An accelerometer-based dataset for monitoring slag in steel manufacturing

Abstract

Abstract Objectives Slag detection in steel manufacturing is essential for ensuring high product quality and process efficiency. The purpose of the accelerometer-based data is to allow for accurate monitoring and differentiation between slag and molten metal flow. This is vital to prevent equipment damage, maintain steel quality, and enhance operational effectiveness. The data is collected specifically to support the development of machine learning models for real-time monitoring in the steel production process, addressing the critical need for precise slag detection. Data description The Steel Slag Flow Dataset (SSFD) offers a comprehensive set of data obtained from a triaxial accelerometer during various stages of steel production. By leveraging this dataset, researchers can effectively analyze and classify the flow of slag versus molten metal. The dataset allows for data-driven approaches so that machine learning researchers can optimize steel manufacturing processes, ensuring high-quality steel production and minimizing the risks associated with slag contamination. The SSFD provides a valuable resource for researchers seeking to enhance predictive maintenance and monitoring in industrial applications.