Volume 1 Issue 1

Research Article
September 08, 2020

Prediction of Hot Metal Temperature Using Multivariate Data Science Approach for Blast Furnace Iron Making Process

Current Trends in Engineering Science (CTES)

This paper presents the determination of health status of blast furnace based on the principle component analysis and multivariate analysis. The health status of Blast Furnace (BF) represented in terms of ‘hot metal temperature’ is an important parameter to regulate the smooth operation coupled with continued production of hot metal to avoid the major danger events to happen. The health index also indicates the performance of BF at early stage so that the operator can take appropriate actions to avoid deterioration in the blast furnace in prior. The health status of blast furnace indicates the stability or instability condition of BF, which might occur during the production process and is used to recognize the fault. The principle component analysis techniques has been widely used in various industrial fields due to its various advantages such as, it does not require the knowledge of the process and faults. In this paper, based on past dataset collected from blast furnace, principle component technique is applied using weka; a software application; that employs pre-processing, clustering, classification and selective attribute modules for development of the health status of BF. The health status has been tested with varying process data and is found to be useful in identification of process abnormality in BF.