转炉炼钢终点碳温预测与控制模型优化研究
Research on the Prediction and Control Model Optimization of End Carbon Temperature in Converter Steelmaking
孙大成1
作者信息
摘要
为提高转炉炼钢的生产效率和经济效益,提出以改进神经网络学习极限机的预测模型为基础,引入改进的粒子群算法作为终点碳温控制优化模型.经过研究表明,转炉炼钢温度偏差在15℃左右的样本有89个,命中率为62.676％.碳含量w(C)偏差在0.015左右的样本有105个,命中率为72.112％.由此可见,预测模型及控制模型对转炉炼钢终点碳温控制的有效性以及降低能源消耗均具有重要意义.
Abstract
In order to improve the production efficiency and economic benefits of converter steelmaking,an improved neural network learning extreme machine prediction model is proposed,and an improved particle swarm optimization algorithm is introduced as the optimization model for endpoint carbon temperature control.After research,it has been shown that there are 89 samples with a converter steelmaking temperature deviation of around 15 ℃,and the hit rate is 62.676％.There are 105 samples with a carbon content deviation w(C)of around 0.015,with a hit rate of 72.112％.It can be seen that the effectiveness of prediction and control models in controlling the endpoint carbon temperature of converter steelmaking and reducing energy consumption are of great significance.
关键词
转炉炼钢/终点碳温/神经网络/学习极限机Key words
converter steelmaking/end carbon temperature/neural networks/learning extreme machine引用本文复制引用
出版年
2024
NETL
NSTL
万方数据