CCT database and neural networks

Simulate phase transformation using the CCT database and neural networks

The simulation of the phase transformation is a key element of material simulation. With the help of the various CCT diagrams, microstructure fractions and hardness values can be determined. Using neural networks, phase transformation for individual steel alloys and austenitizing conditions can be simulated even more precisely.

The CCT database in MatILDa®

In the CCT database of MatILDa®, experimentally recorded and standardized CCT and TTT diagrams are available in a digital library for a large number of steel grades:
• continuous CCT diagrams,
• forming CCT diagrams,
• welding CCT diagrams,
• isothermal CCT diagrams and
TTT diagrams.
CCT database
With the help of the various CCT diagrams, microstructure fractions and hardness values can be determined. Information on the austenitizing temperatures and the austenitizing time as well as the exact chemical analysis can be displayed in the CCT database. In addition, the important temperatures such as Ac3, Ac1, bainite start temperature BS und martensite start temperature MS can be displayed in the CCT data base. By adding individual cooling curves, the phase transformation can be understood according to the conditions in the industrial process. In the continuous CCT diagrams, in addition to the starting temperatures for martensite, bainite, ferrite and pearlite formation the phase fractions and the hardness values can be found.

The database’s open interface permits users to import additional digitalized CCT / TTT diagrams and to implement this information in further calculations involving FEM simulation software.

Simulating phase transformation using neural networks

A disadvantage of individual CCT and TTT diagrams is that the microstructure fractions and hardness values are only valid for the specific parameters (concrete chemical analysis, austenitizing temperature and time) from the experiment. Thus, MatILDa®’s competence is strengthened by the integration of neural networks developed by GMT – referred to in the CCT database as calculated TTT diagram. A major advantage is the validity of a neural network for a predefined range of austenitizing temperature as well as chemical analyses, which can be selected by means of sliders. The use of neural networks allows users to make predictions regarding microstructure fractions, ferrite grain size, pearlite lamellar spacing and mechanical product properties.
TTT database and neural networks
MatILDa® exhibits another special feature: For detailed examination, phase formations are displayed in the form of logarithmic t85 diagrams. User-friendly controls allow quick adjustments of values and application of changes in microstructure portions.
Ask for more details about our CCT database and neural networks!
✔ We have 20+ years of expertise
✔ We successfully advise and support customers from a wide range of industries
✔ Save time & costs through reliable data
✔ Your inquiry is free-of-charge & non-binding

Interfaces & Functionalities

Convenient data exchange via standard interfaces, e.g. with QForm UK, simufact and AutoForm


The material data base for realistic material simulation