Precise control of the end-point phosphorus and sulfur content in converter steelmaking is critical to ensuring steel quality. An end-point prediction model based on LWOA-TSVR is established to better control the BOF end-point content of phosphorus and sulfur. The prediction impact is compared to the models BP, SVM, and TSVR. The results indicate that the LWOA-TSVR model outperforms the other three models in terms of accuracy. And the prediction model is applied to a steel mill. The results showed that the hit rates of phosphorus content and sulfur content were: 96.3%, 81.7%, and 94.8%, 76.9% in the range of ±0.005% and ±0.003%, respectively. The double hit rate was: 87.63% in the range of ±0.005%. Thus, it is demonstrated that the LWOA-TSVR prediction model performs effective prediction of end-point phosphorus and sulfur content with prediction accuracy that exceeds that required by the real steelmaking process in a steel mill.

Bending control is one of the main methods of shape control for the hot rolled plate. However, the existing bending force setting models based on traditional mathematical methods are complex and have low control accuracy, which leads to poor strip exit shapes. Aiming at the problem of complex bending force setting of the traditional algorithm, an improved whale swarm optimization algorithm and twin support vector machine-based bending force model for hot rolled strip steel (LWOA-TSVR) is proposed. Based on the hot rolling field production data of a steel plant, the research group established the bending force prediction model by using the nonlinear approximation ability of the twin support vector machine. The introduction of the Levy flight improvement algorithm improves the generalization ability, prediction accuracy, and convergence speed of the whale swarm optimization algorithm with the help of the convergence of coefficient vectors, solves the problem of a random selection of the parameters of the traditional whale swarm optimization algorithm and optimizes the ability of the whale swarm algorithm to jump out of the local optimum. Based on the actual rolling database, the hit rate of the proposed method reaches 91% (from −5 to 5 KN), which fully meets the requirements of the detection accuracy on the actual production line. The model is not only able to overcome the local search to obtain the global optimal solution, but also has the advantages of fast convergence and higher prediction accuracy. A comparison of the model with twin support vector machines and traditional whale swarm algorithms shows that the prediction accuracy is higher. The experimental results also show that this model has advantages over existing bending force prediction models in terms of improving the accuracy of the strip shape control and providing theoretical guidance for practical bending force settings.

No abstract

No abstract

During the LF refining process, the end-point temperature and carbon content changes at the end of refining are relatively lagging. And most of the traditional prediction models suffer from weak operational generalization ability, long computation time, and the existence of multiple polarization points, which greatly affect the prediction accuracy of the models. In this paper, a wavelet transform based weighted algorithm (WTW) optimized twin support vector machine algorithm (WTWTSVR) prediction model for refining end-point temperature and carbon content is proposed. WTW is introduced into the objective function on the basis of TSVR, and the objective function is converted into an unconstrained optimization solution problem, and then a mathematical model of LF refiner end-point temperature and carbon content is established to complete the prediction of these parameters. The production practice shows that the forecast accuracy of the model for 400 furnace times is 91.5%, 90.2%; 95.6%, 95.5% for refining end-point temperature error and carbon content error within ±5% and ±10%, respectively. The double hit rate within different error ranges (within 10 °C for the temperature model and within 0.005% for the carbon content model) reached 86.5%. The results indicate that the method can provide theoretical guidance for the LF refining production process.