الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 247 |  |  |
| | | from | 247 | | |
Abstract
The present investigation studies the microstructural characteristics of A356 aluminium alloy feedstocks. These feedstocks are produced by cooling slope casting (CSC) technique under different process parameters, typically, tilt angle (θ), cooling slope length (L), and pouring temperature (T). Moreover, the effect of the water cooling on the microstructural characteristics was evaluated. The investigated microstructural characteristics are grain size, shape factor, and aspect ratio of the primary α-Al grains. Also the porosity content was measured at top, middle, and bottom positions of ingots poured with and without water cooling. Additionally, the analysis of variance (ANOVA) was used to emphasize the main parameters and their interactions that affect the aforementioned microstructural characteristics. Moreover regression models were developed to determine the relationship type between CSC parameters and different results. Finally, two fuzzy-logic models were developed to simulate the CSC process with three inputs (θ, L, T) called antecedents and three outputs (grain size, shape factor, aspect ratio) called consequents with and without water cooling. These models can be used to obtain the optimum antecedents that give the best results of consequents.
It was found that the morphology of the primary α-Al grains was non dendritic due to break down of the dendritic microstructure and converts it to globular one by the shear force resulting from natural gravity when the molten metal is passing the slope plate. The feedstocks poured under water-cooling exhibited smaller grain size, but larger shape factor and aspect ratio than those poured without water-cooling. The porosity content was found to be larger in the feedstocks poured under water-cooling than in those poured without water-cooling. The microstructural results including grain size, shape factor, and aspect ratio decrease with increasing tilt angle from 30 to 45o. Additional increase in tilt angle, from 45 to 60o, tends to increase both grain size and shape factor, accompanied by a decreased in aspect ratio. With respect to the effect of cooling slope length, grain
iii
size decreases while both shape factor and aspect ratio increase with increasing cooling slope length. Grain size decreases, but both shape factor and aspect ratio increase with raising pouring temperature from 620 to 630 oC. Additional raise in pouring temperature (from 630 to 640 oC) leads to an increase in grain size, but both shape factor and aspect ratio decrease.
ANOVA results explained that cooling slope length has the highest, while tilt angle has the lowest statistical and physical significance on grain size. For shape factor the water cooling has the highest, but pouring temperature has the lowest statistical and physical significance. With respect to aspect ratio the cooling slope length has the highest, while water cooling has the lowest statistical and physical significance.
The regression models explained that the relationships between CSC parameters (θ, L, and T) and microstructural results (grain size, shape factor, and aspect ratio) are highly nonlinear relationships with a degree of uncertainty of about 40%, 28%, and 45% for grain size, shape factor, and aspect ratio respectively.
The fuzzy logic models for feedstocks poured with and without water cooling explained a significant prediction of grain size, shape factor, and aspect ratio compared to the experimental results. An error percentage is checked for the developed models and the predicted error for feedstocks poured without water cooling did not exceed 5.48% for grain size, 1.97% for shape factor, and 2.97% for aspect ratio. While for the feedstocks poured with water cooling the maximum error percentage was 6.15%, 2. 4% , and 3.89% for grain size, shape factor, and aspect ratio respectively.