Simulation measurement and Image analysis of corrosion PDF

Preview

Summary

Virginia Commonwealth UniversityVCU Scholars CompassTheses and Dissertations Graduate School2011Simulation, measurement and Image analysis ofcorrosion initiation and growth rate for Aluminum2024 and Steel 304Yan Fang Virginia Commonwealth UniversityFollow this and additional works at:scholarscompass...

Description

Virginia Commonwealth University

VCU Scholars Compass Theses and Dissertations

Graduate School

2011

Simulation, measurement and Image analysis of corrosion initiation and growth rate for Aluminum 2024 and Steel 304 Yan Fang Virginia Commonwealth University

Follow this and additional works at: http://scholarscompass.vcu.edu/etd Part of the Engineering Commons © The Author

Recommended Citation Fang, Yan, "Simulation, measurement and Image analysis of corrosion initiation and growth rate for Aluminum 2024 and Steel 304" (2011). VCU Theses and Dissertations. Paper 2569.

This Thesis is brought to you for free and open access by the Graduate School at VCU Scholars Compass. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of VCU Scholars Compass. For more information, please contact [email protected].

ii

School of Engineering Virginia Commonwealth University

This is to certify that the thesis prepared by Yan Fang entitled Image analysis of corrosion growth rate of Aluminum and Steel has been approved by his or her committee as satisfactory completion of the thesis or dissertation requirement for the degree of master of Mechanical & Nuclear Engineer

Dr. Hinderliter Mechanical and Nuclear Engineering, School of engineering

_______________________________________________________________________ Dr. Pidaparti , Mechanical Engineering, School of engineering

Dr. Meng Yu, Computer Science, School of engineering

Dr. Karla Mossi, Mechanical Engineering, School of engineering

Dr. Gary C. Tepper, Mechanical Engineering, School of engineering

Dr. Russell Jamison, School of engineering

Dr. F. Douglas Boudinot, Dean of the School of Graduate Studies

August 12th 2011

iii

© Yan Fang 2011 All Rights Reserved

iv

SIMULATION, MEASUREMENT AND IMAGE ANALYSIS OF CORROSION INITIATION AND GROWTH RATE OF ALUMIUM 2024 AND STEEL 304 A thesis submitted in partial fulfillment of the requirements for the degree of M.S. of Mechanical and Nuclear Engineering at Virginia Commonwealth University. by

YAN FANG Bachelor’s Degree of Vehicle Engineering, Nanjing University of Science and Technology, China, 2009

Director: BRIAN HINDERLITER, ASSOCIATE PROFESSOR, DEPARTMENT OF MECHANICAL ENGINEERING

Director: RAMANA PIDAPARTI PROFESSOR, DEPARTMENT OF MECHANICAL ENGINEERING

Virginia Commonwealth University

v

Richmond, Virginia August, 2011

Acknowledgement I would like to express my gratitude to all those who helped me during the writing of this thesis. First and foremost, I would like to show my deepest gratitude to my advisor Dr. Hinderliter, a respectable, responsible and resourceful scholar, who has provided me with valuable guidance in every stage of the writing of this thesis. Without his enlightening instruction, impressive kindness and patience, I could not have completed my thesis. His keen and vigorous academic observation enlightens me not only in this thesis but also in my future study. I also would like to show my great thanks to Dr. Pidaparti, who helped me a lot with my thesis and his lectures benefited me a lot. He gave me kind encouragement and useful instructions all through my writing. Here is a special thanks to Dr. Mossi, who gave me a lot of help and shared her experience not only in study but also in my usual life since I came to VCU. Especially she helped me successfully apply for the teaching assistantship in 2011 spring, the funding really helped me a lot to finish my work.

vi

Table of Contents Page Acknowledgements ............................................................................................................. ii List of Tables ................................................................................................................... viii List of Figures .................................................................................................................... ix Chapter 1

Introduction ........................................................................................................1 1.1 Literature Review ....................................................................................1 1.1.1 Corrosion costs and Importance .....................................................1 1.1.2 Analysis based on image and color ................................................2 1.1.3 Models ............................................................................................3 1.2 Thesis objectives ......................................................................................6

2

Experiment ....................................................................................................7 2.1 Important instruction before experiment ................................................7 2.2 Experiment preparation ...........................................................................7 2.3 Experiment procedure .............................................................................9

3

Algorithm and Image analysis .....................................................................11 3.1 Flow chart of the Algorithm ..................................................................11 3.2 Algorithm details ..................................................................................12 3.3 Image analysis .......................................................................................17

vii

4

Details of program and validations ..................................................................21 4.1 Detail and explanation of program ........................................................21 4.2 Program validation ................................................................................32

5 6

Result and Discussion ......................................................................................40 Summary and Recommendations .....................................................................48 6.1 Summary ................................................................................................48 6.2 Recommendations ..................................................................................49

Literature Cited ..................................................................................................................50 Appendices.........................................................................................................................56 A

Scanned pictures ..............................................................................................57

B

Matlab code ......................................................................................................67

viii

List of Tables Page Table 1: Coordinate output. ...............................................................................................31 Table 2: Result of test image. ............................................................................................38 Table 3: Corrsoion growth rate for Steel and Al panel 1 ...................................................47 Table 4: Corrsoion growth rate for Steel and Al panel 2 ...................................................48

ix

List of Figures Page Figure 1: Distribution of corrosion ......................................................................................1 Figure 2.1: Steel and Al panels in pH solutions...................................................................8 Figure 2.2: Apparatus..........................................................................................................9 Figure 3.1: Whole panel and single pit of corrosion ..........................................................13 Figure 3.2: Grayscale picture of steel panel.......................................................................14 Figure 3.3: Grayscale image enlarged to pixel scale .........................................................15 Figure 3.4: 4-connected pixel ............................................................................................16 Figure 3.5: Original image .................................................................................................17 Figure 3.6: Grayscale image ..............................................................................................18 Figure 3.7: Red layer image ...............................................................................................19 Figure 3.8: Binary version image.......................................................................................20 Figure 4.1: Original picture marked with color .................................................................29 Figure 4.2: Divided original image ....................................................................................30 Figure 4.3: Dimension of divided piece .............................................................................31 Figure 4.4: A small piece of corrosion image ....................................................................32 Figure 4.5: Output of corrosion area ..................................................................................33 Figure 4.6: Output of matrix ..............................................................................................35 Figure 4.7: test steel image 1 paint with mark pen ............................................................37

x

Figure 4.8: Test steel image 2 paint with mark pen ...........................................................38 Figure 5.1: Matrixes from import ......................................................................................40 Figure 5.2: 2D histogram plots ..........................................................................................42 Figure 5.3: 3D histogram plots ..........................................................................................42 Figure 5.4: Corrosion growth rate of steel (pH 7) .............................................................43 Figure 5.5: Corrosion growth rate of steel (pH 4) .............................................................44 Figure 5.6: Corrosion growth rate of whole steel panel (pH 7) .........................................45 Figure 5.7: Pit initiation of steel (pH 7) .............................................................................45 Figure 5.8: Pit initiation of steel (pH 4) .............................................................................46 Figure 5.9: Pit initiation of Al (pH 4) ................................................................................46 Figure 5.10: Pit initiation of Al (pH 11) ............................................................................46

Abstract SIMULATION, MEASUREMENT AND IMAGE ANALYSIS OF CORROSION INITIATION AND GROWTH RATE OF ALUMIUM 2024 AND STEEL 304 By Yan Fang, MS A thesis submitted in partial fulfillment of the requirements for the degree of M.S. of Mechanical and Nuclear Engineering at Virginia Commonwealth University. Virginia Commonwealth University, 2011

Major Director: Dr. Brian Hinderliter, Ph. D Associate professor, Mechanical and Nuclear Engineering Time: 11 a.m. Place: Room E3210 Engineering East Hall Date: Friday August 12, 2011 Corrosion initiation and growth rate are important properties in maintaining structural integrity, especially for surface and pit corrosion of common infrastructure and transportation metals. In this study, the surface corrosion pit initiation and growth rate on Aluminum 2024(common aerospace alloy) and Steel 304(common alloy for infrastructure) in different pH solutions was measured and values were analyzed by image analysis over a scheduled time. A MATLAB algorithm was developed for detecting the initiation and growth rate of pits as a function of time. The developed algorithm was validated with simulated specimen as well as experiments conducted corrosion specimen. Based on the result of obtained, the MATLAB algorithm predicts the right trends and power law radial xi

xii corrosion pit growth rates and should be useful for corrosion initiation and growth predictions in various metals.

Chapter 1 Introduction 1.1 Literature review 1.1.1 Corrosion costs and Importance The corrosion of metallic structures has brought a significant impact on economy of United States, including infrastructure, transportation, utilities and all kinds of manufacturing. The distribution of corrosion cost in United States, which released by the US Federal Highway Administration is show in Figure 1. In their study, the total direct cost of corrosion was determined by analyzing 26 industrial sectors, in which corrosion is

Figure 1 Distribution of corrosion 1

2 known to exist, and extrapolating the results for a nationwide estimate. The total direct cost of corrosion was determined to be $279 billion per year, which is 3.2 percent of the U.S. gross domestic product (GDP). Indirect costs to the user (society costs) are conservatively estimated to be equal to the direct costs. This means that the overall cost to society could be as much as six percent of the GDP. Often, the indirect costs are ignored because only the direct costs are paid by the owner/operator. Most of the corrosion cost (defense, drinking water & sewer system, motor vehicles, pipelines) come from metallic corrosion [1]. The U.S. infrastructure and transportation system allows for a high level of mobility and freight activity for the nearly 270 million residents and 7 million business establishments. Direct cost of corrosion in the infrastructure category annually is about 22.6 billion dollars, which is 16.4 percent of the total cost of the sector category. Steel and Aluminum are the two most common used metal in the infrastructure category, that is reason why these two material have been used to do the experiment in this thesis [2]. Such high cost of corrosion has become the motivation to find a method to predict the trend of how corrosion grows and the corrosion growth rate. 1.1.2 Analysis based on image and color In the process of reviewing the literatures, many articles are found helpful. There is an article indicates a method to use morphology image to evaluate the corrosivity of a material, they generate a model for the sample material LY12CZ. By the application of the model set in the software of AFGROW, they can predict the pitting depth [3]. Also by

3 digital image processing, another article which focused on the surface damage analysis, shows initial image is characterized by three categories: color, texture and shape features. Machine vision methods are applied to the analysis of corrosion surface damage for the first time. Surface damages instead of electrochemical methods [4]. Both of them inspire me that method of image processing can be an effective method for analyzing the change of corrosion. The research based on the color analysis also becomes a significant part to the study on corrosion. One article indicates the scanner-based image analysis (based on RGB) produced rapid measurements of concrete corrosion that were consistent with the conventional gravimetric method which is more time-consuming. The scanned image analysis technique is a straightforward, inexpensive method for monitoring the concrete corrosion [5]. Despite the method of scanner-based image analysis is generated for the concrete corrosion, material is not the core part of this method, that makes me apply the method on my own experiment for the steel and aluminum. 1.1.3 Models A cellular automated model (by java program) was developed before this study to simulate and show how the corrosion grows [18]. It reveals the corrosion will change the color and reflectance of the metal. (The color in this model represents pit depth which different from the model in this thesis, so the detail of this CA will not be specified)A cellular automaton is a discrete model studied in computability theory, mathematics, physics, complexity science, theoretical biology and microstructure modeling. It consists

4 of a regular grid of cells, each in one of a finite number of states, such as "On" and "Off" (in contrast to a coupled map lattice) [6-8]. However, the model needs parameters (initiation rate) to fix in. So possibility of finding the parameter by both experiment data and program through the image analysis method talked above is considered. This model would also help this study to find the trend of corrosion growth rate to do the prediction, which can really help reduce the cost of corrosion in the industry. After discussing the feasibility with my advisors, it is decided to do the corrosion analysis based on the high resolution scanned image, which means the visualization of the corrosion. Many people have used similar method to do all kinds of analysis for corrosion research, they use the speckle digital photography to measure the corrosion rate of iron in different concentration of H 2 SO 4 corrosive solutions, and compare the result with conventional electrochemical measurements [9]. Some use digital image processing reflection photo elasticity to quantitatively estimate the mechanisms of crack extensions of ferroconcrete due to corrosion [10]. Some others developed a digital image processing and analysis method to classify shape and evaluate size and morphology parameters of corrosion pits. Theoretical geometry data (similar to the part of we examine the program result) have been compared against experimental data obtained for titanium and aluminum alloys subjected to different corrosion tests [11]. There is an alternative corrosion visualization scanning electron microscope (SEM) can be found in many references. SEM is a type of electron microscope that images a sample by scanning it with a high-energy beam of electrons in a raster scan pattern. The electrons

5 interact with the atoms that make up the sample producing signals that contain information about the sample's surface topography, composition, and other properties such as electrical conductivity. This technology can be used in study of high temperature corrosion of pure iron and nickel. Through that, grain boundary grooving, grain growth, surface rearrangement and the corrosion happened in different phases can be observed [12]. Some applications are even used in the medical field. They combined the technology of SEM with X-ray microprobe to observe the change of corrosion of endodontic silver cones [13]. But the SEM needs relatively high condition, it requires that the surfaces be ground and polished to an ultra smooth surface [14]. These kinds of conditions are difficult to get in current condition of experiment and in my image analysis model, it does not need to go into that level of vision, so I did not use SEM in my experiment. The algorithm of program (model) part by Matlab for calculate the irregular corrosion area is a little similar to the flood-filled algorithm, but in a different way. Flood fill, also called seed fill, is an algorithm that determines the area connected to a given node in a multi-dimensional array. It is used in the "bucket" fill tool of paint programs to determine which parts of a bitmap to fill with color, and in games such as Go and Minesweeper for determining which pieces are cleared. When applied on an image to fill a particular bounded area with color, it is also known as boundary fill [15]. Flood-fill can be widely used in all kinds of image analysis system not only in 2D but also in 3D [16, 17].

6 Based on the literature, there is no algorithm which can automatically defect and calculate the corrosion pit growth, thus the approach of the algorithm needs to be developed originally. Here is an overview for my strategy to calculate the irregular shape of corrosion pits: the corros...