论文部分内容阅读
终点碳含量是决定钢的种类和质量的关键因素,是转炉炼钢过程中最难控制的变量之一。建立了基于遗传算法的核偏最小二乘回归(GA-KPLSR)方法的终点碳含量的预测模型。数据仿真结果表明,基于GA-KPLSR的预测模型,不仅能高效的处理变量之间的非线性关系,而且能快速收敛至最优解,得出预测结果的均方误差比主元回归(PCR)和偏最小二乘回归(PLSR)分别降低了25.77%、23.27%;相对误差降低了29.55%、26.83%;绝对误差降低了27.22%、24.84%。该方法可为实际生产中的终点控制提供参考,提高生产效益。
The end point of carbon content is the key factor that determines the type and quality of steel and is one of the most difficult variables to control during BOF steelmaking. A prediction model of end point carbon content based on GA-KPLSR was established. The simulation results show that the GA-KPLSR-based prediction model not only can effectively deal with the nonlinear relationship between variables, but also converges to the optimal solution quickly, and the mean square error of prediction results is higher than that of principal component analysis (PCA) And partial least squares regression (PLSR) decreased by 25.77% and 23.27% respectively; the relative errors decreased by 29.55% and 26.83%; the absolute errors decreased by 27.22% and 24.84% respectively. The method can provide a reference for the end point control in actual production and improve the production efficiency.