Browse Topic: Prognostics

Items (152)
Find
Advancing Reliability Centered Maintenance: A Robust Statistical Approach for MTBF Estimation in Aero-Engine Fleets via Subgrouping Strategy and Dual Weibull Modeling2026-99-02137/31/2026
Fleet heterogeneity, from manufacturing variations and diverse operating conditions, complicates reliability analysis by obscuring true failure patterns in aero-engines. This is a critical challenge in an industry as inaccurate Mean Time Between Failures (MTBF) estimates threaten safety and inflate operational costs, by forcing a choice between inefficiently conservative maintenance or the risk of in-service failures. Conventional analysis often fails by pooling all fleet data. To address this, our paper presents an analytical framework that improves predictive accuracy by filtering, rather than aggregating statistical noise. The methodology uses a Randomized Block Design (RBD) and ANOVA hypothesis test to screen a diverse dataset and isolate statistically homogeneous subgroups. This filtration identifies a core fleet with a consistent failure signature, providing a purified dataset for modeling. This refined data is then modeled using both Weibull and the Exponentiated Inverse Weibull distributions to ensure the results are robust and not model-dependent. Applying this framework to a 25-engine dataset that experienced 66 failures, we isolated a stable failure pattern, yielding a primary MTBF of 171.16 hours and a cross-validated MTBF of 176.35 hours. The close 3% convergence between these models validates our approach. By providing a dependable MTBF, this work establishes a stronger foundation for data-driven Reliability Centered Maintenance (RCM). It empowers maintenance planners to move toward evidence-based intervals, safely extending engine time-on-wing, optimizing spare parts inventory, and significantly reducing direct operational costs for airlines.
Jubaid, Mayin Uddin, Bebe, Gibson, Bigyen, Musa Pethuel, Anik, S M Kullul Mehedee, Yasmin, Ashrafi, Sahran, Mohamed Sideek Mohamed
Hybrid Digital Twin for Aero-Engine Bearing Prognostics: A Physics-Informed Machine Learning Approach to RUL Prediction2026-26-07186/1/2026
Unscheduled maintenance due to the failure of critical components, such as aero-engine rolling element bearings, is a leading cause of costly Aircraft-on-Ground (AOG) events; consequently, current time-based maintenance practices are inefficient and prone to risk. This paper develops a resource-efficient Hybrid Digital Twin (HDT) model for an engine bearing, focusing on the dynamic prediction of spall growth due to Rolling Contact Fatigue (RCF), thereby enabling a condition-based maintenance paradigm. The HDT architecture integrates two core models: (1) a physics-informed model that uses established life and fatigue theory to define initial degradation thresholds, and (2) a data-driven Recurrent Neural Network (RNN), specifically a Long Short-Term Memory (LSTM) network, for dynamic degradation rate modeling. The methodology utilizes a Monte Carlo simulation coupled with RCF progression equations to generate a large, high-fidelity synthetic run-to-failure dataset under varying operational loads, accurately simulating realistic mission profiles. This approach addresses the critical "data scarcity" challenge in aviation. To ensure operational reliability, the framework incorporates Uncertainty Quantification (UQ) using Monte Carlo Dropout and addresses the "Sim-to-Real" gap through Transfer Learning on the NASA IMS bearing dataset. The HDT demonstrates a significant improvement in prognostic accuracy, achieving a Root Mean Square Error (RMSE) reduction of over 71% compared to baseline models. Furthermore, a cost-benefit analysis suggests a potential fleet savings of $240,000 per 100 engines by avoiding false negatives. This computationally efficient approach supports the Digital Engineering Transformation theme by providing a scalable blueprint for the virtual qualification of critical mechanical components.
Mohamed, Abbas
Design and Run-Time Information Exchange for Health-Ready ComponentsJA6268_202303 (Current)5/26/2026
This Surface Vehicle & Aerospace Recommended Practice offers best practices and a methodology by which IVHM functionality relating to components and subsystems should be integrated into vehicle or platform level applications. The intent of the document is to provide practitioners with a structured methodology for specifying, characterizing and exposing the inherent IVHM functionality of a component or subsystem using a common functional reference model, i.e., through the exchange of design-time data and the application of standard vehicle data communications interfaces. This document includes best practices and guidance related to the specification of the information that must be exchanged between the functional layers in the IVHM system or between lower-level components/subsystems and the higher-level control system to enable health monitoring and tracking of system degradation severity. The intent is to provide an IVHM system that can robustly report the degradation of a given component before it reaches the point where it goes outside its operational performance envelope by providing sufficient advance notice to deal with the issue. This document does not specify or address how each layer in the IVHM system produces or uses the data available for exchange.
HM-1 Integrated Vehicle Health Management Committee
Enabling Sustainable E-Mobility: An Edge–Cloud Collaborative Framework for Battery Lifecycle Health Management in Electric Vehicles2026-01-04604/7/2026
The rapid adoption of electric vehicles (EVs) is a cornerstone of the transition to sustainable transportation. However, uncertainty regarding battery degradation remains a significant obstacle, hindering vehicle energy efficiency, operational safety, and the recovery of end-of-life value. Accurate estimation of the battery state of health (SOH) and prediction of the remaining useful life (RUL) are therefore critical for sustainable vehicle lifecycle management. This study proposes an edge–cloud collaborative intelligent framework for in-vehicle deployment that leverages a Transformer-based architecture to jointly model SOH and RUL. The cloud-side model retains the full configuration to capture long-term degradation trajectories for high-accuracy RUL prediction. A lightweight edge-side model, engineered via pruning and knowledge distillation, delivers millisecond-level inference for real-time SOH estimation onboard the vehicle. To ensure efficiency, only four core health indicators are extracted for end-to-end prediction. Experimental validation across 77 battery cells demonstrates that the framework achieves SOH estimation with a root mean square error (RMSE) of 1.41% and RUL prediction with an RMSE of 2.59% (78 cycles). Furthermore, a periodic cloud-side update and over-the-air deployment mechanism ensure long-term adaptability and cross-platform scalability without full local retraining. This intelligent prognostic framework directly enhances EV reliability and sustainability by providing health-informed decision support for optimal vehicle operation, maintenance scheduling, and the reuse of second-life batteries. Consequently, it serves as a vital tool for advancing resource optimization and circular economy principles within the E-mobility ecosystem.
Gao, Weimin, Lv, Zhilong, Ou, Shiqi(Shawn)
Prognostics of Battery State of Health Using Hybrid Modeling of Electric Vehicle Charging Data2026-01-70022/27/2026
Accurate estimation of the state of health (SOH) of lithium-ion batteries is essential for ensuring the safety, reliability, and performance optimization of electric vehicles. In practical operating environments, however, data quality is often compromised by noise interference, frequent fluctuations in load conditions, and the inherently non-stationary nature of battery degradation features. These challenges reduce the effectiveness of conventional modeling approaches, which often struggle to maintain both high prediction accuracy and strong generalization capability. To address these issues, this study develops a comprehensive SOH estimation approach encompassing data quality enhancement, degradation feature extraction, and hybrid deep learning-based modeling. In the first stage, multi-stage anomaly detection techniques are applied to remove noisy or inconsistent measurements. A week-based indexing strategy is introduced to generate temporally coherent labels, ensuring that time-series dependencies are preserved. This procedure ensures a minimum per-vehicle valid-data ratio of 87.59%, thereby guaranteeing consistent data availability across all vehicles and significantly improving both reliability and temporal alignment. In the second stage, a set of degradation-sensitive health indicators is extracted from raw sensor measurements, including voltage, current, and temperature profiles. These features are then aggregated at a weekly resolution to enhance stability and reduce short-term variability. In the third stage, a hybrid deep learning model is constructed by combining a Temporal Convolutional Network (TCN) for local pattern extraction, a Bidirectional Gated Recurrent Unit (BiGRU) for long-term dependency modeling, and an attention mechanism for adaptive feature weighting. Experimental results under 10-fold cross-validation show that the proposed approach achieves a root mean square error of 1.30% and a mean absolute error of 1.03% on the test set, outperforming single-model baselines in both accuracy and robustness. The proposed framework offers a practical and scalable solution for high-precision SOH estimation in real-world scenarios and provides a strong basis for deployment under diverse and extreme operating conditions.
Wang, Sijing, Jiao, Meiyuan, Huang, Weixuan, Lin, Yiting, Liu, Honglai, Lian, Cheng
Data-Driven Capacity Estimation Method for Retired Lithium-Ion Batteries Based on Electrochemical Impedance Spectroscopy2026-01-70152/27/2026
With the rapid expansion of global electric vehicles (EVs) deployment, the echelon utilization of retired lithium-ion batteries (LIBs) has emerged as a critical issue. Although these batteries typically retain over 70% of their initial capacity and remain suitable for stationary energy storage systems, the substantial variability in aging states poses safety risks. Conventional capacity estimation methods are often time-intensive and costly, while data-driven approaches face challenges from complex degradation mechanisms and limited historical usage data. This study uses the electrochemical impedance spectroscopy (EIS) method to create a model that estimates the capacity of retired batteries. EIS offers fast measurement, requires no historical cycling data, and provides rich state-of-health (SOH) information. An EIS dataset was acquired from 18650-type LFP and NCM cells aged under multiple cycling conditions. The real part and magnitude of the impedance spectra were extracted as input features for model training. A hybrid deep learning framework integrating the sparrow search algorithm (SSA), convolutional neural networks (CNN), gated recurrent units (GRU), and an attention mechanism was developed. SSA automatically optimize model hyperparameters, mitigating the overfitting risks, while the attention mechanism highlighted informative frequency-domain features, reducing manual feature engineering and enhancing prediction accuracy. Results show excellent performance: for LFP cells, the root-mean-square error (RMSE) and mean absolute error (MAE) are 0.24% and 0.19%, respectively, with a coefficient of determination (R2) of 98.96%; for NCM cells, the RMSE and MAE are 0.99% and 0.88%, with R2 of 97.97%. On the mixed-material dataset, the RMSE, MAE, and R2 reach 0.79%, 0.67%, and 97.84%. These results confirm that the proposed method maintains high accuracy across different cathode chemistries, while significantly reducing testing and modeling costs. The approach shows strong potential for large-scale, automated screening and classification of retired LIBs in practical second-life applications.
Hou, Zhengyu, Luan, Weiling, Sun, Changzheng, Chen, Ying
Rapid Battery RUL Prediction under Different Fast-Charging Protocols via Domain-Adaptive Deep Learning and Model Compression2026-01-70032/27/2026
Accurate and rapid remaining useful life (RUL) prediction of batteries under various extreme conditions is crucial for battery management systems. However, existing methods often face challenges such as limited datasets under extreme conditions, high model complexity, and weak interpretability. Therefore, this paper proposes a hybrid framework based on pruning domain-adaptive convolutional neural networks (CNN) and long short-term memory (LSTM) to study RUL prediction under different fast-charging conditions using the MIT dataset. First, four voltage-related feature matrices are extracted. Using maximum mean discrepancy (MMD) constraints, the CNN-LSTM is trained with source domain and limited target domain data to align distributions. Neuron pruning is then applied to the fully connected layer to compress the model. Results demonstrate that under sparse target domain data, the domain adaptation approach achieves significantly lower prediction errors than fine-tuning. The pruned model maintains low prediction errors while reducing parameters by 42.32%. Further, an explainable algorithm quantifies regional data contributions to identify critical voltage intervals. Ultimately, precise predictions are achieved using only key data from the 2.9–3.2V range, fully demonstrating the method's efficiency. This study provides a lightweight and interpretable solution for cross-domain battery RUL prediction under fast-charging conditions.
Huang, Mingyue, Chen, Hongxu, Luan, Weiling
A Literature Review on Application of Maintenance and Management Strategies for Electric-Vertical Take-Off and Landing Aircraft’s Maintenance, Repair, and Overhaul Organizations01-18-03-001612/18/2025
In the context of emerging technology developed for advanced air mobility concept, its maintenance protocols are not yet mature and existing aviation maintenance systems may not support electric-vertical take-off and landing (e-VTOL) needs. Thus, the operation of e-VTOL aircraft during its deployment stage necessitates the need for qualitative maintenance support. The main purpose of this study is to develop the basic structural principles of the projected new maintenance, repair, and overhaul (MRO) organization for e-VTOL air vehicles, which will support airworthiness through comprehensive maintenance approaches. Thus, the operation of e-VTOL aircraft during its deployment stage necessitates the need for qualitative maintenance support. The importance of the study is to offer standard procedures based on management and maintenance strategies, application of predictive and prescriptive maintenance tools, which pose a significant contribution to ensuring safety, reliability, and cost-effectiveness in e-VTOL operations. The methodology based on leveraging modern management theory in combination with maintenance strategy ensures reaching a goal, creating an effective MRO organization. The findings of the analysis, conducted on the current study, reveal the suitability of the traditional aircraft maintenance approach for e-VTOL air vehicle maintenance processes that can support multi-model aircraft with different design configurations and architectures. To facilitate comprehensive engagement among all relevant stakeholders, the result of the analysis assumes the establishment of an effective aircraft maintenance ecosystem. Effective agreement between e-VTOL operators and MRO providers will contribute to ensuring appropriateness with evolving aircraft architectures in compliance with regulatory standards. This study fills a research gap in the literature relating to aircraft maintenance by proposing a digitally integrated approach and, regulation-compliant framework tailored for e-VTOL aircraft. The suggested multi-strategy maintenance model incorporates predictive analytics, modular diagnostics, and contingency planning tailored for e-VTOL operations, synergizing with AI implementation distinguishes it with its novelty implemented in the modern aviation sector.
Imanov, Tapdig, Bozdereli, Arzu
Prognostic Framework for Failure Analysis and Smart Farming2025-28-034011/6/2025
Prognostics and Health Management (PHM) is framework for electrical/mechanical components in heavy machines represents a transformative approach that harnesses cutting-edge sensing technologies and analytics to predict and elevate reliability and efficiency of agricultural/construction machinery. By using advanced data collection and sophisticated analytics, PHM achieves real-time monitoring of critical performance parameters such as voltage, current, temperature, and operational cycles, along with field data mapped with GPS coordinates as well as environmental conditions. This capability allows for the early detection of anomalies and potential failures, thereby enhancing operational reliability. Data collected from the machine will be pushed to the server periodically and whenever any failure is detected advanced AI algorithms on machine and server will analyze the information and link to collected data which will be used to identify possible failures or assess the safety of the machine for future instances. This proactive monitoring ensures that any potential issues are flagged in real-time, allowing for immediate intervention and maintenance actions. A comprehensive failure mode analysis is conducted to elucidate common failure patterns, followed by facilitating targeted and proactive maintenance strategies. Importantly, the prognostic data generated not only aids in predicting failures but also plays a crucial role in failure cause analysis. The advantages of adopting these prognostic approaches are manifold, including a significant reduction in unplanned downtime, lower maintenance costs, and enhanced safety for operators through timely interventions. The findings underscore that the implementation of PHM not only extends the lifespan of electrical/mechanical components but also advances the principles of precision agriculture and construction.
Shinde, Ketan Kishor
Motion Amplification (MA) Technology Applications for Ground Vehicle (GV) Systems2025-01-04919/16/2025
Ground vehicles in operation produce a unique vibration signature. This signature is a key indicator of vehicle system, sub-system, and component health but is often not visible to the naked eye or detectable without specialized equipment. Vibration analysis tools can capture these signatures and unlock their value by establishing a signature baseline and detecting changes to that baseline. Changes are strong and consistent indicators of incipient failure and failure progression, and therefore useful as diagnostic reports and prognostic markers. Existing vibration analysis tools and techniques make these signatures quantifiable, but these tools require on-equipment sensors and lengthy data collection processes. Motion Amplification (MA), however is a powerful new vibration analysis technology that overcomes sensor limitations and speeds data collection and analysis by replacing sensor-based vibration analysis tools with video recordings. The recordings use each individual picture element (Pixel) in a recording image as an individual motion sensor. These point data are processed by an innovative software program that integrates all Pixels into a single quantifiable and visible vibration signature and displays the amplified recording for analysis. This paper will show how MA advances the state of the art in vibration analysis and has significant potential to improve diagnostic and prognostic outcomes in complex ground vehicle systems over speed, accuracy, and sensitivity metrics and provide actionable vehicle health intelligence at the tactical level.
Aebischer, David
Battery Datasets for Neural Network-Based State-of-Health Estimators: A Review and Implementative Workflow2025-24-01399/7/2025
Management of battery systems for electric vehicles has great importance to ensure safe and efficient operation. State-of-Charge and State-of-Health (SoH) are fundamental parameters to be taken under control even though they cannot be directly measured during vehicle operation. Some control approaches have gained increasing interest thanks to advances in sensor availability, edge computing and the development of big data. In particular, SoH estimation through machine learning (ML) and neural networks (NNs) has been thoroughly investigated due to their great flexibility and potential in mapping non-linear relations within data. The numerous studies available in the literature either employ different extracted features from data to train NNs, or directly use measurement signals as input. Additionally, many studies available in the literature are based on a limited number of publicly available datasets, which mainly encompass cylindrical battery cells with small capacity. Starting from the workflow analysis for developing and implementing ML SoH estimators, this work aims to give an overview of the latest application studies in this field, with a special focus on the analysis of the main datasets available in the literature. In the end, the workflow for the implementation of NN-based SoC estimation is demonstrated with a step-by-step procedure on a publicly available dataset, and a final comparison with non-neural regression algorithms is performed.
Chianese, Giovanni, Capasso, Clemente, Veneri, Ottorino
Dynamic Modeling of Torque & Emissions: Digital Twin to Simulate Combustion Process and Exhaust Emissions for Diagnostics & Prognostics Applications2024-28-017212/5/2024
As a journey to green initiatives, one of the focus areas for automotive industry is reducing environmental impact especially in case of internal combustion engines. Latest digital twin technology enable modelling complicated, fast and unsteady phenomena including the changes of emission gases concentration and output torque observed during diesel emission and combustion process. This paper presents research on the emission and combustion characteristics of a heavy vehicle diesel engine, elaborating an engineered architecture for prognostics/diagnostics, state monitoring, and performance trending of heavy-duty vehicle engine (HDVE) and after treatment system (ATS). The proposed architecture leverages advanced modeling methodologies to ensure precise predictions and diagnostics, using data-driven techniques, the architecture accurately model’s engine and exhaust system behaviors under various operating conditions. For exhaust system, architecture demonstrates encouraging predictive performance in estimating engine/tailpipe NOx-emissions. This development introduces novel method for calculating health scores, particularly for Selective Catalytic Reduction (SCR) systems which enhances diagnostic capabilities, enabling early detection of issues such as reduced conversion efficiency. By accurately predicting emissions and identifying potential problems early, the architecture helps ensure compliance with regulatory requirements. Additionally, the architecture considers vehicle dynamics, especially in the context of drivetrain health. The Nonlinear Autoregressive with Exogenous Inputs (NARX) model for torque estimation is crucial for understanding the dynamic behavior of the engine and its impact on overall vehicle performance. By monitoring and analyzing deviations in predicted torque, the architecture provides insights into the health and performance of the drivetrain, facilitating timely interventions and maintenance actions to ensure optimal vehicle dynamics and reliability. This study presents an architecture that integrates emission and vehicle dynamics models with prognostics health asset framework offering a holistic approach to predictive maintenance for HDVE and ATS
Singh, Prabhsharn, Thakare, Ujval, Hivarkar, Umesh
Tackling Limited Labeled Field Data Challenges for State of Health Estimation of Lithium-Ion Batteries by Advanced Semi-Supervised Regression2024-01-22004/9/2024
Accurate estimation of battery state of health (SOH) has become indispensable in ensuring the predictive maintenance and safety of electric vehicles (EVs). While supervised machine learning excels in laboratory settings with adequate SOH labels, field-based SOH data collection for supervised learning is hindered by EVs' complex conditions and prohibitive data collection costs. To overcome this challenge, a battery SOH estimation method based on semi-supervised regression is proposed and validated using field data in this paper. Initially, the Ampere integral formula is employed to calculate SOH labels from charging data, and the error of labeled SOH is reduced by the open-circuit voltage correction strategy. The calculation error of the SOH label is confirmed to be less than 1.2%, as validated by the full-charge test of the battery packs. Subsequently, statistical features are extracted from charging data, and health indicator sets are selected by two correlation analysis methods (Pearson correlation and grayscale correlation). Moreover, two regressors are trained by learning the mapping between labeled SOH and various health indicator sets. To enhance the training dataset, semi-supervised with co-training is utilized to estimate pseudo-labels for unlabeled charging data. The final SOH estimation is achieved through the fusion of these two regressors. Finally, the proposed method is validated using field data from 20 electric forklifts collected over approximately one year. Remarkably, even with only 10 labeled data points, the proposed method achieves a mean absolute error in SOH estimation of a mere 3.96%. This represents a significant reduction of 20% compared to the traditional supervised learning method. Compared with the two benchmarks without co-training, the estimation error drops by 7.69% and 8.76%, respectively.
Li, Jinwen, Chen, Wenqiang, Khalatbarisoltani, Arash, Liu, Hongao, Lin, Xianke, Hu, Xiaosong
A Study on a Prognostics and Health Management (PHM) Based on Fracture Mechanics Using Deep Learning2024-01-22484/9/2024
This paper presents deep learning-based prognostics and health management (PHM) for predicting fractures of an electric propulsion (eP) drivetrain system using real-time CAN signals. The deep learning algorithm, based on autoencoders, resamples time-series signals and converts them into 2D images using recurrence plots (RP). Subsequently, through unsupervised learning of DeepSVDD, it detects anomalies in the converted 2D images and predicts the failure of the system in real-time. Also, reliability analysis based on fracture mechanics was performed using the detected signals and big data. In particular, the severity of the eP drivetrain system is proportional to the maximum shear stress (τmax) in terms of linear elastic fracture mechanics (LEFM) and can be calculated by summarizing the relationship between cracks (a) and the stress intensity factor (KIII). During this process, the system status can be checked by comparing the stress intensity factor and fracture toughness (KIIIc), and the time from the detection of an abnormal signal in the system to complete failure can be quantitatively determined. Therefore, it is possible to continuously maintain the status of the system by detecting failure signals using deep learning before vehicle parts fail, and with the detected failure prediction signals, a process can be established to enable users to repair defects in the vehicle system before breakdown occurs. By predicting the remaining life of the system and calculating field reliability through these procedures, we introduce innovative technologies aimed at preventing safety accidents, reducing economic costs, and addressing quality issues. In the future, we expect to achieve high business performance by extending and applying this deep learning-based PHM approach to all vehicle components.
Moon, Byungwoo, Lee, SangWon, Nam, DongJin, Kim, Jeonghwan, Bae, JaeWoong, Shin, JeongMin
Connected Vehicle Data – Prognostics and Monetization Opportunity2023-01-168510/31/2023
In recent years, the automotive industry has seen an exponential increase in the replacement of mechanical components with electronic-controlled components or systems. engine, transmission, brake, exhaust gas recirculation (EGR), lighting, driver-assist technologies, etc. are all monitored and/or controlled electronically. Connected vehicles are increasingly being used by Original Equipment Manufacturers (OEMs) to collect and transmit vehicle data in real-time via the use of various sensors, actuators, and communication technologies. Vehicle telematics devices can collect and transmit data about the vehicle location, speed, fuel efficiency, State Of Charge (SOC), auxiliary battery voltage, emissions, performance, and more. This data is sent over to the cloud via cellular networks, where it can be processed and analyzed to improve their products and services by automotive companies and/or fleet management. This data can also be used for a variety of purposes, including enhancing the driving experience, improving safety, understanding customer driving patterns, vehicle functionality utilization, SOC monitoring, battery thermal monitoring, prognostic health review, and providing new services to drivers and passengers. By collecting and analyzing this data, connected vehicles can provide a wealth of insights that can be used to improve safety, reduce congestion, and risky driving behavior, review pending diagnostics codes in vehicles, and enhance the overall driving experience. This paper investigates and explores the opportunities related to connected vehicle data analytics, vehicle health prognostics, and possible monetization associated scope.
Kumar, Vivek, Zhu, Di, Dadam, Sumanth Reddy
Prediction of the Remaining Useful Life of the Proton Exchange Membrane Fuel Cell with an Integrated Health Index2023-01-701310/30/2023
To improve the prediction accuracy of the remaining useful life (RUL) of the proton exchange membrane fuel cell (PEMFC), an integrated health index (IHI) including electrical and non-electrical parameters of PEMFC is established, and the RUL prediction is conducted based on the above index. Firstly, several operating conditions including the PEMFC degradation information are selected according to the information theory method. Moreover, the IHI is established by the sequential quadratic programming method. Secondly, RUL predictions based on the power and IHI are conducted by the adaptive neuro fuzzy inference system (ANFIS), respectively. Finally, different results comparisons including power and IHI differences, differences between experimental and training/predicting results, amounts of different differences in training and predicting phases, and RUL prediction results are presented in detail. The results show that the accuracy of the RUL prediction based on the IHI is higher than that based on the power. The accuracy at the time of 654 h of the ANFIS based on the IHI is improved by 40.8% and 30.4% compared with the linear fitting and ANFIS based on the power, respectively.
Fan, Lei, Zhou, Su, Wen, Chaokai, Gao, Jianhua
The Adoption of Digital Twins in Integrated Vehicle Health ManagementEPR202302410/26/2023
To many, a digital twin offers “functionality,” or the ability to virtually rerun events that have happened on the real system and the ability to simulate future performance. However, this requires models based on the physics of the system to be built into the digital twin, links to data from sensors on the real live system, and sophisticated algorithms incorporating artificial intelligence (AI) and machine learning (ML). All of this can be used for integrated vehicle health management (IVHM) decisions, such as determining future failure, root cause analysis, and optimized energy performance. All of these can be used to make decisions to optimize the operation of an aircraft—these may even extend into safety-based decisions. The Adoption of Digital Twins in Integrated Vehicle Health Management, however, still has a range of unsettled topics that cover technological reliability, data security and ownership, user presentation and interfaces, as well as certification of the digital twin’s system mechanics (i.e., AI, ML) for use in safety-critical applications. Click here to access the full SAE EDGETM Research Report portfolio.
Phillips, Paul
Physics Informed Machine Learning for Advanced Diagnostics & Prognostics of Ground Combat Vehicles2024-01-40898/10/2023
We introduce novel approaches utilizing Physics Informed Machine Learning (PIML) for advanced diagnostics & prognostics of ground combat vehicles (CV). Specifically, we present the development of a PIML model designed to predict the health of engine oil in diesel engines. The condition of engine oil is closely linked to engine wear, thus serving as a crucial indicator of engine health. Our model integrates a physics-based simulation of engine wear in diesel engines, leveraging a time history of engine oil viscosity and engine speed as key input parameters. Furthermore, we conduct uncertainty quantification to assess the impact of varying parameters on engine oil health prediction. Additionally, our model demonstrates the capability to enhance low-fidelity physics models through the integration of a limited set of experimental data. By combining data-driven techniques with physics-based insights, our approach offers enhanced diagnostics and prognostics capabilities for ground combat vehicles, thereby facilitating proactive maintenance and optimization for operational readiness.
Betts, Juan F., Alizadeh, Arash
Criticality of Prognostics in the Operations of Autonomous Aircraft01-16-03-00226/28/2023
This article addresses the design, testing, and evaluation of rigorous and verifiable prognostic and health management (PHM) functions applied to autonomous aircraft systems. These PHM functions—many deployed as algorithms—are integrated into a holistic framework for integrity management of aircraft components and systems that are subject to both operational degradation and incipient failure modes. The designer of a comprehensive and verifiable prognostics system is faced with significant challenges. Data (both baseline and faulted) that are correlated, time stamped, and appropriately sampled are not always readily available. Quantifying uncertainty, and its propagation and management, which are inherent in prognosis, can be difficult. High-fidelity modeling of critical components/systems can consume precious resources. Data mining tools for feature extraction and selection are not easy to develop and maintain. And finally, diagnostic and prognostic algorithms that address accurately the designer’s specifications are not easy to develop, verify, deploy, and sustain. These are just the technical challenges. On top of these are business challenges, for example, demonstrating that the PHM functionality will be economically beneficial to the system stakeholders, and finally, there are regulatory challenges, such as, assuring the authorities that the PHM system will have the necessary safety assurance levels while delivering its performance goals. This article tackles all three aspects of the use of PHM systems in autonomous systems. It outlines how some of the technical challenges have been overcome and demonstrates why PHM could be essential in this ecosystem and why regulatory authorities are increasingly open to the use of PHM systems even in the most safety-critical areas of aviation.
Vachtsevanos, George, Rajamani, Ravi
Study on State-of-the-Art Preventive Maintenance Techniques for ADS Vehicle Safety2023-01-08464/11/2023
1 Autonomous Driving Systems (ADS) are developing rapidly. As vehicle technology advances to SAE level 3 and above (L4, L5), there is a need to maximize and verify safety and operational benefits. As a result, maintenance of these ADS systems is essential which includes scheduled, condition-based, risk-based, and predictive maintenance. A lot of techniques and methods have been developed and are being used in the maintenance of conventional vehicles as well as other industries, but ADS is new technology and several of these maintenance types are still being developed as well as adapted for ADS. In this work, we are presenting a systematic literature review of the “State of the Art” knowledge for the maintenance of a fleet of ADS which includes fault diagnostics, prognostics, predictive maintenance, and preventive maintenance. We are providing statistical inference of different methodologies, comparison between methodologies, and providing our inference of different techniques that are used in other industries for maintenance that can be utilized for ADS. This paper presents a summary, main result, challenges, and opportunities of these approaches and supports new work for the maintenance of ADS.
Sanket, Rohit, Hanif, Athar, Ahmed, Qadeer, Monohon, Mark
Prognostics and Machine Learning to Assess Embedded Delamination Tolerance in Composites01-15-02-00168/26/2022
Laminated composites are extensively used in the aerospace industry. However, structures made from laminated composites are highly susceptible to delamination failures. It is therefore imperative to consider a structure’s tolerance to delamination during design and operation. Hybrid composites with laminas containing different fibers were used earlier in laminates to achieve certain benefits in strength, stiffness, and buckling. However, the concept of mixing laminas with different fibers was not explored by researchers to enhance delamination tolerance levels. This article examines the above aspect of hybridization by employing machine learning algorithms and proposes a reliable method of analysis to study delamination, which is crucial to ensure the safety of airframe composite panels. In this article, fracture-mechanics-based structural integrity results related to mode I Strain Energy Release Rate (SERR) are obtained using geometrical non-linear three-dimensional finite element analysis. The parametric study helped to subject the data to multiple polynomial regressions for predictive model development. A standalone executable program deploys the machine learning model to predict the delamination tolerance of laminated composite panels. It is confirmed from the current study that appropriate hybridization with glass plies in between a few top and bottom carbon layers enhances the levels of delamination tolerance.
Nambisan, Savitha Narayanan, Dattaguru, B.
Connected On-Board Inert Gas Generation System with Real-Time Data Access & Machine Learning Approach for Predictive / Prognostics Maintenance2022-26-00235/26/2022
The purpose of the OBIGGS is to reduce the amount of oxygen in the fuel tank to a 'safe' level to significantly reduce the possibility of ignition of fuel vapors. There are circumstances where equipment of OBIGGS like ASMs, Ozone Converter Catalysts, etc. gets degraded earlier than the provided MTBF. This paper studies the present conventional systems limitations, like due to memory constraints only the faults and limited shop data are being recorded, hence there is no provision to store/report the stream of data margins with which we can pass/fail the performance tests. This paper also explains how a new design of the Connected concept achieves access to real-time data from the system and how the data is pushed to the cloud network. A connected solution for the OBIGGS is the technology to access real-time data (Systems LRUs Performance data and Custom data Parameters) from the Systems controller data bus, this data is further applied to AI/ML methods for predictive/prognostics features to compare why the performance of the ASMs in some systems may degrade quicker than others and to inform of when equipment of OBIGGS may need to be inspected/tested/replaced. VOCs related to ASMs degradation, FQIS field issues, sensors, valves, and other equipment's data parameters can be monitored over time that would be of value and an interest in more details for the Suppliers, Manufacturers, and Customers side.
Kumar, Naveen, Kotnadh, Shivaprasad, MorkondaHaribapu cEng, Arvind, Kanneboyina cEng, Rajesh, Rao cEng, Manjunatha
A Prognostic Based Control Framework for Hybrid Electric Vehicles2022-01-03523/29/2022
Electrified transportation has received significant interest recently because of sustainable and clean energy goals. However, the degradation of electrical components such as energy storage systems raises system reliability and economic concerns. In this paper, a prognostic-based control strategy is proposed for hybrid electric vehicles (HEVs) to abate the degradation of energy systems. Degradation forecasting models of electrical components are developed to predict their degradation paths. The predicted results are then used to control HEVs in order to reduce the degradation of components.
Hoang, Phuong H., Ozkan, Gokhan, Badr, Payam, Timilsina, Laxman, Edrington, Christopher
Deep Uncertainty Quantification of Prognostic Techniques for Proton Exchange Membrane Fuel Cell2022-01-70012/14/2022
An accurate voltage prediction associated with uncertainty quantification is of great importance to predict the remaining useful life for proton exchange membrane fuel cell in automobile applications. This paper achieves the remaining useful life prediction using deep neural networks, with an emphasis on uncertainty quantification in voltage prognostics for proton exchange membrane fuel cell systems. The trend and pattern of voltage degradation data was investigated by using long-short term memory and the voltage prediction trend was represented with prediction interval. The experimental results show that the deep learning model with corresponding uncertainty techniques can achieve prediction root mean square error values within 0.02 and represent the voltage prediction with a prediction interval.
Yang, Xin, Chen, Fengxiang
Electronic Prognostics – A Case Study Using Global Positioning System (GPS)TBMG-408582/1/2022
Many types of circuits compose avionics systems. One of the following categories can be used to classify each circuit topology at the time this research was performed:
Electronic Prognostics - A Case Study Using Global Positioning System (GPS)22AERP02_072/1/2022
Prognostic health management (PHM) of electronic systems presents challenges traditionally viewed as either insurmountable or not worth the cost of pursuit, but recent changes in weapons platform acquisition and support requirements has spurred renewed interest in electronics PHM, revealing possible applications, accessible data sources, and previously unexplored predictive techniques. Naval Air Warfare Center, Patuxent River, Maryland Many types of circuits compose avionics systems. One of the following categories can be used to classify each circuit topology at the time this research was performed: High frequency analog Low frequency analog Low impedance High impedance Common failure mode mechanisms for analog circuits depend largely on the architecture and relative operating frequencies of the circuit. In this research, high frequency analog circuits are categorized as operating above 1GHz, while low frequency analog circuits operate below 1GHz. High frequency analog circuits are sensitive to small changes in device parameters, resulting in non-destructive, or operational, failure modes. Unlike physical device failures, the cause of operational failures cannot be traced back to individual components. Low frequency analog circuits are more likely to undergo physical device failure. The accompanying figure illustrates the relationship between the operating frequency of an analog circuit and the different types of failure modes.
Prognostic Metrics for Engine Health Management SystemsAIR5909A (Current)11/30/2021
This SAE Aerospace Information Report (AIR) presents metrics for assessing the performance of prognostic algorithms applied for Engine Health Management (EHM) functions. The emphasis is entirely on prognostics and as such is intended to provide an extension and complement to such documents as AIR5871, which offers information and guidance on general prognostic approaches relevant to gas turbines, and AIR4985 which offers general metrics for evaluating diagnostic systems and their impact on engine health management activities.
E-32 Aerospace Propulsion Systems Health Management
Surveying Off-Board and Extravehicular Monitoring and Progress Towards Pervasive Diagnostics12-04-04-002710/26/2021
We survey the state of the art in off-board diagnostics for vehicles, their occupants, and environments, with particular focus on vibroacoustic (VA) approaches. We identify promising application areas including data-driven management for shared mobility and automated fleets, usage-based insurance, and vehicle, occupant, and environmental state and condition monitoring. We close by exploring the particular application of VA monitoring to vehicle diagnostics and prognostics and propose the introduction of automated vehicle- and context-specific model selection as a means of improving algorithm performance, e.g., to enable smartphone-resident diagnostics. Towards this vision, four strong-performing, interdependent classifiers are presented as a proof of concept for identifying vehicle configuration from acoustic signatures. The described approach may serve as the first step in developing “universal diagnostics,” with applicability extending beyond the automotive domain.
Siegel, Joshua E., Coda, Umberto, Terwilliger, Adam M.
RELIABILITY MODELING TO INFORM THE DEVELOPMENT OF ON-PLATFORM PREDICTIVE ANALYTICS2024-01-39158/10/2021
ABSTRACT Implementing Prognostic and Predictive Maintenance (PPMx) for the U.S. Army’s ground vehicle fleet requires the design and integration of on-platform predictive analytics. To support the design process, U.S. Army DEVCOM Ground Vehicle Systems Center (GVSC) and Applied Research Laboratory (ARL) Penn State researchers are developing a systematic approach that uses reliability modeling in a guiding role. The key steps of the process are building the initial reliability model from available data (e.g., system diagrams and physical layouts), augmenting with information on observed states and failure modes via subject matter experts, and then conducting trades on additional sensors and algorithms to determine a suitable predictive analytics capability. In this paper we provide an example of this process as applied to an Army ground vehicle, first focusing on a simplified sub-problem to demonstrate the technique, then providing statistics on the large scale process. Citation: M. Majcher, L. Bennett, J. Banks, M. Lukens, E. Nulton, M. Yukish, J. Merenich, “Reliability Modeling to Inform the Development of On-Platform Predictive Analytics”, In Proceedings of the Ground Vehicle Systems Engineering and Technology Symposium (GVSETS), NDIA, Novi, MI, Aug. 10-12, 2021.
Majcher, Monica, Bennett, Lorri A., Banks, Jeffrey, Lukens, Matthew, Nulton, Eric, Yukish, Michael A., Merenich, John J.
A HIGH-PERFORMANCE DATA TO DECISION SOLUTION FOR ALL ECHELON PARTICIPATION IN ARMY GROUND VEHICLE PPMX2024-01-39148/10/2021
ABSTRACT As the Army leverages Prognostic and Predictive Maintenance (PPMx) models to migrate ground vehicle platforms toward health monitoring and prescriptive maintenance, the need is imminent for a pipeline to quickly and constantly move operational and maintenance data off the platform, through analytic models, and push the insights gained back out to the edge. This process will reduce data-to-decision time and operation and sustainment costs while increasing reliability for the platform and situational awareness for analysts, subject matter experts, maintainers, and operators. The US Army Ground Vehicle Systems Center (GVSC) is collaborating with The US Army Engineer Research and Development Center (ERDC) to develop a system of systems approach to stream operational and maintenance data to appropriate computing resources, collocating the data with DoD High-Performance Computing (HPC) processing capabilities where appropriate, then channeling the generated insights to maintainers and operational decision makers where this decision support will have the greatest impact. The team has accomplished proof of concept or prototyping for several foundational components of the system and has demonstrated the effectiveness of combining data analytics with high-performance computing on large data in discovering and developing PPMx models. Citation: W. Glenn Bond, Andrew Pokoyoway, David Daniszewski, Cesar Lucas, Thomas L. Arnold, Haley R. Dozier, “A High-Performance Data to Decision Prototyping Solution for All Echelon Participation in Army Ground Vehicle PPMx”, In Proceedings of the Ground Vehicle Systems Engineering and Technology Symposium (GVSETS), NDIA, Novi, MI, Aug. 10-12, 2021.
Bond, W. Glenn, Pokoyoway, Andrew, Daniszewski, David, Lucas, Cesar, Arnold, Thomas L., Dozier, Haley R.
Use of Health Monitoring Systems to Detect Aircraft Exposure to Volcanic EventsAIR6212 (Current)5/13/2021
This document collates the ways and means that existing sensors can identify the platform’s exposure to volcanic ash. The capabilities include real-time detection and estimation, and post flight determinations of exposure and intensity. The document includes results of initiatives with the Federal Aviation Administration (FAA), the European Aviation Safety Agency (EASA), the International Civil Aviation Organization (ICAO), Transport Canada, various research organizations, Industry and other subject matter experts. The document illustrates the ways that an aircraft can use existing sensors to act as health monitoring tools so as to assess the operational and maintenance effects related to volcanic ash incidents and possibly help determine what remedial action to take after encountering a volcanic ash (VA) event. Finally, the document provides insight into emerging technologies and capabilities that have been specifically pursued to detect volcanic ash encounters but are not yet a part of an airplane’s standard fit.
HM-1 Integrated Vehicle Health Management Committee
Prognostics Framework to Enable a Maintenance Free Operating Period F-0077-2021-168035/10/2021
ABSTRACT
DeWind, Eric, Sopko,  Richard
Prognostics and Health Management Guidelines for Electro-Mechanical ActuatorsAIR8012 (Current)11/12/2020
Reducing the power consumption—and hence, the fuel burn—is a major target for the next generation of aircraft, and electrical actuation is perceived as a technological area able to provide power saving. Electrical actuation can in fact contribute to the reduction of the non-propulsive power because electro-mechanical actuators, when compared to the conventional hydraulic actuators, rely on a form of power subjected to lower distribution losses and in general can lead to a weight savings at the aircraft level if the required power remains under a break-over point. Moreover, electro-mechanical actuators (EMAs) present higher reliability and maintainability with a lower life-cycle cost. Two critical issues with electrically powered actuation are the temperature rise in the electric motor windings and in the power electronics, and the sensitivity to certain single point of failures that can lead to mechanical seizures, that has so far thwarted the use of EMAs for safety-critical applications. In order to address the issue of a possible actuator seizure, many research and development activities have been performed to identify ways of making an electro-mechanical actuator jam-free or jam-tolerant, since this would allow more flexibility in defining the overall architecture of the flight control system. Although interesting and ingenious design solutions have been proposed, they all have so far resulted in complex mechanical designs that on one hand allow the actuator to operate after a jam of an internal component, but on the other hand bring about increased weight, volume, and cost, and a reliability reduction due to the much larger number of parts. With primary flight control actuators being some of the more safety-critical components of an aircraft, an undetected actuator failure can lead to serious consequences. Furthermore, failures that are not flight safety critical can still have negative consequences ranging from flight disruptions to unscheduled and costly maintenance. The development of an effective and reliable PHM system for EMAs is thus seen as a possible way to contribute to the acceptance of EMAs as primary flight control actuators in commercial aircraft, for an effective PHM system will feature appropriate diagnostic functions to detect anomalous behavior, and prognostic capabilities indicating when an incipient fault develops and estimating how, under continuing usage, the fault will eventually become a failure. Providing EMAs with a PHM capability will entail positive results in different areas: Progressively improve the EMA robustness; PHM algorithms can initially provide indication of the onset of a fault, while more sophisticated reasoning functions can then assess the remaining useful life and contribute to the safety of flight. Minimize the unscheduled maintenance events, thereby increasing the EMA availability. Facilitate maintenance operations and troubleshooting. Simplify the supply chain. It is important to note that a more extensive use of EMAs facilitated by PHM algorithms must be associated to all design provisions necessary to attain mandatory safety requirements.
HM-1 Integrated Vehicle Health Management Committee
Prognostics for Aerospace Propulsion SystemsAIR5871A (Current)10/14/2020
This document applies to prognostics of aerospace propulsion systems. Its purpose is to define the meaning of prognostics in this context, explain their potential and limitations, and to provide guidelines for potential approaches for use in existing condition monitoring environments. This document also includes some examples. The current revision does not provide specific guidance on validation and verification, nor does it address implementation aspects such as computational capability or certification.
E-32 Aerospace Propulsion Systems Health Management
Diagnostic and Prognostic Metrics for Aerospace Propulsion Health Management SystemsAIR7999 (Current)10/14/2020
This Aerospace Information Report (AIR) presents metrics for assessing the performance of diagnostic and prognostic algorithms and systems delivering propulsion health management functions.
E-32 Aerospace Propulsion Systems Health Management
Automated Component Tracking Technologies for Future Vertical LiftF-0076-2020-1627310/5/2020
Over the last decade or more there has been a concerted push to move from on condition to predictive maintenance to improve rotorcraft availability and cost competitiveness of sustainment (Ref. 1-2). The US Army, along with industry partners, have been working on the development of prognostics for complete rotorcraft coverage. It has been identified that accurately capturing maintenance actions is needed to improve the accuracy of prognostics for better component health state awareness. Further to achieve the Army's vision for Zero Maintenance rotorcraft and meet the Maintenance Free Operating Period (MFOP) (Ref. 3) requirements for the Future Vertical Lift (FVL) program, it's essential to have an automated configuration management system. To help meet these objectives, the Army and Honeywell are working on the Rotorcraft Automated Component Tracking (RACT) Science and Technology (S&T) development program. This paper discusses the research being conducted to enable the Army's RACT concept done by the Honeywell team and the CCDC AvMC. It identifies the current state of RACT technologies and challenges of integrating such technologies into the rotorcraft environment.
Bharadwaj, Raj, Moffatt, John, Borck, Hayley
Using the IoT for Predictive MaintenanceTBMG-376619/1/2020
Efficient and effective maintenance is critical to the success of all types of enterprises. In the manufacturing, extractive, utilities, and logistics industries, any unnecessary downtime, directly affects the bottom line. Failures of building systems, both commercial and residential, lead to extra expense and unhappy occupants.
Unsettled Issues Concerning Integrated Vehicle Health Management Systems and Maintenance CreditsEPR20200065/27/2020
The “holy grail” for prognostics and health management (PHM) professionals in the aviation sector is to have integrated vehicle health management (IVHM) systems incorporated into standard aircraft maintenance policies. Such a change from current aerospace industry practices would lend credibility to this field by validating its claims of reducing repair and maintenance costs and, hence, the overall cost of ownership of the asset. Ultimately, more widespread use of advanced PHM techniques will have a positive impact on safety and, for some cases, might even allow aircraft designers to reduce the weight of components because the uncertainty associated with estimating their predicted useful life can be reduced. We will discuss how standard maintenance procedures are developed, who the various stakeholders are, and – based on this understanding - outline how new PHM systems can gain the required approval to be included in these standard practices. There have been a few limited successes in this field already, and we will discuss the lessons learned in developing these systems. Finally, we will review the progress that the structural health management (SHM) community has made, and continues to make, to change the way the industry regards automated SHM systems. NOTE: SAE EDGE™ Research Reports are intended to identify and illuminate key issues in emerging, but still unsettled, technologies of interest to the mobility industry. The goal of SAE EDGE™ Research Reports is to stimulate discussion and work in the hope of promoting and speeding resolution of identified issues. SAE EDGE™ Research Reports are not intended to resolve the issues they identify or close any topic to further scrutiny. Click here to access the full SAE EDGETM Research Report portfolio.
Rajamani, Ravi
Determination of Costs and Benefits from Implementing an Engine Health Management SystemARP4176A (Current)4/28/2020
This ARP provides an insight into how to approach a cost benefit analysis (CBA) to determine the return on investment (ROI) that would result from implementing a propulsion Prognostics and Health Management (PHM) system on an air vehicle. It describes the complexity of features that can be considered in the analysis, the different tools and approaches for conducting a CBA and differentiates between military and commercial applications. This document is intended to help those who might not necessarily have a deep technical understanding or familiarity with PHM systems but want to either quantify or understand the economic benefits (i.e., the value proposition) that a PHM system could provide.
E-32 Aerospace Propulsion Systems Health Management
Rationale, Considerations, and Framework for Data Interoperability for Health Management within the Aerospace EcosystemAIR6904 (Current)4/21/2020
This SAE Aerospace Information Report (AIR) offers an overview of the aspects of intellectual property (IP) protection, legislative compliance, business model, and technologies which need to be considered and addressed to implement a data interoperability, secure business model and technology platform to enable prognostics and health management (PHM) in the digital age. While this information report is restricted to the aerospace domain and also to commercial aviation, the concepts are applicable to any other domain that employs data for supporting health management functionality.
G-31 Digital Transactions for Aerospace
In-Vehicle Diagnostic System for Prognostics and OTA Updates of Automated/Autonomous Vehicles2020-01-13734/14/2020
The increasing complexity of microcontroller-based automotive E/E systems that control road-vehicles and non-road mobile machinery comes with increased self-diagnosis functions and diagnosability via external test equipment (diagnostic tester). Technicians in the development, production and service depend on diagnostic test equipment that is connected to the E/E system and performs diagnostic communication. Examples of use cases of diagnostic communication include but are not limited to condition monitoring, data acquisition, (guided) fault finding and flash programming. More and more functions of a modern vehicle are realized by software (firmware). Powerful multicore servers replace the numerous control units and many control unit functions can be performed directly by smart sensors and actuators. New E/E system architectures come with increased self-diagnostic capabilities. They automatically perform tests, log diagnostic data and push such data for prognostics purposes and condition (health) monitoring to the cloud. They also support over-the-air firmware updates (FOTA). This paper describes the components of an E/E system that is equipped with an in-vehicle diagnostic tester. The tester consists of standardized components, including MVCI-Server (ISO 22900), ODX (ISO 22901), OTX (ISO 13209) and UDS on IP (ISO 14229-5). The paper includes a description of cybersecurity measures to protect the vehicle against malicious attacks.
Subke, Peter, Moshref, Muzafar, Erber, Julian
Unsettled Technology Opportunities for Vehicle Health Management and the Role for Health-Ready ComponentsEPR20200033/17/2020
Game-changing opportunities abound for the application of vehicle health management (VHM) across multiple transportation-related sectors, but key unresolved issues continue to impede progress. VHM technology is based upon the broader field of advanced analytics. Much of traditional analytics efforts to date have been largely descriptive in nature and offer somewhat limited value for large-scale enterprises. Analytics technology becomes increasingly valuable when it offers predictive results or, even better, prescriptive results, which can be used to identify specific courses of action. It is this focus on action which takes analytics to a higher level of impact, and which imbues it with the potential to materially impact the success of the enterprise. Artificial intelligence (AI), specifically machine learning technology, shows future promise in the VHM space, but it is not currently adequate by itself for high-accuracy analytics. The recent push for health-ready components offers hope in resolving some of the issues slowing the implementation of VHM technology. Health-ready components are those components that provide the necessary functionality or information to allow them to be gracefully integrated into an overall VHM solution. Our primary focus area in this SAE EDGE™ Research Report is on maintaining the health of vehicles in various transportation sectors with the greatest content coming from automotive. Tremendous synergies can be achieved by applying these very broad concepts from automotive to aerospace and other sectors, and vice versa. As will be seen, these concepts are also important for key emerging product features and for the manufacturing systems that produce the vehicles. The barriers impeding progress are organizational, historical, technological, and legal, among others. We offer some insights into how these barriers arose with some potential courses of action to mitigate them as well as to stimulate further discussion. NOTE: SAE EDGE™ Research Reports are intended to identify and illuminate key issues in emerging, but still unsettled, technologies of interest to the mobility industry. The goal of SAE EDGE™ Research Reports is to stimulate discussion and work in the hope of promoting and speeding resolution of identified issues. SAE EDGE™ Research Reports are not intended to resolve the issues they identify or close any topic to further scrutiny. Click here to access the full SAE EDGETM Research Report portfolio.
Holland, Steve
Digital Twins for Prognostic Profiling2019-28-245611/21/2019
Ability to have least failures in products on the field with minimum effort from the manufacturers is a major area of focus driven by Industry 4.0 initiatives. Amidst traditional methods of performing system/subsystem level tests often does not enable the complete coverage of a machine health performance predictions. This paper highlights a workable workflow that could be used as a template while considering system design especially employing Digital Twins that help in mimicking real-life scenarios early in the design cycle to increase product’s reliability as well as tend to near zero defects. With currently available disruptive technologies, systems integrated multi-domain 'mechatronics' systems operating in closed-loop/close-interaction. This poses great challenge to system health monitoring as failure of any component can trigger catastrophic system failures. It may be the reason that component failures, as per some aerospace reports, are found to be major contributing factors to aircraft loss-of-control. Essentially, it is either too expensive or impossible to monitor every component or subsystem of a complex machine and the current state of the Integrated Health Monitoring Systems seem to be quite inadequate. In this paper, we propose an approach that combines the best of the diagnostics and feature extraction techniques coupled with Artificial Intelligence as a solution to address the challenges of Prognostics Health Management (PHM) for complex systems. The paper also documents a standard procedure to apply the right technologies/tools at every stage so that a clear process can be applied for any similar complex system across the product development life cycle. In this paper we derive the health status of subsystems by looking at system level responses [1]. Distinguishing features are derived from the overall system level response through feature extraction methodologies and then fed into decision making frameworks that are implemented using both Convolutional Neural Networks [7, 18, 19], Machine Learning [4] and Deep Learning. Models are trained with distinguishable features through system simulations [20]. Employing rightly designed ML models provide the ability of classifying the failure modes as well as to analyze system faults/responses. Predictive modelling techniques are applied to the ML processed data to deliver useful prognostics on the criticality of the failure mode, RUL of the components/subsystems while system is in operation can be determined. The proposed concept can be easily adapted to various systems from varying domains [2]. The methodology evolved in this work can be easily extended for various use cases for instance in the Transportation domain the user can get alerts not only of failures ahead of time but also the remaining useful lifer as well as possible causes of such a failure. This would let prevent downtime of the overall vehicle/fleet and thereby ensures smooth operation of the entire service. As a case study, the present work demonstrates the a DPHM solution applied to electrical energy generator where failure mode effects of subsystems and their effect on the overall system performance are studied using Modeling and Simulation techniques. The overall work would finally lead in demonstrating a working recommendation/advisory system that understand the behavior as if it was a pure Digital Twin [24] and thereby giving a quick turn around for different use cases like study/analysis/what if/predict behavior under various operating conditions with a high level of confidence before the changes are tried on a real system.
Thukaram, Painuri, Mohan, Sreeram
Improvement of Aircraft Availability and Optimization of Component Costs by Pre-Emptive Removal of Targeted Components2019-01-13413/19/2019
Availability of large repairable systems, like aircraft, are critical for commercial operators to generate revenue, and for military organizations to achieve their mission readiness objectives. Of the relatively few studies that deal with improving availability, most have focused on increasing reliability, and not on the biggest driver of low availability - Unscheduled Maintenance Events (UMEs). The cost of maintenance has long been a target of cost-cutting measures, and one common strategy focuses on extracting as much service life as possible out of various non-critical system components by letting those components “run to failure” (as defined in SAE JA1012). However, one of the biggest drawbacks of the “run to failure” approach is that it comes at the cost of lower asset availability because the failure of one of those components will nearly always lead to a UME, typically just when the operator wants to use, or is currently using, that asset. To combat the impact of UMEs, many OEMs, operators, and component manufacturers are looking to prognostics to get advanced notice of impending failures, so monitored components can be replaced before they completely fail. But, for technological and/or economic reasons, prognostics are not a viable option for the vast majority of components. Furthermore, the idea that running components to failure will reduce costs is fundamentally flawed because it fails to account for the extra operational costs incurred from those UMEs. As an alternative to running components to failure or relying only on prognostics, asset operators and maintainers need other strategies to minimize the operational impact of UMEs for components without prognostics that also balances component utilization and the operational costs associated with UMEs against overall asset availability. This paper presents a methodology for evaluating the trade-offs between these factors and shows how this approach can potentially reduce overall asset life-cycle costs.
Lesmerises, Alan
Framework Standard for Prognosis: An Approach for Effective Prognosis Implementation2019-26-03641/9/2019
Prognosis is used to improve system availability. This is achieved by minimizing system downtime with the help of mechanisms that senses the degradation in the system health to predict the ‘time-to-failure’ of the system. Degradation in the system’s health is measured by sensing the early signs of aging and wear and tear of the system components. This requires knowledge of all the failure modes of the system along with patterns of behavioral changes in the individual components of the system while it continues to age. Prognosis methods and mechanisms are still evolving. So, no comprehensive guidelines or framework standards exist as of today that can provide reliable and standardized prognosis solutions to the end user customers. The intent of devising such a framework and guidelines is to improve and standardize the implementation of prognosis solutions so that; it will be more effective to all stakeholders from the perspective of safety, cost and convenience. At present, there is a lot of variation in the implementation of a prognostic mechanism, although having well developed methods for the same. This is due to the lack of availability of a common framework that guides the design of the prognostic mechanism for a product. The framework standard proposed in this paper can facilitate the development of the prognosis methods that are more stable, accurate and reliable, in terms of providing information about the health of the system at the right time making it more useful from the perspective of maintenance. The framework standard provides guidance throughout system life cycle, i.e. right from the development till maintenance phase.
Pawar, Sanket
Condition-Based Maintenance in Aviation: The History, The Business and The TechnologyPT-19312/11/2018
Condition-Based Maintenance in Aviation: The History, The Business and The Technology describes the history and practice of Condition-Based Maintenance (CBM) systems by showcasing ten technical papers from the archives of SAE International, stretching from the dawn of the jet age down to the present times. By scientifically understanding how different components degrade during operations, it is possible to schedule inspections, repairs, and overhauls at appropriate intervals so that any incipient failure can be detected well in advance. Today, this includes more sensors and analytics so that periodic inspections are replaced by automated "continuous" inspections, and analytical methods that detect imminent failures and predict degradation issues more economically and efficiently. Similar concepts are also being developed for delivering prognostics functions, such as tracking of remaining useful life (RUL) of life-limited parts in aircraft engines. The discipline within CBM that deals with this is called prognostics and health management (PHM), which covers all aspects of diagnostics and prognostics, including modeling of systems and subsystems, sensing, data transmission, storage and retrieval, analytical methods, and decision making. Traditionally, nondestructive testing (NDT) methods have been employed during the major airplane checks to assess structural damage. These techniques are enhanced with in- situ sensing techniques that can continuously monitor aircraft structures and report on their health. The move to condition-based assessment of maintenance needs to be balanced by the assurance that safety is not compromised, that initial cost of new equipment is amortized by the savings, and that regulatory authorities are on board with any modifications to the planned maintenance schedule. The trend is clearly to include more CBM functions into Maintenance, Repair and Overhaul (MRO) processes so better cost control can be achieved without ever comprising passenger safety.
Rajamani, Ravi
Diagnostics and Prognostics of Aerospace EnginesPT-19511/28/2018
The propulsion system is arguably the most critical part of the aircraft; it certainly is the single most expensive component of the vehicle. Ensuring that engines operate reliably without major maintenance issues is an important goal for all operators, military or commercial. Engine health management (EHM) is a critical piece of this puzzle and has been a part of the engine maintenance for more than five decades. In fact, systematic condition monitoring was introduced for engines before it was applied to other systems on the aircraft. Diagnostics and Prognostics of Aerospace Engines is a collection of technical papers from the archives of SAE International, which introduces the reader to a brief history of EHM, presents some examples of EHM functions, and outlines important future trends. The goal of engine health maintenance is ultimately to reduce the cost of operations by catching problems before they become major issues, by helping reduce repair times through diagnostics, and by facilitating logistic optimization through prognostic estimates. Diagnostics and Prognostics of Aerospace Engines shows that the essence of these goals has not changed over time.
Rajamani, Ravi
SIMPLIFIED APPROACH ON DEVELOPING PREVENTIVE AND CORRECTIVE MAINTENANCE STRATEGY2024-01-36918/7/2018
ABSTRACT Developing preventive and corrective maintenance strategies for military ground vehicles based on asset readiness and lifecycle cost is a challenge due to the complexity associated with the collection and storage of maintenance and failure data in the operational environment. Many of the past reliability centered maintenance efforts have encountered significant challenges in collecting, identifying, accessing, cleaning, enhancing, fusing, and analyzing the data. Another challenge is creating and maintaining complex simulation models that require significant effort and time to produce business value. The work described in this paper is the result of a collaborative effort among multiple US Army organizations to simplify the approach in order to gain valuable insight from the existing data. It is shown how the resulting process can be used to develop simplified models to optimize corrective and preventive maintenance programs. Details are provided on how to work with the existing data sources in order to develop and implement methods at the program management level. The simulation results demonstrate the benefits for the maintenance teams, logistics teams, operation teams, fleet planners, and warfighters.
Gugaratshan, Guga, Srinivasan, Syamala, Harrison, Dean, Castanier, Matthew P., Wade, Jody D., Jones, J. Isaac “Ike”
Flexible Integrated Intelligent Network for Prognostic Health Management SystemsSM_2018_CBM-27672/21/2018
Weapons systems depend on the health of their components to perform reliably over extended periods of time. Maintenance is an important aspect of reliability, assuring equipment is available when needed. Trends toward electronic/networked information present opportunities to improve maintainer effectiveness, but new tools must be developed to manage interactions between the information sources and maintainers. AVNIK, with subcontract team member ISI, researched techniques for Flexible Integrated Intelligent Network (FIIN) for Prognostic Health Management (PHM) systems, based on the artificial intelligence field of cooperative multi-agent distributed work environments. AVNIK also developed prognostics algorithms using statistical anomaly detection and trend analysis. Our approach applies a multi-agent architecture to produce a distributed information management toolset framework enabling the sharing of system health information to aircrew members, aircraft to transmit inflight aircraft health/diagnostics information to ground crews for maintenance operations and allow for the prepositioning of assets and maintenance execution. These tools gather relevant information from diverse sources upon user request or autonomous notification, filter information, and format it for presentation to users.
Little, W.Thomas, Platt, Michele, Yalowitz, Jeff
Wireless Sensor Suite for Electrical System Diagnosis and PrognosisSM_2018_CBM-49942/21/2018
The authors have developed a wireless sensor suite for rotorcraft Generator Control Unit (GCU) and Main Power Relay (MPR) health monitoring. This sensor suite monitors for changes in component characteristics, temperature, vibration extremes, voltage surges, and other factors that will indicate the unit is close to the end of its service life. The sensor suite logs event/triggered or time sequenced/snapshot "smart data" for prognostic and diagnostic estimations, then wirelessly transmits the logged data to maintenance personnel and/or a Health Monitoring and Usage System (HUMS) mounted on the rotorcraft. The system assists in diagnosis and corrective maintenance by capturing data at the time of a fault and aiding in its visualization. This system also assists with Condition Based Maintenance (CBM) by calculating prognostic signatures that indicate to maintenance personnel when failure of select electrical system components is imminent.
Usrey, Michael, Depew, R. Ray, Frediani, Laurel, Schaible, Brian, Schoonover, Dale, Wright, Lewis, Brand, Alex, Kalgren, Patrick, McKown, R. Steve, Duke, Al, Lyman, Chris
Items per page:
1 – 50 of 152