Structural Health Monitoring : Role of Modelling and Simulation

Fumio KOJIMA Organization of Advanced Science and Technology Graduate School of Sytem Informatics, Kobe University, Kobe, Japan (Phone/Fax +81-78-803-6493; E-mail: kojima@koala.kobe-u.ac.jp)

システムモデル論（第３回）　H23.10.19 モデリングI

Scenario of the talk

1.  Background and Aim of Research 2.  Structural Health Monitoring using Inverse Analysis 3.  Current and New Direction for SHM

Research Project

Projects of System Group:   Inverse Problems in Nondestructive Evaluation

  Condition Monitoring

  New NDE for Nanoscale Defects

Cooperative Organization

 CRSC, North Carolina Sate University, NC USA  National Institute for Applied Mechanics, HCMC, Vietnam

Inverse Problems in Nondestructive Evaluation NASA

AIST

JAXA

Cyber-Infrastructure of Remote NDE System: Electrical Potential Method:

Scenario of the talk

1.  Background and Aim of Research 2.  Structural Health Monitoring using Inverse Analysis 3.  Current and New Direction for SHM

Remark: Structural health monitoring is a promising technology in keeping safety of large scale complex systems, such as nuclear power plants, airplanes, etc. It involves the broad concept of assessing ongoing and in-service performance of structures using variety of measurements. Those elements include sensors in structures, data acquisition, data management, data interpretation, diagnosis, etc.

First Project

NASA Langley Research Center, Hampton Virginia

Nondestructive Evaluation Science Branch (1987~)

January 28, 1986

Abridgement

IX MAINTENANCE SAFEGUARDS

Installation, test, and maintenance procedures must be especially rigorous for Space Shuttle items designated Criticality 1. NASA should establish a system of analyzing and reporting performance trends of such items. Maintenance procedures for such items should be specified in the Critical Items List, especially for those such as the liquid-fueled main engines, which require unstinting maintenance and overhaul. With regard to the Orbiters, NASA should: Develop and execute a comprehensive maintenance inspection plan. Perform periodic structural inspections when scheduled and not permit them to be waived. Restore and support the maintenance and spare parts programs, and stop the practice of removing parts from one Orbiter to supply another.

Measurement Equipment at NASA Langley Research Center

(1988, ABC World News)

Shape Reconstruction using Thermal Data

Heat Lamp Heat Lamp

Infrared Camera Input ：Heat-up at Surface

Output：Surface Measurement

Inspection Model

(Inverse Problems, 1991)

Why do we need to estimate material defect?

  It is important to evaluate material system used in passenger flight, bridge, nuclear power plant for their safety.

  Health monitoring techniques are available using various kinds nondestructive test but it is not enough information from current technologies.

  Computational techniques play key role in maintenance technologies

Nondestructive Evaluation Science Measurements

+ High performance Computing

Type of Nuclear Power Plants There are two separate coolant loops in PWRs. Heating the water in the primary coolant loop by thermal conduction through the fuel cladding and it is pumped into the steam generator, where heat is transferred to the lower pressure secondary coolant.

BWR has only one coolant loop. Heat is produced by nuclear fission in the reactor core and the producing steam is directly used to drive a turbine. In BWR, the core shroud is a large stainless steel cylinder of circumferentially welded plates surrounding the reactor fuel core.

The fast breeder reactor (FBR) is the next generation of NPPs and a fast neutron reactor designed to breed fuel by producing more fissile material than it consumes.

JSFR Monjyu

Conventional Inspection Technologies　(ECT)

In PWR, the steam generator tubes constitute one of the primary barriers between the radioactive and non-radioactive sides of the plant. Therefore, in-service inspections (ISI) of the steam generator tubes are essential in keeping safety of operations.

Crack Reconstruction using Inverse Analysis

Eddy Current Signal Shape Reconstruction

International Benchmark Problem (1999-2000)

Primary coolant loop

Secondary coolant

Conventional Inspection Technology (UT)

Scenario of the talk

1.  About Our Laboratory 2.  Background and Aim of Research 3.  Structural Health Monitoring using Inverse Analysis 4.  Current and New Direction for SHM

Remark: Structural health monitoring is a promising technology in keeping safety of large scale complex systems, such as nuclear power plants, airplanes, etc. It involves the broad concept of assessing ongoing and in-service performance of structures using variety of measurements. Those elements include sensors in structures, data acquisition, data management, data interpretation, diagnosis, etc.

Our Strategies

 Mathematical modelling of NDT and defect profiles  Forward analysis for constructing virtual inspection images  Inverse analysis for model-based NDT

A direct problem is to design a real NDT system mathematically using the input and output relation with the appropriate admissible class of material flaws, while an inverse problem is to construct a method for recovering and/or visualizing material flaw information under the mathematical formulation of the direct problem.

Three Procedures:

SCC Monitoring In BWR, the core shroud is a large stainless steel cylinder of circumferentially welded plates surrounding the reactor fuel core. The shroud provides for the core geometry of the fuel bundles. Extensive cracking of circumferential welds on the core shroud has been discovered in an increasing number of Japanese BWRs plants since 2002.

Network Tomography

Grand test (Natioanl Project 2004-2006)

Network tomography implements the hybrid use of real-virtual system in NDT environments

Ground Test

Inspection Module

Computational Module

Communication Interface Module

Inverse Analysis

Computational Agent

Evaluator

Mobile Interface

Data Sending

Underwater Manipulator

Sensor Plate Curved Corner

Measurement Server

Probe

ECT Sensor

Operator

National Project (Hitachi Ltd., Tohoku Univ., Polytech. Univ. and Kobe Univ., 2004-2006)

Mathematical Modeling of NDT (Procedure I)

CrackTransmitterRecieverEddy CurrentConductor

Induced voltage

Electromagnetic Field in Sample

Simulator (FEM)

FEM matrix BEM matrix

crack

Fast Solver using Domain Decomposition Method

Finite Elements

Crack Elements

Crack model

Conductivity vector

Modeling of SCC

Surface of natural crack Finite element cracking model

Stress corrosion cracking is quite complicated structures for sizing. In our approach , complexity of real cracks are approximated by a set of rectangular solids. Each solid is characterized by its own properties, such as length, width, depth, and physical parameters.

Modeling of Defect Profiles (SCC)

Forward Analysis for ECT (Procedure II)

Measurement

  Frequency Control   Scanning Strategy

ＥＣＴ Simulator

Defect Parameter   Conductivities   Position   Number of Defects

Virtual Image

Inverse Analysis (Procedure III)

ECT Measurements

Crack shape ?

Crack Parameter

Crack Model Evaluation Update crack

parameter

Virtual Image

Forward Analysis

Inverse Analysis

＋

‐

Measurement Image Crack Parameter :

Test Example (SUS sample with SCC)

-0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 0 0.01 0.02 0.03 0.04X1 (m) 0 0.01 0.02 0.03 0.04 0.05X2 (m)

True Shape Estimate Shape

  In this lecture, it was shown that modeling and simulation of NDT is indispensable problem-solving methodology for implementing high performance monitoring of NPPs.

  Electromagnetic inverse methodologies for sizing stress corrosion crack were considered for boiled water reactor (BWR) plants. It is very crucial to characterize target cracking of SUS304 and SUS316L materials used in BWR plants since those involve various kind of complexities, such as orientation, multiple deep-lying branching, partially conducting, etc.

  The model based health monitoring system was outlined for sizing natural crack in three dimensions. The feasibility and validity of our inverse algorithm using laboratory data were demonstrated.

Scenario of the talk

1.  About Our Laboratory 2.  Background and Aim of Research 3.  Structural Health Monitoring using Inverse Analysis 4.  Current and New Direction for SHM

Remark: Structural health monitoring is a promising technology in keeping safety of large scale complex systems, such as nuclear power plants, airplanes, etc. It involves the broad concept of assessing ongoing and in-service performance of structures using variety of measurements. Those elements include sensors in structures, data acquisition, data management, data interpretation, diagnosis, etc.

Critical Issues From Time Based Maintenance (TBM) to Condition Based Maintenance (CBM),… What kinds of philosophy (or concept) of NDT is necessary?

Condition monitoring plays an essential role!

Modeling methodologies of material failure progress

Research Direction

Data assimilation

Predictive model

Feasible management for sensor networks in aircraft

Lifetime estimate of aircraft components using stochastic inverse analysis

New Methodology for Health Monitoring based on s Stochastic Simulations