, this paper proposes to estimate low frequency impedance of AEC, which is dominated by its capacitance value, thereby allowing health monitoring based on capacitance value. The technique is based on injection of low frequency current ripple into AEC using duty ratio control of the dc-dc converter. The parameters of new and aged capacitors are experimentally obtained at various temperatures and are used to establish the failure criteria. Furthermore, the proposed method is applicable for both continuous conduction mode (CCM) and discontinuous conduction mode (DCM) of operation. For DCM operation, sampling instant to recover low frequency waveform is suggested based on mathematical analysis. Detailed simulation studies are performed and results are included in the paper. Experimentation is carried out on a dc-dc boost converter integrating solar PV with the dc system. Experimental results are found to be in agreement with simulation results.
A remaining useful life prediction algorithm and degradation model for electrolytic capacitors is presented. Electrolytic capacitors are used in several applications ranging from power supplies on critical avionics equipment to power drivers for electro-mechanical actuators. These devices are known for their low reliability and given their criticality in electronics subsystems they are a good candidate for component level prognostics and health management research. Prognostics provides a way to assess remaining useful life of a capacitor based on its current state of health and its anticipated future usage and operational conditions. In particular, experimental results of an accelerated aging test under electrical stresses are presented. The capacitors used in this test form the basis for a remaining life prediction algorithm where a model of the degradation process is suggested. This preliminary remaining life prediction algorithm serves as a demonstration of how prognostics methodologies could be used for electrolytic capacitors.
GPS is the most widely used global navigation satellite system. By design, there is no provision for real time integrity information within the Standard Positioning Service (SPS). However, in safety critical sectors like aviation, stringent integrity performance requirements must be met. This can be achieved externally or at the receiver level through receiver autonomous integrity monitoring (RAIM). The latter is a cost effective method that relies on data consistency, and therefore requires redundant measurements. An external aid to provide this redundancy can be in the form of an Inertial Navigation System (INS). This should enable continued performance even during RAIM holes (when no redundant satellite measurements are available). However, due to the inclusion of an additional system and the coupling mechanism, integrity issues become more challenging. To develop an effective integrity monitoring capability, a good understanding of the potential failure modes of the integrated system is vital. In this paper potential failure modes of integrated GPS/INS systems are identified. This is followed by the specification of corresponding models that would be required to investigate the capability of existing integrity algorithms and to develop enhancements or new algorithms.
1 The estimation of remaining useful life (RUL) of a faulty component is at the center of system prognostics and health management. It gives operators a potent tool in decision making by quantifying how much time is left until functionality is lost. RUL prediction needs to contend with multiple sources of errors like modeling inconsistencies, system noise and degraded sensor fidelity, which leads to unsatisfactory performance from classical techniques like Autoregressive Integrated Moving Average (ARIMA) and Extended Kalman Filtering (EKF). Bayesian theory of uncertainty management provides a way to contain these problems. The Relevance Vector Machine (RVM), the Bayesian treatment of the well known Support Vector Machine (SVM), a kernel-based regression/classification technique, is used for model development. This model is incorporated into a Particle Filter (PF) framework, where statistical estimates of noise and anticipated operational conditions are used to provide estimates of RUL in the form of a probability density function (PDF). We present here a comparative study of the above mentioned approaches on experimental data collected from Li-ion batteries. Batteries were chosen as an example for a complex system whose internal state variables are either inaccessible to sensors or hard to measure under operational conditions. In addition, battery performance is strongly influenced by ambient environmental and load conditions.
This paper presents an integrated approach to switching mode power supply health management that implements techniques from engineering disciplines including statistical reliability modeling, damage accumulation models, physics of failure modeling, and sensor-based condition monitoring using automated reasoning algorithms. Novel features extracted from sensed parameters such as temperature, power quality, and efficiency were analyzed using advanced fault detection and damage accumulation algorithms. Using model-based assessments in the absence of fault indications, and updating the model-based assessments with sensed information when it becomes available provides health state awareness at any point in time. Intelligent fusion of this diagnostic information with historical component reliability statistics provides a robust health state awareness as the basis for accurate prognostic predictions. Complementary prognostic techniques including analysis of projected operating conditions by physics-based component aging models, empirical (trending) models, and system level failure progression models will be used to develop verifiable prognostic models. The diagnostic techniques, and prognostic models have been demonstrated through accelerated failure testing of switching mode power supplies
This paper proposes the experiments and setups for studying diagnosis and prognosis of electrolytic capacitors in DC-DC power converters. Electrolytic capacitors and power MOS-FET's have higher failure rates than other components in DC-DC converter systems. Currently, our work focuses on experimental analysis and modeling electrolytic capacitors degradation and its effects on the output of DC-DC converter systems. The output degradation is typically measured by the increase in Equivalent series resistance and decrease in capacitance leading to output ripple currents. Typically, the ripple current effects dominate, and they can have adverse effects on downstream components. A model based approach to studying degradation phenomena enables us to combine the physics based modeling of the DC-DC converter with physics of failure models of capacitor degradation, and predict using stochastic simulation methods how system performance deteriorates with time. Degradation experiments were conducted where electrolytic capacitors were subjected to electrical and thermal stress to accelerate the aging of the system. This more systematic analysis may provide a more general and accurate method for computing the remaining useful life (RUL) of the component and the converter system.
Switching power supplies have become ubiquitous in electronic modules and systems. From converting power types, power levels, or driving actuators, these power converters embody varying topologies but usually have high switching rates of up to 500 KHz, power devices such as MOSFETs, microelectronic components and a mix of passive components that store and release energy. They are complex modules that have an unfortunate history of observed high failure rates, yet they may be required to support critical systems.Electronic prognostics/health management, or ePHM, provides the ability to predict the failure of the electronic before it actually occurs. This paper will examine the considerations associated with key system- and device level prognostics that designers need to evaluate when “PHM-enabling” their power system.
This paper presents a prognostics approach which detects the performance degradation of multilayer ceramic capacitors under temperature-humidity-bias conditions, and then predicts remaining useful life. In the tests, three performance parameters (capacitance, dissipation factor and insulation resistance) were monitored in-situ. By comparing the predicted results with the experimental results, the prognostics approach provided advanced warning of failures for capacitors and predicted their remaining useful life.
The breakdown phenomena during formation of non-porous anodic films on Al, Ta, Nb and Zr are investigated. The effect of the current density,the composition and concentration of the electrolyte and the nature of the anodized metal on the features of electric breaking down and on the value of the breakdown voltage is studied. New data on the breakdown behaviour are obtained alongside with additional information for some well-known breakdown characteristics. The results are discussed in the light of the theory of electric breakdowns during anodization.
The anodic formation of barrier oxide films on valve metals can be carried on till the electrode potential has attained the “breakdown voltage” (UB, after which a sustained sparking appears. The systematization of the available data on this phenomenon revealed that UB depends fundamentally on the nature of the anodized metal, as well as on the composition and resistivity of the electrolyte. Many other factors like the current density, the surface topography of the electrode, the history of the film formation, etc do not affect UB noticeably. This phenomenology of the breakdowns, as well as their appearance, was explained by a breakdown model. According to this model, sparking is considered as an avalanching in the bulk of the anodic film, the initial electrons being injected into the film from the electrolyte. The peculiar features of the sparking were found to arise both from the specific injecting behaviour of the electrolyte and from the simultaneous oxide film formation.
The dielectric breakdown field of anodic oxide layers on aluminum is independent of thickness in the region 230- 1700 Å and is close to the forming field.
Purpose – This paper seeks to present a prognostics approach using the Mahalanobis distance (MD) method to predict the reliability of multilayer ceramic capacitors (MLCCs) in temperature-humidity-bias (THB) conditions. Design/methodology/approach – Data collected during THB testing of 96 MLCCs were analyzed using the MD method. In the THB tests, three parameters (capacitance (C), dissipation factor (DF), and insulation resistance (IR)) were monitored in situ. A Mahalanobis space (MS) was formed from the MD values of a set of non-failed MLCCs. MD values for the remaining MLCCs were compared with an MD threshold. Data for MLCCs which exceeded the threshold were examined using the failure criteria for the individual electrical parameters to identify failures and precursors to failure. Findings – It was found that the MD method provided an ability to detect failures of the capacitors and identify precursors to failure, although the detection rate was not perfect. Research limitations/implications – It was observed that the quality and construction of the MS, together with the choice of the MD threshold, were the critical factors determining the sensitivity of the MD method. Recommendations are offered for improved sensitivity to enable assessment of intermittent failures. Originality/value – MD analysis of the multivariate MLCC data set illustrates how detection of failures can be simplified in a system for which several parameters were monitored simultaneously. This makes the MD method of great potential value in a health-monitoring system.
This paper proposes a model based approach for prognosis of DC-DC power converters. We briefly review the prognosis process, and present an overview of different approaches that have been developed. We study the effects of ca-pacitor degradation on DC-DC converter perfor-mance by developing a combination of a thermal model for ripple current effects and a physics of failure model of the thermal effects on capaci-tor degradation. The derived degradation model of the capacitor is reintroduced into the DC-DC converter model to study changes in the system performance using Monte Carlo methods. The simulation results observed under different con-ditions and experimental setups for model verifi-cation are discussed. The paper concludes with comments and future work to be done.
Understanding the ageing mechanisms of electronic components critical avionics systems such as the GPS and INAV are of critical importance. Electrolytic capac-itors and MOSFET's have higher failure rates among the components of DC-DC power converter systems. Our current work focuses on analyzing and modeling elec-trolytic capacitor degradation and its effects on the out-put of DC-DC converter systems. The output degrada-tion is typically measured by an increase in ESR (Equiv-alent Series Resistance) and decrease in the capacitance value over long periods of use even under nominal oper-ating conditions. Typically the primary effect of degra-dation is increased ripple current and this has adverse effects on downstream components. For example, in avionics systems where the power supply drives a GPS unit, ripple currents can cause glitches in the GPS posi-tion and velocity output, and this may cause errors in the Inertial Navigation (INAV) system, causing the aircraft to fly off course. In this paper, we present the details of our ageing methodology along with details of experi-ments and analysis of the results.
This paper outlines a non-invasive method for the detection of failure precursors for optical isolators used in switch-mode power supplies (SMPS). The method relies on the transfer characteristics of the closed loop operation of the power supply circuit to evaluate gain, rather than direct measurement of current at the isolator terminals. With voltage regulation, a power supply will continue to operate within specifications well beyond the onset of opto-isolator wear, but then fail suddenly when the damage accumulation reaches a threshold. We show that it is possible to detect the onset of degradation of the isolator before performance is adversely affected, and track the progression of a fault to provide greater lead time for field replacement. The method presented is simple to implement on many commercial off-the-shelf (COTS) supplies, and offers the advantage of longer maintenance cycles at lower cost. We show how this method can improve the accuracy of remaining useful life (RUL) prediction, reduce the occurrence of No Trouble Found (NTF), and improve manufacturing of highly reliable SMPS.
He received the M.S. degree in EECS from Vanderbilt University
- S Chetan
- Research Kulkarni
- Isis Assistant At
- Vanderbilt University
Chetan S Kulkarni is a Research Assistant at ISIS, Vanderbilt University. He received the M.S. degree in EECS from
Vanderbilt University, Nashville, TN, in 2009, where he is
currently a Ph.D student.
He received a Ph.D. degree in Decision Sciences and Engineering Systems in
- R José
José R. Celaya is a research scientist with SGT Inc. at
the Prognostics Center of Excellence, NASA Ames Research
Center. He received a Ph.D. degree in Decision Sciences and
Engineering Systems in 2008, a M. E. degree in Operations
Research and Statistics in 2008, a M. S. degree in Electrical
Engineering in 2003, all from Rensselaer Polytechnic Institute, Troy New York; and a B. S. in Cybernetics Engineering
in 2001 from CETYS University, México.
-03). 60384-4-1 fixed capacitors for use in electronic equipment
- Iec
IEC. (2007-03). 60384-4-1 fixed capacitors for use in electronic equipment (Tech. Rep.).