Numerical Simulation to Predict the Effect of Process Parameters on Hardness during Martempering of AISI4140 Steel
| Autor: | K. Narayan Prabhu, K. M. Pranesh Rao |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
010302 applied physics
Work (thermodynamics) Martempering Materials science Computer simulation Mechanical Engineering Process (computing) 02 engineering and technology Heat transfer coefficient 021001 nanoscience & nanotechnology 01 natural sciences Nonlinear system Mechanics of Materials Distortion 0103 physical sciences General Materials Science Composite material 0210 nano-technology Residence time (statistics) |
| Zdroj: | Journal of Materials Engineering and Performance. 30:3416-3435 |
| ISSN: | 1544-1024 1059-9495 |
| DOI: | 10.1007/s11665-021-05630-6 |
| Popis: | Martempering is a widely practiced industrial heat treatment process to harden steel parts with minimum distortion. A numerical experiment to predict hardness distribution in AISI 4140 steel cylinders of various diameters during martempering is presented in this work. Apart from the diameter (D), the effect of other process variables such as heat transfer coefficient (h), bath temperature (Tb), and residence time (tr) was also studied. The relationship between hardness distribution and the aforementioned process variables was highly nonlinear. An artificial neural network (ANN) model with a single hidden layer and 30 hidden layer neurons was thus developed to predict the hardness distribution in martempered AISI 4140 steel cylinders. The increase in bath temperature, diameter, and residence time decreased the average hardness, and an increase in the heat transfer coefficient increased the average hardness of martempered AISI 4140 cylinders. The weights of the ANN model were used to calculate the relative importance of all input variables and they followed a decreasing order of Tb>D>tr>h. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink