JapanChemicals and materials2021 – ongoingNissan Chemical
Hybrid AI model of an ammonia-plant steam reformer
Nissan Chemical built a more accurate steam-reformer model 2 times as fast with hybrid AI
Problem
Nissan Chemical needed a simulation that reproduced its ammonia plant's behavior. Its conventional reactor model struggled with furnace temperature distributions.
Approach
The company used Aspen Hybrid Models within Aspen Plus to combine physical process knowledge with neural-network calibration. Engineers cleaned plant data before training the model. The resulting model reproduced operating data more accurately than the conventional approach.
Results
Takeaway
The demonstrated benefit was faster model development. The cited operating-cost reduction was potential savings from future optimization.
Categories
- Industry
- Chemicals and materials
- Business function
- IT and R&DProduction and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Vendor solution
- Evidence
- Vendor case study
Sources
Source summaries
AspenTech reported that Nissan Chemical created a more accurate reformer model 2 times as fast. The customer presented results in May 2021; operating-cost savings were described as potential.
Full details are in the original sources.
