ML-Powered Pot Performance Prediction in Aluminium Smelter
摘要
Hirakud SmelterSmelter (HKD), a unit of Hindalco IndustriesIndustry Limited is a part of Aditya Birla Group (ABG). Hirakud Aluminium is an integrated aluminium smelting complex that uses GAMI Technology and is one of the oldest smelters in India, established in 1959. The potlines, converted from Soderberg to prebake in 2009, have inherent challenges in terms of technology and retrofitting the old pots to prebakes. Facing rising energy costs, aluminium smeltersAluminium smelter aim to reduce specific energy consumptionEnergy consumption by enhancing current efficiencyCurrent efficiency (CE), which lowers energy use and boosts productivity. This article presents a predictive modelPredictive model using Logistic RegressionLogistic regression to identify low-efficiency pots in 235kA aluminium smelting operations. Utilizing data from a 235kA potline, the model predicts pot efficiency with high accuracy and precisionPrecision by analyzing key performance indicators and operational metrics. By providing real-time insights, it pinpoints underperforming pots, enabling targeted interventions that enhance overall process efficiency and productivityProductivity. The findings demonstrate the potential of machine learningMachine learning in optimizing industrial processes and contribute to significant improvements in operational efficiency and resource utilization.