Abstract
The application of data mining techniques in the design of modern foundry materials allows achieving higher product quality indicators. Designing of a new product always requires thorough knowledge of the effect of alloying elements on the microstructure and hence also on the properties of the examined material. The conducted experimental studies allow for a qualitative assessment of the indicated relationships, but it is the use of intelligent computational techniques that enables building an approximation model of the microstructure and, owing to this, make predictions with high precision. The developed model of prediction supports the technology-related decisions as early as at the stage of casting design and is considered the first step in selecting the type of material used.
Highlights
Materials Engineering is the field of science characterized by an interdisciplinary nature
It is essential to define the rules, laws, and relationships existing in the specified area, and the data mining tools can support the experimental research process
In the paper [28], the effect of the shape of the reaction chamber on the graphite shape in the cast iron obtained by Inmold technology was investigated
Summary
Materials Engineering is the field of science characterized by an interdisciplinary nature. With models predicting the microstructure, it becomes possible to improve and automate the process of designing new materials with desired properties. The cast iron with compacted graphite has a lower coefficient of thermal expansion, higher thermal conductivity, greater resistance to dynamic changes in temperature, higher vibration damping capacity, and better castability. All these advantages predestine this material for a variety of applications. The first utilitarian use of this cast iron covered brake discs for high-speed railway stock This material is mainly used for the construction of blocks of internal combustion engines, exhaust manifolds, and the like cast automotive parts. There are various embodiments of the neural networks to approximate the cement structure [22] and [23], and well-known is
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have