摘要
以转炉炼钢中炉料配比模型为研究背景,采用RBF神经网络进行终点预测模型的构建,并结合目标价格函数实现炉料配比模型的构建。同时,利用修正的最速下降法进行了最优配比参数的求解。此外,为提高了模型计算结果的准确性,利用最小二乘法对金属收得率进行修正。经验证,模型计算所得的配比数据较生产实际数据有了较大的改善,实现了初始设定目标——降本增效,具有很好的推广应用价值。
Based on a model of steelmaking in converter burden ratio,the endpoint forecast model was constructed by RBF neural network.Combining with the target price function,a burden ratio model was constructed.At the same time,the modified steepest descent method was used to solve the optimal ratio.In addition,in order to improve the accuracy of the model calculation,the model used the least square method to correct the metal collection rate.Further,compared with the actual data,the calculated data is improved and it realizes the initial setting goal of reducing the cost and improving the benefit.It also has a good application value.
出处
《钢铁研究学报》
CAS
CSCD
北大核心
2016年第7期32-37,43,共7页
Journal of Iron and Steel Research
基金
安徽省"千人培养计划"资助项目(20120024)
中国安徽省高校优秀青年人才基金重点资助项目(2013SQRL024ZD)
关键词
炉料配比模型
终点预测模型
最速下降法
最小二乘法
降本增效
burden ratio's model
end point forecast model
steepest descent method
least square method
cost reduction and benefit improvement