2021 Fiscal Year Research-status Report
Theoretical and numerical analysis for a phase-field model describing the crack growth phenomenon
| Project/Area Number |
19K14605
|
|---|
| Research Institution | Hokkaido University |
|---|
Principal Investigator |
GAO Yueyuan 北海道大学, 電子科学研究所, 特任助教 (80807793)
|
|---|
| Project Period (FY) |
2019-04-01 – 2023-03-31
|
| Keywords | Crack growth phenomenon / Phase-field Model / Fracture toughness / Inverse estimation / Data science |
|---|
| Outline of Annual Research Achievements |
We study the crack growth phenomenon by means of phase-field models. After having performed simulations of the crack growth phenomenon and computed the effective toughness of the material which is called the J-integral, we investigated the inverse estimation of the fracture toughness of the phase-field model based on the crack path. We have applied several developed data science technique, and then well adapted the techniques to the specific crack propagation problem. We proposed an algorithm based on the data processing and linear regression and then applied the study to compare 2 types of phase-field model based on different surface energy, which are called AT1 and AT2 models. We have investigated the advantages and disadvantages of these 2 models in the forward and inverse problems.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
For the year 2020-2021, we mainly performed numerical simulations and check the J-integral, which is the forward problem to ensure our analysis and programming are on the right direction. The estimation of parameter has become a topic of inverse problem to explore. For the year 2021-2022, we started the inverse problem by applying different data science techniques. After several discussion and analysis, we have well adapted the techniques to the specific crack propagation problem and we proposed an algorithm that is original which suggests to sample points in the region where the crack has not arrived. We considered and compared different inhomogeneous materials in 2 different models which reveals the inherent difference between the 2 models. At this moment, we are summarizing the results.
|
| Strategy for Future Research Activity |
The novel point of this study is the proposed algorithm and we apply the algorithm to study the difference among different types of models, so we plan to continue the research work for more complicated models. The models we are studying now describe the elastic materials, we aim to extend our study to cover larger range of materials in the real world, for example the viscoelastic materials. At the same time, we observe different patterns of the crack path in inhomogeneous case, such as by-passing, pinning and penetrating. We plan to perform the physical and theoretical study of the formation of different patterns.
|
| Causes of Carryover |
Now we are focusing on summarizing the results and preparing for the publication. The main aim for the year is to sum up the results and write and submit the manuscript. One part of the remaining funding will be used for purchasing software that we apply to verify and improve the simulation results and the other part will be used for the publication fee of the paper.
|
