E9
Digital Twins 3

Back to overview

17:35
conference time (CEST, Berlin)
Using True Digital Twins to Develop, Test, and Optimize Complex Systems
26/10/2021 17:35 conference time (CEST, Berlin)
Room: E
J. Jarrett (Kinetic Vision, USA)
J. Jarrett (Kinetic Vision, USA)
The benefits of automation are well known: dramatically lower direct labor costs with increased productivity, and improved fulfillment time and accuracy. Automation also lowers ancillary labor-related costs for areas such as safety, healthcare, employment litigation, and labor negotiation. The important benefits of automation come at a price, however, as the development time and cost to design these systems have also dramatically increased. Physical prototypes of complex systems like large-throughput warehouses, with the associated integration of equipment such as in-line sensors, computer vision, robotics, control software and other systems, as well as the evaluation of those systems with live human test subjects, is extremely costly, time consuming, and potentially unsafe. Presented is a method where complex systems and processes are developed using a virtual environment, commonly known as a Digital Twin. The twin replicates the proposed physical space, and includes all mechanical, robotic, sensor, vision, AI, and computer systems, while simultaneously incorporating real human action and interaction. Humans experience the system in virtual reality (VR), with concurrent motion capture driving their avatars in real time. The avatars interact with virtual objects and their physical “stand ins” which are also motion-tracked. All aspects of the system, including hardware, software, AI training data/models, and human interaction, are developed, tested, and optimized virtually using a software platform running on industrial GPUs such as the NVIDIA RTX 2080 Ti. The result is faster design discovery and iteration at a much lower cost. Once tested and optimized, the virtual system is deployed physically. Team collaboration is also improved, as key stakeholders in training, operations, safety, and other groups can join the environment virtually as needed, even if located thousands of miles apart. The presentation will provide examples of the development techniques and technologies that are used to create and optimize True Digital Twins. This includes multiple video examples of real life applications and their identical virtual twins in the robotic automation, retail, and medical fields.
Digital twins, true digital twin, virtual replica, 3D environments, automation, AI training data, multi-sensor, virtual reality, modeling and simulation, motion-capture, digital replica.
17:55
conference time (CEST, Berlin)
Thermo-Mechanical Modelling, Test Correlation, and Physics/AI-based Model Order Reduction in Gas Turbine Applications
26/10/2021 17:55 conference time (CEST, Berlin)
Room: E
C. Blake, C. Semler (MAYA HTT, CAN)
C. Blake, C. Semler (MAYA HTT, CAN)
In order to remain competitive, gas turbine OEMs must ensure high availability of their products. In addition to designing for product performance and product quality, OEMs provide long-term maintenance plans to their customers to keep their engines working smoothly. Such maintenance plans are made possible by the monitoring and advanced analytics of telemetry data coming from engine assets. For example, detection of problematic operating conditions and prediction of future part failures needing maintenance or overhaul requires collection of data and analytical or machine learning models to process that data. However, implementing and deploying such real-time models necessitates upstream simulation tools. The goal of this paper is to present 3D simulation technology that enables these predictive-maintenance use cases. We will discuss: a) Thermo-mechanical simulation of gas turbines. Here, Siemens Simcenter 3D Thermal-Multiphysics is used to build a complete representation of the whole engine through the use of thousands of separate boundary conditions that adequately represent the transient behavior of the whole engine. Those boundary conditions represent the flow conditions, in various regimes, for a range of rotational speeds and in different sections of the engine. They make use of built-in thermal correlations, of “expressions” that can automatically access specific model data (e.g., fluid temperature, material properties, solve-time computed results, rotational speed or radius), or of proprietary correlations obtained from in-house measurements, experience or CFD. b) Correlation of simulation data to test data. Here, we will discuss a novel adjoint-based sensitivity approach to rapidly compute sensitivities to many design parameters, such as heat-transfer coefficients, and heat-pickup rates, and utilize those sensitivities to rapidly correlate and update a transient thermal-mechanical model to temperature measurements. c) Model-order reduction. Here, we will discuss physics-based (intrusive) and AI-based (non-intrusive) methods to extract lower-dimensional approximations of the whole engine thermo-mechanical model that run in near-real time. The impact of model non-linearities will be emphasized. These reduced-order models can be deployed within edge devices or in the cloud to act as “virtual sensors” or “digital twins” that augment measured telemetry data. Development and deployment of these analytics tools involves several technology bricks and the exchange of large amounts of real-time data between them. As such, we will briefly touch upon the requirements for an industrial data pipeline to facilitate the connection between these bricks. We will demonstrate these topics by means of a real gas-turbine engine model.
thermal analysis, digital twin, reduced-order models, thermal correlation, adjoint solver, multiphysics
18:15
conference time (CEST, Berlin)
On-Demand Auto-Generation of Predictive Digital Twins for Cyber-Physical Systems
26/10/2021 18:15 conference time (CEST, Berlin)
Room: E
S. Coy, G. Gershanok (TimeLike Systems,USA )
S. Coy, G. Gershanok (TimeLike Systems,USA )
Predictive digital twins of cyber-physical systems can be very useful for evaluating and choosing among decision alternatives that arise during development, operations, and maintenance of these systems; applications include model-based engineering, model-based decision support for human decision-makers, and automated decision-making for autonomous systems. In the context of development, the digital twin would typically be based on the current intended design for the given system. In the case of operations and maintenance, however, it is crucial to ensure that the digital twin represents the precise current state of the system, accurately reflecting any and all changes that have occurred over the system’s lifetime, from wear and tear on hardware components, to updates to software components, and any other modifications that may have occurred, e.g. components that have been added or removed, connections that have been made or broken, and any new software that may have been uploaded. One way to accomplish this is to create the digital twin at the same time we create the actual system, and then, every time any change occurs to the real system, we would make the corresponding change to the digital twin. However, this kind of “parallel maintenance” strategy is inefficient, error-prone, and burdensome. A much better approach, in our view, is to implement a facility that enables us to autogenerate an up-to-date digital twin on demand, whenever we need it. The design of such a facility is no small challenge, not when the objective is to make it work for any kind of cyber-physical system, no matter how complex, but it can be done. In this presentation we will describe an innovate component-based software technology for model-based engineering and DevOps of cyber-physical systems that provides just such a facility. An earlier version of this same technology was described in our presentation "Combining Heterogenous Models" at the 2017 NAFEMS conference on Multiscale and Multiphysics Modeling & Simulation - Innovation Enabling Technologies.
Digital Twin, Autonomous Things, Systems Modeling & Simulation, System-Level Simulation, Model-Based Engineering, Model-Based DevOps, Model-Based Decision Support, Model-Based Decision-Making for Autonomous Systems, Combining Heterogeneous Models
18:35
conference time (CEST, Berlin)
Cloud-Based Digital Twins of Overlay Metal Deposition for Responsive Control of Distortion
26/10/2021 18:35 conference time (CEST, Berlin)
Room: E
M. Asadi, M. Fernandez, M. Tanbakuei Kashani, M. Smith (Applus (SKC Engineering), CAN)
M. Asadi, M. Fernandez, M. Tanbakuei Kashani, M. Smith (Applus (SKC Engineering), CAN)
The cognitive computation of a digital-twin becomes time-intensive beyond the requirement of a smart system when it uses simulation tools that solve governing constitutive equations in the form of partial differential equations (PDE). However, such physics-based digital twins offer a critical core-competency that enables the smarting through limited data. The current data-driven digital twins that use machine learning (ML) algorithms take significant initial data to mature for smarting. In most manufacturing processes, there is no such data. Therefore, cyber-manufacturing systems work with limited data and adaptive learning. For example, when fabrication deals with overlay metal deposition (also known as cladding), the digital twin can manage the adverse effects such as distortion and residual stress during the deposition. However, the process of finding an effective deposition pattern is a challenging task given a large number of possible deposition scenarios. We build a hybrid digital-twin that takes advantage of ML algorithms for quick response while gaining fidelity through adaptive learning with FEA simulation tools. Our digital twin consists of a quick learner that is data-driven to be responsive, and it uses an active learning algorithm to wisely navigate the data selection toward a higher training rate and gain fidelity using critical data points rather than an aggregated data set. We use our hybrid digital-twin to explore various overlay scenarios in real-time to form a platform for smart cladding. We present our platform with an actual overlay deposition application on panel structures, including countless depositions scenarios. This digital twin has gained an acceptable fidelity with 100 interactive and iterative labelling queries to FEA. The learner uses a physics-guided machine learning approach to become more informed and reducing data dependency. This hybrid digital-twin is packaged as a cloud-based tool that enables engineers to analyze and compare different patterns to assess fabrication scenarios without computational time delay. Smart systems are not explicit programming; they are architectured to learn continually. Our hybrid digital twin learns from using. The more to use, the higher fidelity evolves in the cloud.
cyber-manufacturing, digital twin, machine learning, metal deposition, overlay pattern, welding, distortion control, adaptive learning
×

[TITLE]

[LISTING

[ABSTRACT]

[DATE]

[ROOM]

[KEYWORDS]