Ready, Steady, Predict!
Last updated
Was this helpful?
Total Materia Predictor
Last updated
Was this helpful?
Total Materia Predictor simplifies the process of predicting material properties. With advanced machine learning models, it allows users to predict mechanical and physical properties for both metallic and polymer materials, including stress-strain curves and fatigue strength for metallic materials.
Each prediction involves the following steps:
Begin by selecting either a specific material designation or by entering a chemical composition. You can search for a material using standard designations or filter by material group, standard, or producer. This flexibility ensures that whether you know the material name or only its chemical composition, you can proceed to accurate predictions.
Once the material is selected, you need to choose the desired property for prediction:
Property: A list of available properties to predict is provided, such as tensile strength, heat capacity, or thermal expansion.
Model ID: An internal identification number for the model that will be used for prediction.
Applicability: This indicates how well the model applies to the selected material and inputs. It helps users understand the reliability of the prediction. Applicability levels such as "Fully Applicable" or "Extrapolation" inform whether the input values are within the tested domain of the model.
Next, provide specific parameters relevant to your prediction:
Product form: Specify if the material is in bar form, sheet form, etc.
Dimension: Input the relevant dimension or size for the product.
Heat treatment: Select any applied heat treatments. Temperature: Define the operational temperature or conditions under which you want the property prediction.
These details ensure that the prediction is as accurate and applicable as possible for the specific scenario you're evaluating.
Once all parameters have been provided, you will receive the prediction results. The results are based on the machine learning model chosen and provide values for the predicted properties under the given conditions.
Note: If you need to modify any of the inputs, such as the material, heat treatment, or other parameters, you can easily go back to the relevant step by navigating through the process. This ensures flexibility in refining your predictions.
