Machine Learning

Steel Electrical Load Prediction

Python ClassificationTime-SeriesEnergy Optimization

Steel Electrical Load Prediction - project visual

The Problem

Steel production requires effective load management; predicting electricity load levels (Light, Medium, Maximum) supports scheduling and energy-saving strategies.

Approach

Built a multi-class classification model using historical consumption data, time-based features, grid measurements, and power factor indicators.

Feature engineering on timestamps, reactive power, and usage patterns improved accuracy.

Results

Enabled prediction of load levels across 35k+ production cycles, supporting optimized production scheduling and better load management.

Key result: Predicted load levels for 35k+ production cycles

Tools Used

PythonScikit-learnPandasNumPy

GitHub - Coming Soon