📘 Resource Overview & Documentation Details

You are viewing the technical directory listing for Continuous Casting Modernization for Smarter Steelmaking. This engineering resource is curated under our Intelligent Manufacturing division to support automation specialists, control systems engineers, and manufacturing technicians looking for verified reference materials.

🔍 Contextual Abstract & Technical Scope

Share Summary:LinkedInEmail

Continuous casting modernization is not simply a matter of replacing aging controls. It also involves using process models and operational data to improve how a caster is monitored and controlled. For steelmakers, the challenge is to translate complex phenomena—including heat transfer, solidification, thermal stress, segregation and strand deformation—into useful signals and decisions during production.

A model-based approach can address selected parts of the process rather than trying to represent every interaction in one all-encompassing model. Modular automation packages can then apply those models to specific tasks, from detecting mold conditions to controlling secondary cooling and soft reduction. This offers a way to add intelligent manufacturing functions while retaining existing plant infrastructure.

Why continuous casting modernization depends on process models

Continuous casting combines multiple interacting physical effects. A useful model must account for the relevant process conditions and convert essential parameters into boundary conditions that can be applied to a defined operational problem. The objective is not complexity for its own sake: model parameters must be verified and correlated with process data, and the model should be tuned to the section of the caster or production issue being addressed.

In an intelligent manufacturing system, equipment and software do more than collect measurements. They help operators recognize process conditions, support decisions and, where configured, act on those decisions. This is particularly valuable when production data comes from different systems and decisions must be made while casting is underway.

A modular design can support this work in both online and offline modes. Online applications monitor or control the process; offline tools can support simulation, practice development, scheduling, data analysis and troubleshooting. Sharing a model kernel between simulation and production also helps align predicted behavior with the operating process. The modular structure is relevant to brownfield plants, where modernization may need to fit an existing architecture.

Building a connected casting platform

The packages described for continuous casting cover different process functions, but operate within a broader platform concept. A tracking supervisor records events such as ladle opening and closing, strand starts and stops, and semi-product cutting, weighing and sampling. It also records data, estimates remaining process time, tracks temperature, calculates superheat, coordinates process applications and tracks equipment life to support online quality-level assignment.

This shared foundation gives operators and process systems a more organized view of the casting sequence. The individual applications can then focus on specific needs, rather than requiring every plant to adopt one identical configuration.

Monitoring mold behavior and stabilizing mold level

Detecting mold phenomena

Mold monitoring can use thermal distributions, lubrication information and solidification-related signals to identify conditions that may require attention. These include longitudinal cracks, slag entrapment, irregular mold powder behavior and signs associated with sticking. Where temperature sensors are installed across mold faces, signal analysis can also help identify events such as sticking, breakout and cold-tooth conditions.

For sticking detection, changes in meniscus temperature and mold-level variability can be considered together. In the described application, a decrease in meniscus temperature combined with increased mold-level standard deviation generates an alarm. An operator can then be prompted to reduce casting speed before sticking occurs. The underlying process concern is lubrication: changes in mold powder behavior can reduce the supply of molten flux and contribute to insufficient lubrication at the mold surface.

Controlling mold level

Mold-level control must contend with disturbances such as bulging, standing waves and roll eccentricity. A model-based control system can estimate process variables and apply filtering or control strategies to counter known disturbances. A linear quadratic estimator can provide estimated process data, while a linear quadratic regulator uses estimated states to determine actuator references. The system’s parameters are tuned for the caster, and its response can be adjusted through the controller setup.

Reported plant results in the source compared mold-level standard deviation under a proportional-integral-derivative controller with results using the advanced control system. The reported improvements varied across the illustrated cases, reinforcing that performance depends on the plant and the condition being addressed rather than implying one universal outcome.

Managing cooling and solidification

Secondary cooling control links a temperature model to spray-water flow. The model estimates the strand’s temperature profile, and the control system compares the estimate with target conditions for the product. The temperature difference provides feedback for adjusting water flow. The described approach uses a dynamic three-dimensional solidification model and nozzle layouts intended to distribute water across the slab while limiting uneven cooling. Dynamic control of outer nozzles can also help manage corner temperatures.

Maintaining appropriate surface temperatures through critical caster zones matters for product quality and downstream handling. Cooling models can support practical operations and offline simulations to assess cooling profiles for different steel types. The source describes applications intended to reduce corner cracking and avoid uneven slab profiles associated with overcooling or nonuniform heat transfer.

Applying soft reduction at the right point

Soft reduction applies mechanical reduction to the strand as solidification progresses. Its effectiveness depends on both where it is applied and the amount of reduction. Applied too early, it can limit the reduction of centerline segregation; applied too late, it can risk internal cracking. Process models estimate the evolution of solidification and help position the reduction practice accordingly.

Dynamic soft reduction control uses this information to adjust the relevant segment settings in line with a defined practice. Simulation can be used offline to examine model behavior and test practices. This is particularly pertinent during transient conditions, including sequence start-up, fly tundish and steel-grade changes, when the casting process is not at steady state.

A separate method for verifying the solidification endpoint uses a light oscillation of an upper segment to generate a pressure wave in the liquid pool. Frequency analysis of mold-level signals helps determine whether a liquid core remains. This information can support tuning of the machine setup and the application of soft reduction.

Practical implications for plant teams

Modernization can be staged around the caster’s needs. The reported installations used different combinations of mold monitoring, level control, cooling, solidification verification and soft-reduction functions. In one example, automation modernization integrated with existing machines and required limited changes to the existing framework; the plant’s production stoppage was reported as two weeks. Other installations added specific packages to support cooling simulations or combine control functions with endpoint verification and soft reduction.

For engineering and operations teams evaluating a project, the practical questions are which process problems matter most, what instrumentation and data are available, how models will be tuned and validated, and how online controls will fit with existing automation. Offline simulation and modular deployment can help structure that evaluation.

Conclusion

Continuous casting modernization can combine targeted process models with modular automation to improve visibility and control across the caster. Mold monitoring, mold-level regulation, cooling control and soft reduction each address different process challenges, so their value depends on appropriate data, model tuning and integration with operating practices. For plant decision-makers, the central opportunity is to add focused intelligent functions while building on the equipment and automation already in place.

Access Official Technical PDF