Reproduce “Predictive control of blast furnace temperature in steelmaking with hybrid depth-infused quantum neural networks” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Nayoung Lee, Minsoo Shin, Asel Sagingalieva, Ayush Joshi Tripathi, Karan Pinto, Alexey Melnikov
Researchers from POSCO Holdings Inc. and Terra Quantum AG developed a hybrid depth-infused quantum neural network to predict and control blast furnace temperatures in steelmaking. The system stabilized molten iron temperature fluctuations to within ","pm7.6",
Key insights
- Gradient Boosting-based feature importance analysis identified previously underutilized sensor data, such as specific slag compositions (TiO2, MnO, S) and ObyC, as highly influential for accurate blast furnace temperature prediction.
- Integrating a Quantum Depth-Infused (QDI) layer within a classical LSTM architecture offers superior temperature prediction accuracy by leveraging quantum-enhanced feature space exploration while maintaining compatibility with NISQ devices.
- A predictive control loop, even with a relatively simple classical linear model for the optimization component, can effectively stabilize a complex industrial process by iteratively forecasting outcomes and adjusting control variables to meet precise operational targets.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.