This study presents a comprehensive optimization methodology for low-pressure casting process parameters of A356 aluminum alloy sleeves, integrating numerical simulation, response surface methodology (RSM), and genetic algorithms to enhance casting quality and mechanical properties. The research focuses on minimizing shrinkage porosity volume and secondary dendrite arm spacing (SDAS) through systematic parameter tuning.
Experimental Design and Numerical Simulation
The investigation employs Box-Behnken response surface design with four critical parameters: upper die temperature (A), side die temperature (B), lower die temperature (C), and pouring temperature (D). ProCAST software simulates the process for 27 experimental combinations, quantifying shrinkage porosity volume (Y₁) and SDAS (Y₂) as quality indicators. The computational model accurately replicates thermal gradients and solidification behavior inherent to squeeze casting operations.
Response Surface Analysis
Quadratic regression models establish relationships between process parameters and quality metrics. For shrinkage porosity volume (Y₁), the significant model terms are:
$$Y_1=0.32 +0.02801A +0.03389B-0.01158C -0.01988D-0.01193AB+0.009225AC-0.00225AD-0.003375BC-0.006375BD+0.0027CD+0.03648A^2-0.003288B^2+0.0209C^2+0.02976D^2$$
ANOVA reveals parameter significance for shrinkage porosity:
| Parameter | F-value | p-value | Significance |
|---|---|---|---|
| B (side die) | 106.479 | <0.0001 | ** |
| A (upper die) | 72.763 | <0.0001 | ** |
| D (pouring) | 36.618 | <0.0001 | ** |
| C (lower die) | 12.438 | 0.0042 | ** |
For SDAS (Y₂), the predictive model is:
$$Y_2=36.4-0.004167A+0.3792B+1.191C+0.17088D-0.0025AC-0.005AD-0.0325BC+0.015BD +0.1275CD +0.09833A^2 +0.2708B^2 +0.3133C^2+0.1433D^2$$
Parameter significance for SDAS:
| Parameter | F-value | p-value | Significance |
|---|---|---|---|
| C (lower die) | 439.858 | <0.0001 | ** |
| B (side die) | 44.593 | <0.0001 | ** |
| D (pouring) | 9.052 | 0.0109 | * |
Multi-Objective Genetic Algorithm Optimization
A genetic algorithm addresses the multi-objective optimization problem:
$$\text{min } f_1=Y_1(z)$$
$$\text{min } f_2=Y_2(z)$$
$$\text{s.t. } 290℃≤A≤350℃, 330℃≤B≤390℃, 370℃≤C≤430℃, 690℃≤D≤710℃$$
The weighted sum method converts this to a single objective:
$$Y(z)=\omega_1 f_1+\omega_2 f_2$$
where weights ω₁=7 and ω₂=3 emphasize shrinkage reduction while maintaining mechanical enhancement. After 200 generations with population size 20, crossover probability 0.8, and mutation probability 0.1, the Pareto-optimal solution emerges.
Validation Results
The optimal squeeze casting parameters significantly improve quality metrics:
| Parameter | Optimal Value | Initial Value |
|---|---|---|
| Upper die temperature | 307°C | 320°C |
| Side die temperature | 330°C | 360°C |
| Lower die temperature | 394°C | 400°C |
| Pouring temperature | 701°C | 700°C |
| Quality Metric | Initial | Optimized | Improvement |
|---|---|---|---|
| Shrinkage porosity (cm³) | 0.3199 | 0.2777 | 13.2% |
| SDAS (μm) | 36.41 | 36.36 | 0.14% |
Conclusions
The integrated optimization framework demonstrates significant improvements in A356 aluminum alloy sleeve quality. Key findings include:
- Side die temperature exerts the strongest influence on shrinkage defects (F=106.479), while lower die temperature dominates SDAS development (F=439.858)
- The Pareto-optimal solution (upper die 307°C, side die 330°C, lower die 394°C, pouring 701°C) reduces shrinkage porosity by 13.2%
- SDAS distribution becomes more uniform throughout the sleeve, enhancing mechanical properties despite marginal maximum value reduction
- Genetic algorithms effectively resolve multi-parameter coupling effects in squeeze casting optimization
This methodology provides a robust framework for parameter optimization in industrial squeeze casting applications, particularly for complex geometries like engine sleeves. Future research will focus on cooling system optimization to further mitigate residual porosity in thick sections.