Digital Twin Technologies to Improve Mission Readiness and Sustainment

Navy SBIR 20.2 - Topic N202-105

Naval Air Systems Command (NAVAIR) - Ms. Donna Attick [email protected]

Opens: June 3, 2020 - Closes: July 2, 2020 (12:00 pm ET)

 

 

N202-105       TITLE: Digital Twin Technologies to Improve Mission Readiness and Sustainment

 

RT&L FOCUS AREA(S): Artificial Intelligence/ Machine Learning, General Warfighting Requirements (GWR)

TECHNOLOGY AREA(S): Air Platform

 

OBJECTIVE: Demonstrate the application of digital twin technologies by developing a virtual model of any naval aircraft product, derive benefits (such as predictive capabilities), and show its impact on total life cycle cost of the product. The models representing any naval aircraft product must allow real-time monitoring of performance and facilitate in-flight service of the product.

 

DESCRIPTION: A need exists for digital twin technologies, which allows the digital footprint of any product to permeate from design inception, through development, sustainment, and finally to disposal (i.e., the entire product lifecycle). Digital twin technology is viable and allows access to the digital image of the asset in real time, leading to secure actionable information that will improve a process, product, or service of any organization [Ref 1].

 

The concept has been around for a while, as shown during the disaster of the Apollo 13 mission. NASA demonstrated the technology with a mirrored system on the ground, which rescued the flight, and is further illustrated in Reference 2. Digital twin technology involves creating a virtual representation of a physical product. Digital twins are powered by machine learning algorithms and are continuously learning systems. The products are connected in a cloud-based environment that receives the data from the sensors. The input data is analyzed and compared to the organization�s baseline data to identify actionable information. The goal for the digital twin technology is to create, test, and build a product in a virtual environment and demonstrate improved product design, monitor product health to identify potential degradation, and simulate manufacturing processes. This will allow real-time monitoring and preventive maintenance of the product, which will reduce product life cycle costs. In the future, every physical product will have a virtual replica (model) hosted in the cloud and enriched every day with operational data to make the model more robust.

�

Currently, model-based system engineering (MBSE) approaches, which move the record of authority from documents to digital models managed in a data-rich environment, are used to build products. MBSE approaches enable organizations to understand design change impacts, communicate design intent, and analyze and predict product design before it is built. System architecture models are developed when MBSE integration occurs across multiple domains such as program management, product support, manufacturing (involving analytical), verification, software, and mechanical and electrical models.

 

Advancements in information technologies (such as computational capabilities, the internet, cloud environment, internet of things enabled by sensors with connectivity and bandwidth factors, and cyber communication) are making the virtual space significant; in this virtual space, analog data from the physical space is converted into digital data that can be easily stored, analyzed, and displayed. Currently, there are huge gaps in the information technologies mentioned; therefore, significant research and development is needed. The combination of information technologies with MBSE enables digital twin technologies [Ref 3].���

 

Digital twin technology is not widespread due to the requirements of prohibitive computing power needs, accessibility, bandwidth, and storage issues. Lack of robust data analytics aided by artificial intelligence, machine learning techniques, and visualization tools is impeding technology development. Digital twin technology has the potential to improve supply chain integrity, flight safety, in-flight service, condition-based maintenance, foreign object detection, and predictive maintenance. For example, developing any predictive maintenance algorithm requires sensor data, which can be utilized to train a classification algorithm for fault detection. This algorithm is used for verification and is installed as a code to the control unit of the product. It is nearly impossible to create the fault conditions necessary for training a predictive maintenance algorithm on the actual product. A solution to this challenge is to create a digital twin of the product (a model), and apply simulation and analysis of sensor data for various fault conditions. A neural network detects abnormal patterns of the sensor data, reflects the trends in predictive models, which are then used to predict failures, and allows tests for all fault conditions with severity. The entire procedure should be automated, thereby allowing tests of �what-if� scenarios on the digital twin model.

 

Predictive maintenance helps to determine when an aircraft product needs maintenance. It reduces downtime and prevents product failure by enabling maintenance to be scheduled based on the actual need rather than at predetermined intervals. It can be used to calculate maintenance-related parameters (MTBF � Mean Time between Failures), forecast the behavior of the product under different circumstances, and simulate different maintenance scenarios. Thus, predictive maintenance capability helps to extend the product life and reduce total ownership costs. Collectively, it will contribute to improving the Navy�s mission readiness and sustainment significantly.

 

It is envisioned that we will be able to develop a virtual integrated model-based representation of a physical product, allow the simulation of the product in a real setting in a dynamic fashion, and demonstrate closed loops between the virtual and physical space. Challenges for this include developing an accurate model that precisely reflects the physical twin�s properties. To improve the models further, a digital twin also requires remodeling based on the changes in the product�s configuration. For predicting failures, detailed blueprints of a product�s failure modes are required. Since the digital twin is a replica of the physical product itself, the requirements, qualification, and certification necessary to determine the flight worthiness of the product are the same for the virtual model as well. The expected outcomes of the effort are real-time monitoring and in-flight service of the product, since the digital twin represents an advanced engineered product. This will enable prolonged product life to deliver capabilities continually.

 

Any product used in naval aircraft can be considered for the proof of concept demonstration (e.g., propulsion engine, electrical power system, fuel system, avionics, air vehicle, auxiliary support equipment, electronic warfare system, human-machine interface).

 

Work produced in Phase II may become classified. Note: The prospective contractor(s) must be U.S. owned and operated with no foreign influence as defined by DoD 5220.22-M, National Industrial Security Program Operating Manual, unless acceptable mitigating procedures can and have been implemented and approved by the Defense Counterintelligence and Security Agency (DCSA). The selected contractor and/or subcontractor must be able to acquire and maintain a secret level facility and Personnel Security Clearances. This will allow contractor personnel to perform on advanced phases of this project as set forth by DCSA and NAVAIR in order to gain access to classified information pertaining to the national defense of the United States and its allies; this will be an inherent requirement. The selected company will be required to safeguard classified material IAW DoD 5220.22-M during the advanced phases of this contract.

 

PHASE I: Design and develop a concept to create a digital twin of a product to show its present state using a model. Develop digital twin processes in the product life cycle � design stage to the field use, maintain and sustain in the real-world case. For validation, demonstrate the closed loop that would exist between physical and virtual space. Apply modeling, simulation, and analysis as necessary. Phase I will include prototype plans for Phase II.

 

PHASE II: Develop a prototype product (a high-fidelity model) by integrating the physical asset to the digital twin and demonstrate the closed loop between physical - virtual - physical space. Demonstrate the applicability of readiness and sustainment influencing factors such as condition-based maintenance, foreign object detection, predictive maintenance, and flight safety with quantifiable metrics. Quantify the cost benefits, such as reduction in the operation cost and total lifecycle cost, as applicable.

Demonstrate the applicability of �what-if� scenarios tested against factors such as product performance management, manufacturing processes, and Navy-unique harsh environmental operating conditions. Demonstrate the scalability of the digital technology to multiple products of an aircraft.

Work in Phase II may become classified. Please see note in the Description section.

 

PHASE III DUAL USE APPLICATIONS: Develop robust architecture, showing the linkage between connectivity and services. Demonstrate the integration of the product into naval aircraft and perform final testing. Successfully transition, implement, and insert the technology for warfighter benefits. Develop mobile application solutions as applicable.

 

Aerospace industry, Manufacturing, Automobile sectors will benefit from the digital twin technology. The successful demonstration of the digital twin of the product that is operationalized will enable the applicability of the approach to any product/process/service industry to achieve cost benefits.

 

REFERENCES:

1. �Industry 4.0 and the Digital Twin: Manufacturing Meets Its Match.� Deloitte University Press, 2017. https://www2.deloitte.com/content/dam/Deloitte/cn/Documents/cip/deloitte-cn-cip-industry-4-0-digital-twin-technology-en-171215.pdf

 

2. Grieves, M. & Vickers, J. �Digital Twin: Mitigating Unpredictable, Undesirable Emergent Behavior in Complex Systems (Excerpt).� Springer, 2017. https://research.fit.edu/media/site-specific/researchfitedu/camid/documents/Origin-and-Types-of-the-Digital-Twin.pdf

 

3. Tao, F., Zhang, M. & Nee, A. �Digital Twin Driven Smart Manufacturing.� Elsevier, 2019. https://www.sciencedirect.com/book/9780128176306/digital-twin-driven-smart-manufacturing

 

KEYWORDS: Digital Double, Artificial Intelligence, Machine Learning, Data Strategy, Architecture, Internet of Things, Cloud, Digital Twin