5th International Conference on Advances in Solidification Processes (ICASP-5) 5th International Symposium on Cutting Edge of Computer Simulation of Solidification, Casting and Refining (CSSCR-5)

Numerical simulation technology has been widely used in the field of casting because of its visibility and forward-looking. However, for numerical simulation prediction of complex shape castings, the designer usually considers pouring under ideal conditions. The thermal property parameters used are uniform standardized parameters, and less consideration is given to the physical property parameters and boundary condition differences caused by process fluctuations in the casting process. Therefore, the numerical simulation prediction results are quite different from the actual production casting defects, which restricts the application and development of numerical simulation technology in casting process design and optimization. In this paper, a single-piece simulation and quality control method for complex shape casting is proposed considering the condition of process fluctuations. Firstly, the sensor, industrial data acquisition and casting single-piece full-cycle management technology is used to accurately obtain the key process parameters such as alloy casting temperature and time, initial temperature of casting and auxiliary materials during the forming process, and the volatility characteristics of the process parameters are derived. Secondly, the thermal property parameter reverse technique is used to carry out the pouring temperature measurement experiment of the standard test piece, and The orthogonal test and the anti-heat conduction method are applied to accurately obtain the thermal property parameters, and series of simulation of the whole process for casting process is performed. Thirdly, combined to the defect level analysis method, Fuzzy evaluation method and deep neural network method, the relationship model between key process parameters and differences is established, and then correct the model through differential reverse feedback to achieve accurate prediction of individual casting defects. Finally, based on the results of accurate prediction, multi-objective optimization algorithm is applied to find volatility characteristics of key parameters and sensitivity and influence mechanism of defects, and combining defect repair method of adaptive operation of casting to digitally control casting products quality in whole process. The research and implementation of the single-piece simulation and quality control method will provide scientific theory and technical support for high-quality and high-performance complex shape casting.