欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对钢水动态冶炼数据的特点,利用非线性多元回归分析的麦夸特算法得出钢水温度和含碳量与其他过程变量之间的数学模型,实现各个过程变量的合理搭配。通过中值滤波器对数据进行滤波,对滤波后的部分数据进行非线性回归分析,预测关于钢水终点温度和含碳量的数学模型,利用麦夸特算法对模型进行求解,用建模余下数据对所建的数学模型进行校验。结果表明：所获得的终点温度和含碳量模型与实际模型相比均方差分别为0.11和0.09,所建模型具有较好的拟合度。
英文摘要：
In this paper, aiming at the dynamic characteristics of the steel smelting data, Marquardt algorithm of nonlinear regression analysis is used to obtain a mathematical models between the carbon content, molten steel temperature and other process variables, which will give a reasonable mix of various process variables. Firstly, filtering the data by a median filter, Secondly, making non-linear regression analysis of some data on the filtered to predict the mathematical models of molten steel temperature and carbon content, while using the Marquardt algorithm to solve the model; Finally, checking the mathematical model with the remaining data modeling. Simulation results showed that the mean square errors were 0.11 and 0.09 respectively by comparing between the obtained models and the practical models.
同期刊论文项目
同项目期刊论文
Palmprint Recognition Based on Multilinear Principal Component Analysis and Fisher Linear Discrimina
期刊信息
位置：