“Innervated” Pipelines: A New Technology Platform for In-Situ Repair and Embedded Intelligence PI: Dr. Paul Ohodnicki, University of Pittsburgh (Pitt) Team Members / Co-PIs: Dr. Kevin Chen, Dr. Jung-Kun Lee, University of Pittsburgh (Pitt) Dr. Glenn Grant, Dr. Kayte Denslow, Dr. Christopher Smith, Pacific Northwest National Laboratory (PNNL) Demonstrating in-situ repair and fiber optic sensor deployment through robotic deployable cold-spray, combined with a fusion of acoustic NDE and distributed fiber optic sensing in an artificial intelligence-based classification and diagnostic framework for asset health monitoring. Total Project Cost: $1.0M Length 12 mo. Project Vision
36
Embed
“Innervated” Pipelines: A New Technology Platform for In ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
“Innervated” Pipelines: A New Technology Platform
for In-Situ Repair and Embedded Intelligence
PI:Dr. Paul Ohodnicki, University of Pittsburgh (Pitt)
Team Members / Co-PIs:Dr. Kevin Chen, Dr. Jung-Kun Lee, University of Pittsburgh (Pitt)
Dr. Glenn Grant, Dr. Kayte Denslow, Dr. Christopher Smith, Pacific
Northwest National Laboratory (PNNL)
Demonstrating in-situ repair and fiber optic sensor deployment
through robotic deployable cold-spray, combined with a fusion
of acoustic NDE and distributed fiber optic sensing in an
artificial intelligence-based classification and diagnostic
framework for asset health monitoring. Total Project Cost: $1.0M
Length 12 mo.
Project Vision
The Concept : Big Picture
“Innervated” Pipelines of the Future
In-Situ Repair + Embedded Intelligence + Digital Asset Modeling
Focus of Initial 12-Month Project Efforts : Targeted Feasibility Demonstrations
• Cold-spray process: Cold spray is a proven high-rate metal deposition process where metal powders (~5-45 μm particles) are combined with hot gas, accelerated to high velocity (Mach 1-4) and deposited on repair area to build up thickness
• Fully dense, thick metal deposit, metallugically bonded to substrate
• Can repair through-holes and “build back” corrosion allowance
• Can embed sensors in the wall for condition assessment
The concept is to make a pipe-in-a-pipe that may offer structural credits far superior to polymer liner options
The Concept :
Advanced Acoustic NDE / Optical Fiber Methods
Propose New Advanced NDE Interrogation Techniques, and Low-Cost Cold-Spray Embedded Fiber Optics Sensing.
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Team Capabilities and Contributions
Custom Developed, Low-Cost Distributed Fiber Optic Sensors and Interrogators
Fiber Optic Coatings for Protection / Corrosion Sensing
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Team Capabilities and Contributions
Reduced Order Modeling
The Team : Overall Project StructureProject Organizational Chart for Year 1 Efforts
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Team Capabilities and Contributions
Project Objectives
Primary Focus in First Year Efforts
‣ Key Innovations to Be Pursued in year 1
– Cold-spray in-situ repair (primary year 1 focus)
– Internally deployed distributed fiber optics
– Multiphysics modeling for acoustic signatures of defects
‣ Primary Risks to Be Mitigated in year 1
– Basic feasibility of cold spray repair (primary year 1 focus)
– Acoustic, temperature, and strain distribution – defect free
– Characteristic distributions for representative defects
– Sampling of expected sensing signatures for both fiber optics and acoustic NDE methods
Computational Simulations
Inputs: Pipeline + defect + gas transport parameters
Simulated Dataset
AI Model
Defect
Output: Defect detection
Lab Experiments
Reduced Order Model
Advanced Data Analytics
Distributed Sensor dataset
Use multi-physics
model to simulate
sensor response
Accelerate simulations
using reduced order
modeling
Simulate datasets of
healthy & defect
scenarios for training
AI model
Conduct lab-scale
experiments to collect
sensor data for healthy
& defect scenarios
Use advanced data
analytics to automate
the processing of large
datasets
Collect labeled
distributed sensor data
for AI model training
Initial Emphasis
Project Objectives
Technology to Market
‣ Establishment and Meeting of Industry Advisory Group
– Assess cold-spray coating and optical fiber embedding
– Economic assessment
– Regulatory considerations
– Scoping of various deployment scenarios
‣ Draft / first completed technology to market plan
10Economic Modeling, Regulatory, Deployment, Related R & D
Dewitt Burdeaux Mat Podskarbi
Marius Ellingsen
Kent Weisenberg
Ruishu Wright
Industry Advisory Committee
Project Objectives : First Year Project Timeline
11
Project Management / T2M
Cold-Spray Process Development and Validation
Fiber Optic Sensor Deployment, Embedding, and Acoustic Modeling
Q3Q2Q1 Q4
Initial Coating Validation Experiments
Robotic Deployment Tool Modification
Coating Validation Experiments
Go/No-Go : Robotic
Deployment Tool Modification
Industry Advisory Group Meeting
First Iteration of T2M Plan
First Completed Draft T2M Plan
Industry Advisory Group Meeting
Fiber Optic Sensor Embedding with
Cold SprayFiber Optic
Interrogation with Acoustic
Excitation
Completed Fiber Optic Robotic
Deployment Tool Design
Results : Prior PNNL Motivating Cold-Spray Repair Research
12
‣ Cold Spray Repair of Hydroturbine Impellers in-situ at large hydroelectric facilities
‣ Cold spray repair of SCC on Dry storage containers for nuclear spent fuel
Results indicate cold-spray repair can
produce conditions with 5 to 10x
improvement in cavitation erosion
resistance over arc welding repair
and 2 to 4x improvement over original
material
Robotic platforms
have been designed to
support cold-spray
nozzles and feed
hoses and have been
tested for repair of dry
storage containers
Results : Prior Pitt Motivating Optical Fiber Sensing + AI
13
Discrete In-Fiber Sensor Devices Allow for Quasi-Distributed Sensing
Ultrafast (fs) Laser Processing In Line Fabry-Perot Sensors
FFT
M. Wang, P. Ohodnicki, P. Lu, K. Chen et al., Optics Express, 28 (14), 20225 (2020).
Ultrafast Laser Processing to Fabricate In-Fiber Devices for Low-Cost Sensing
Convolutional Neural NetworkSupervised Learning
Defects Acoustic
excitation
s
Discrete Acoustic Sensing Devices Combined with AI Classification Frameworks Allow for Development of Pattern Recognition Schemes (Infrastructure Security, Faults, etc.)
H. Wen, P. Ohodnicki, K. Chen et al., 2018 Asia Communications and Photonics Conference.
Challenges and Risks
Challenges
‣ Optimization of process parameters for cold-spray repair coatings and fiber optic sensor embedding
‣ Protective packaging of fiber optic sensors for cold-spray embedding internal to the pipe
‣ Development of robotic deployment strategies for both cold-spray repair and fiber optic installation and embedding
‣ Pitt and PNNL at Partial Capacity Due to COVID-19
Risk Mitigation
‣ Industry Advisor Group Engagement During Initial Feasibility Demonstrations of Year 1 to Ensure Deployment Compatibility
‣ Fiber Optic Deployment Tool Demonstration in Year 1 to Integrate with Cold-Spray Embedding in Subsequent Years
Potential Partnerships
‣ Potential to Enhance Project Outcomes / Impacts Including:
– Identifying defects and failure modes for physics-based simulations of acoustic signatures
– Expanded capabilities for performance testing of coated pipeline materials by cold-spray
– Opportunities for field validation of distributed fiber optic sensor technology, including in-pipe integration
– Adopting deployment strategies from other technologies
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Team Capabilities and Contributions
Cold-Spray Electromagnetic Transducers
The Concept :
Distributed Optical Fiber SensingExternal Deployed Fiber Optic Sensors are Commercially Available
Propose New Internal Deployed, Low-Cost Cold-Spray Embedded Fiber Optics and Functionalization for Corrosion Monitoring.
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Team Capabilities and Contributions
Custom Developed, Low-Cost Distributed Fiber Optic Sensors and Interrogators
Fiber Optic Coatings for Protection / Corrosion Sensing
▸ Lower temperature process compared to conventional repair techniques (e.g., welding), and method of application results in reduced thermal input to area of repair
– No heat affected zone (HAZ) that alters the microstructure of the base metal as occurs with traditional fusion-based welding operations
– No need for post-repair stress relieving
– No ignition hazard from flame, required pre-grinding operations, or other operations consisting of spark-generating moving parts
▸ Can achieve high deposition rates, on the order of 350 g/min
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Team Capabilities and Contributions
Research Focus: Cold Spray, Solid Phase Joining and Processing (Friction Stir Welding), Leads projects in Robotic platform integration
Research Focus: Advanced manufacturing, forming, joining, thermomechanical processing. Leads programs in Solid Phase Processing
Cold-Spray Coating Technology and DeploymentCoating Materials and Corrosion ProtectionFiber Optic Sensor EmbeddingNon-destructive Evaluation (NDE) Techniques
Digital Twin Framework DevelopmentPhysics-Based Modeling with Accelerated SolverFEA Based Application for Defect SignaturesCoating and Sensor Valuation and Optimization
Regulatory FrameworkIndustrial PerspectiveIn-Line Inspection DataGIS / Digital Twin Framework Interface
Team Capabilities and Contributions
NDE / EMAT Facilities Summary
Pipeline Test Loop Facilities
Cold Spray Facility Summary• VRC GEN 3 High Velocity System• Applicator end effector nozzle on
stationary robot platform• Rotary Positioning System
Cold Spray Project Objective
‣ Develop the process to apply a fully dense metal deposit of >5 mm thickness to the inside of a legacy pipeline to repair damaged material, build back corrosion allowance, and potentially heal small through-wall penetrations….all without excavation.
‣ Benchmark the process speed and material costs (gas and powder) on a 4 foot length of pipe 10 -12 inches in diameter
‣ Transfer the process, materials, and equipment designs to commercial entities to implement the process on a robotic crawler capable of travelling inside a pipe.
26
Cold Spray Project Approach
27
Title Task Description (76716)Task 1 PM Project Management
Task 2 Cold Spray
Development Flat Plate
Apply cold spray coating to bare steel / cast iron coupons
(Milestones M1.2,M2.1)
Task 3 Cold Spray
Development Pipe Inner
subsections
Apply cold spray coating to bare steel / cast iron pipe sections,
develop nozzle for inside diameter CS (Milestone 2.2)
Discrete In-Fiber Sensor Devices Allow for Quasi-Distributed Sensing
Ultrafast (fs) Laser Processing In Line Fabry-Perot Sensors
In Line Fabry-Perot Sensors Array with Different Cavity Lengths
FFT
M. Wang, P. Ohodnicki, P. Lu, K. Chen et al., Optics Express, 28 (14), 20225 (2020).
Ultrafast Laser Processing to Fabricate In-Fiber Devices for Low-Cost Sensing
Results : Prior Pitt Motivating Distributed Acoustic Sensing + AI
33
Discrete Acoustic Sensing Devices Combined with AI Classification Frameworks Allow for Development of Pattern Recognition Schemes (Infrastructure Security, Faults, etc.)
Convolutional Neural NetworkSupervised Learning
H. Wen, P. Ohodnicki, K. Chen et al., 2018 Asia Communications and Photonics Conference.
Defects Acoustic
excitation
s
Low-Cost Distributed Sensing Integrated with AI Classification Frameworks
Pipeline internal corrosion monitoring (Jiang et al, Struct Control Health Monit. 2017)
Application of Distributed Strain Sensing for Local Corrosion Detection
Results : Prior Pitt Motivating AI Based Reduced Order Modeling
35
Accelerated Simulation of Sensing Responses for Defects and Pipeline Configurations to Accelerate AI-Framework Training for Defect Localization / Identification
Convolutional Neural NetworkSupervised Learning
Physics-Based AI Approach to Reduced Order Modeling of Complex Physical PhenomenaData sampling (b) Dimensionality reduction