Mold Data Acquisition and Predicting Defects in High-Pressure Die Casting (HPDC)
- 주제어 (키워드) High-pressure die-casting , Soldering , Shrinkage pore , Deformation , Process Optimization
- 발행기관 서강대학교 일반대학원
- 지도교수 김낙수
- 발행년도 2025
- 학위수여년월 2025. 2
- 학위명 박사
- 학과 및 전공 일반대학원 기계공학과
- 실제 URI http://www.dcollection.net/handler/sogang/000000079663
- UCI I804:11029-000000079663
- 본문언어 영어
- 저작권 서강대학교 논문은 저작권 보호를 받습니다.
초록 (요약문)
This study presents an approach to defect prediction and quality optimization in HPDC through real-time data acquisition and simulations. A system was developed to monitor temperature and pressure within the mold using K-type thermocouples and load cells, embedded into the mold's ejector pins. Data is transmitted wirelessly via Bluetooth, enabling integration into smart factory environments and remote monitoring. Commercial casting software was used to simulate thermal and flow behaviors, with results validated against experimental data and CT scans. This enabled the development of a predictive model to identify defect-prone areas, focusing on soldering defects, ejector pin deformation, and shrinkage porosity. A predictive model for soldering defects estimated optimal solidification times, showing that exceeding an injection temperature of 674.3°C led to incomplete solidification and soldering. The model achieved a high accuracy (R² = 0.9988), allowing precise mold control to mitigate defects. To address deformation from ejector pins, simulations using Abaqus indicated that cooling inconsistencies and force variations led to deformation, with an average prediction accuracy of 80.2%. Real-time monitoring allows proactive adjustments to maintain dimensional accuracy. Shrinkage porosity was analyzed using a dimensionless Niyama criterion based on temperature gradients and cooling rates, validated through CT data inspections. Integrating the predictive model with computational fluid dynamics (CFD) helped optimize cooling rates and minimize shrinkage defects, particularly in complex geometries like motor housings. The study demonstrates that integrating real-time data acquisition, simulations, and predictive models improves HPDC reliability. By leveraging temperature and pressure monitoring with predictive analytics, manufacturers can optimize processes, reduce defects, and enhance product quality. This approach supports defect reduction and aligns with Industry 4.0, enabling smarter, automated manufacturing and ensuring HPDC remains efficient for high-performance component production.
more목차
1. Introduction 1
2. Mold Data Measurement 5
1. Measurement System Design 5
2. Installation of Data Collection Hardware 8
3. Monitoring Results 11
3. Casting Simulation 14
1. Boundary Conditions 15
2. Verification of Results 17
3. Analysis of Differences Between Process Data 20
4. Defects Prediction 21
4.1. Soldering 21
4.1.1 Example of Defect Occurrences 22
4.1.2 Predictive Model Derivation 24
4.1.3 Model Validation 28
4.1.4 Process Conditions Control 29
4.2. Deformation from Eject Pin 34
4.2.1 Prediction by Casting Simulation 35
4.2.2 Geometric Analysis and Verification 39
4.3. Shrinkage pore 43
4.3.1 Prediction model 44
4.3.2 Model validation 48
4.3.3 Process Conditions 51
5. Conclusion 56
References 60
