Development of Data-Driven Machine Learning Models for the Prediction of Casting Surface Defects

This paper presents an approach for the application of machine learning in the prediction and understanding of casting surface related defects. The manner by which production data from a steel and cast iron foundry can be used to create models for predicting casting surface related defect is demonstrated. The data used for the model creation were collected from a medium-sized steel and cast iron foundry in which components ranging from 1 to 100 kg in weight are produced from wear and heat resistant cast iron and steel materials. This includes all production-relevant data from the melting and casting process, as well as from the mold production, the sand preparation and component quality related data from the quality management department. The data are tethered together with each other, the information regarding the identity and number of components that resulted in scrap due to the casting surface defect metal penetrations was added to the dataset. Six different machine learning algorithms were trained and an interpretation of the models outputs was accomplished with the application of the SHAP framework.

Control of the Mg-Treated Iron Casting Skin Formation by S-Diffusion Blocking at the Metal–Mould Interface

Having established that sulphur presence in the mould materials appears to have an important contribution in graphite degeneration at least in the casting surface layer, a research program is undertaken to explore the possible beneficial effect of sulphur diffusion blocking at the metal–mould interface. Test samples, with and without a thin steel sheet (up to 3 mm thickness) application on the inner surface of the mould cavity, before iron melt pouring, are considered for structure analysis. A higher nodulizing potential (0.048% Mgres, 0.015% Ceres, and 0.006% Lares) decreases the occurrence of surface graphite degeneration in castings obtained in rigid chemically bonded resin sand moulds, using P-toluol sulfonic acid (PTSA) hardener (S-including), but it is not enough to avoid this phenomenon (200–400 μm skin in present experimental conditions). The casting skin appears to have different values, depending on the evaluation technique (un- and Nital-etching direct measurement, or graphite parameters variation on the casting section). In the presence of a thin steel sheet at the metal–mould interface, the casting skin thickness decreases or is just excluded. It is supposed that it acts as a barrier, blocking S-diffusion from the mould media into the iron melt. Without this S-diffusion, the graphite degeneration in the casting surface layer could be avoided, or at least diminished. For industrial application, the increasing of residual content of nodulizing elements is a limited solution, and it is recommended to use barriers to block S transfer on the mould/metal surface, such as dense coatings or coatings with desulphurization capacity.

Quality evaluation of iron casting EN-GJL-200 produced on the disamatic line

This article discusses the issues concerning the quality of casting surface with a small unit weight manufactured on the Disamatic automatic casting lines in a flaskless mould with a vertical parting line. The research was conducted in manufacturing conditions in one of the casting facilities in the country. The identification of casting defects for a monthly manufacturing cycle was made and agreed that the main causes of scrap casting are defects on the raw surface: lumps of moulding sand which after removal form cavities and draws as well as defects of gas origin such as blowholes. There are also defects of form namely short runs. In the later part of the work, the research results of the properties of moulding sand were presented and casting defects were selected the occurrence of which is connected with the quality of moulding sands.