Background

Aleph Tech worked with a leading aluminium producer to optimise energy usage for its aluminium processing plant and its coal-based power plant.

Challenges

The plant struggled with undersupplying energy to smelter, FRP, and CRM operations, leading to costly grid energy imports. This was due to:

1. Inaccurate predictions of energy demand for aluminium processing.
2. Minimal visibility into power plant efficiency (boiler, turbines, etc.).
3. Variability in generation capacity from seasonal factors (rainy vs. dry) and coal quality (moisture, calorific value, carbon content).

Solution

Aleph Tech created digital twins of the power plant and aluminium processes, incorporating seasonal data. Advanced machine learning optimised turbine load allocation, coal flow rate, and other setpoints.

Result

Achieved 95-99% energy demand prediction accuracy and cut energy costs by 21%.

Background‍

Alumina calcination, a high-energy process central to aluminium refining, involves a furnace where hot air removes the surface and bound moisture to produce alumina of specified quality. Aleph Tech worked with an aluminium plant to reduce fuel consumption while maintaining product specifications.

Challenges

The calcination process is highly interconnected and complex, involving:

1. Variability in aluminium trihydrate feed conditions.
2. Impact of ambient air quality on furnace performance.
3. Balancing fuel consumption with desired product moisture content.

Solution

Aleph Tech developed a digital twin of the calcination plant (furnaces, compressors, exchangers, cyclones, etc.) to predict performance under varying conditions. Subsequently, our advanced machine-learning algorithm predicts the impact of feed, ambient conditions, and fuel quality, and optimises operational parameters for minimal fuel use while achieving the desired product quality.

Results

Aleph's solution accurately predicts plant performance and optimises the feed material rate, fuel flow rate, combustion, and secondary air flow rates. This approach reduced energy usage and emissions by 10-15%.