An official website of the United States government
Digital twin is a vital enabling technology for smart manufacturing in the era of Industry 4.0. Digital twin effectively replicates its physical asset enabling easy visualization, smart decision-making and cognitive capability in the system. In this paper, a framework of dynamic data driven digital twin for complex engineering products was proposed. To illustrate the proposed framework, an example of health management on aircraft engines was studied. This framework models the digital twin by extracting information from the various sensors and Industry Internet of Things (IIoT) monitoring the remaining useful life (RUL) of an engine in both cyber and physical domains. Then, with sensor measurements selected from linear degradation models, a long short-term memory (LSTM) neural network is proposed to dynamically update the digital twin, which can estimate the most up-to-date RUL of the physical aircraft engine. Through comparison with other machine learning algorithms, including similarity based linear regression and feed forward neural network, on RUL modelling, this LSTM based dynamical data driven digital twin provides a promising tool to accurately replicate the health status of aircraft engines. This digital twin based RUL technique can also be extended for health management and remote operation of manufacturing systems.
more »
« less
A Digital Twin Framework Utilizing Machine Learning for Robust Predictive Maintenance: Enhancing Tire Health Monitoring
Abstract We introduce a novel digital twin framework for predictive maintenance of physical systems with long term operations. Using monitoring tire health as an application, we show how the digital twin framework is used to enhance automotive safety and efficiency, while overcoming technical challenges using a three-step approach. Firstly, for managing the data complexity over a long operation span, we employ data reduction techniques to concisely represent physical tires using historical performance and usage data. Relying on this data, for fast real-time prediction, we train a transformer-based model offline on our concise dataset to predict future tire health over time, represented as Remaining Casing Potential (RCP). Based on our architecture, our model quantifies both epistemic and aleatoric uncertainty, providing reliable confidence intervals around predicted RCP. Secondly, we incorporate real-time data by updating the predictive model in the digital twin framework, ensuring its accuracy throughout its life span with the aid of hybrid modeling and the use of a discrepancy function. Thirdly, to assist decision making in predictive maintenance, we implement a Tire State Decision Algorithm, which strategically determines the optimal timing for tire replacement based on RCP forecasted by our transformer model. This three-step approach ensures that our digital twin not only accurately predicts the health of a system, but also continually refines its digital representation and makes predictive maintenance decisions for removal from service. Our proposed digital twin framework embodies a physical system accurately and leverages big data and machine learning for predictive maintenance, model update and decision making.
more »
« less
- Award ID(s):
- 2133630
- PAR ID:
- 10582207
- Publisher / Repository:
- American Society of Mechanical Engineers
- Date Published:
- ISBN:
- 978-0-7918-8834-6
- Format(s):
- Medium: X
- Location:
- Washington, DC, USA
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
We articulate the design imperatives for machine learning based digital twins for nonlinear dynamical systems, which can be used to monitor the “health” of the system and anticipate future collapse. The fundamental requirement for digital twins of nonlinear dynamical systems is dynamical evolution: the digital twin must be able to evolve its dynamical state at the present time to the next time step without further state input—a requirement that reservoir computing naturally meets. We conduct extensive tests using prototypical systems from optics, ecology, and climate, where the respective specific examples are a chaotic CO2 laser system, a model of phytoplankton subject to seasonality, and the Lorenz-96 climate network. We demonstrate that, with a single or parallel reservoir computer, the digital twins are capable of a variety of challenging forecasting and monitoring tasks. Our digital twin has the following capabilities: (1) extrapolating the dynamics of the target system to predict how it may respond to a changing dynamical environment, e.g., a driving signal that it has never experienced before, (2) making continual forecasting and monitoring with sparse real-time updates under non-stationary external driving, (3) inferring hidden variables in the target system and accurately reproducing/predicting their dynamical evolution, (4) adapting to external driving of different waveform, and (5) extrapolating the global bifurcation behaviors to network systems of different sizes. These features make our digital twins appealing in applications, such as monitoring the health of critical systems and forecasting their potential collapse induced by environmental changes or perturbations. Such systems can be an infrastructure, an ecosystem, or a regional climate system.more » « less
-
Arai, K. (Ed.)Digital twin technology has found wide applications in the aerospace industry, for fleet monitoring, diagnostics, predictive maintenance, and manufacturing. Sensing the aero-engine in a test environment is challenging, due to issues with noisy and failed sensors, introduced by extreme conditions. Downstream diagnostics and prognostics require that key sensed values are available for the duration of the test for both real-time and off-line analysis. Virtual sensors that mimic the behavior of the physical sensors provide a resilient solution in the presence of such failures. This paper describes virtual sensors designed using a low-cost digital twin platform with off-the-shelf analytical software components and libraries. The platform is a no-code environment, that permits users to rapidly experiment with different analytical models and configurations. We show that virtual sensors can emulate the behavior of the physical sensor(s) in the event of multiple physical sensor failures with high accuracy, the design of which is facilitated by the Digital Twin platform. From a process standpoint, the Digital Twin results in several advantages to the organization including the breaking down of departmental data silos, reevaluation of key assumptions regarding system design, and the standardization of monitoring process. The digital twin approach is seen to be a catalyst for reengineering the design and monitoring lifecycle for industrial organizations.more » « less
-
Urban drainage systems face increased floods and combined sewer overflows due to climate change and population growth. To manage these hazards, cities are seeking stormwater digital twins that integrate sensor data with hydraulic models for real-time response. However, these efforts are complicated by unreliable sensor data, imperfect hydrologic models, and inaccurate rainfall forecasts. To address these issues, we introduce a stormwater digital twin system that uses online data assimilation to estimate stormwater depths and discharges under sensor and model uncertainty. We first derive a novel state estimation scheme based on Extended Kalman Filtering that fuses sensor data into a hydraulic model while simultaneously detecting and removing faulty measurements. The system’s accuracy is evaluated through a long-term deployment in Austin’s flood-prone Waller Creek watershed. The digital twin model demonstrates enhanced accuracy in estimating stormwater depths at ungauged locations and delivers more accurate near-term forecasts. Moreover, it effectively identifies and removes sensor faults from streaming data, achieving a Receiver Operating Characteristic Area Under the Curve (ROC AUC) of over 0.99 and significantly reducing the potential for false flood alarms. This study provides a complete software implementation, offering water managers a reliable framework for real-time monitoring, rapid flood response, predictive maintenance, and active control of sewer systems.more » « less
-
Digital twins provide a cost-effective means of evaluating performance, predicting network changes, and enhancing network administration and decision-making processes. However, acquiring detailed data for digital twin development remains a challenge due to commercial system constraints. City-scale testbeds, like COSMOS, offer practical solutions, aiding in data collection for modelling of digital twins. In this paper, we utilize data from experiments on the COSMOS testbed to design a digital twin model for the accumulation of gain ripple in cascaded Erbium-doped fiber amplifiers (EDFAs). We quantify the gain ripple in terms of its peak-to-valley ratios and identify optimal EDFA combinations. Moreover, we explore the impact of system parameters, and validate the proposed digital model through comparison with experimental results. We show that the differences between the digital twin predictions and experimental results are <0.1 dB.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1115/DETC2024-140496
