基于深度信念网络模型的多目标优化
|
摘要
为降低炼焦能耗,提高焦炭产量和质量,准确建立生产目标模型,提出基于深度信念网络模型的多目标优化研究方案。根据现场专家经验及生产现状确定能耗和产量为生产目标,对采集的炼焦数据进行处理和相关性分析,分别建立能耗和产量的深度信念网络模型及质量径向基神经网络模型,并且采用差分扰动的粒子群多目标优化算法进行集气管压力设定值优化,通过仿真研究验证了该方案的可行性。实验表明,该方案能准确地挖掘数据间的复杂特性,建立精准的目标模型,并得出最佳的集气管压力设定值,使炼焦能耗降低并且产量提高,可以为实际生产提供理论指导。
In order to reduce coking energy consumption,improve coke yield and quality,accurately establish a production target model,a multi-objective optimization model was proposed,which was based on deep belief network. According to the field expert experience and production status,the energy consumption and the coke yield were determined as production targets. The collected coking data was processed and correlated,and a deep belief network model for the coking energy consumption and coke yield and a mass radial basis neural network model for the coke quality were established respectively. The differential particle swarm optimization was used to optimize the set value of the gas collector pressure. The feasibility of the method was verified by simulation. Experiments show that the method can accurately mine the complex characteristics between data,establish a precise target model,and obtain the best set value of the collector pressure,which can reduce the energy consumption of coking and increase the coke yield. This method can also provide theoretical guidance for actual production.
In order to reduce coking energy consumption,improve coke yield and quality,accurately establish a production target model,a multi-objective optimization model was proposed,which was based on deep belief network. According to the field expert experience and production status,the energy consumption and the coke yield were determined as production targets. The collected coking data was processed and correlated,and a deep belief network model for the coking energy consumption and coke yield and a mass radial basis neural network model for the coke quality were established respectively. The differential particle swarm optimization was used to optimize the set value of the gas collector pressure. The feasibility of the method was verified by simulation. Experiments show that the method can accurately mine the complex characteristics between data,establish a precise target model,and obtain the best set value of the collector pressure,which can reduce the energy consumption of coking and increase the coke yield. This method can also provide theoretical guidance for actual production.
引文
1张莹.炼焦工业的发展趋势[J].科技情报开发与经济,2005,15(8):149-150Zhang Ying. The developing trend of coking industry[J]. Sci/Tech Information Development and Economy,2005,15(8):149-150
2 吕太,刘振华.基于随机孔模型的单颗粒多孔焦炭富氧燃烧特性[J].科学技术与工程,2018,18(7):12-17LüTai,Liu Zhenhua. Characteristics of single porous char particle oxy-fuel combustion with random pore model[J]. Science Technology and Engineering,2018,18(7):12-17
3 赖旭芝,李爱平,吴敏,等.基于多目标遗传算法的炼焦生产过程优化控制[J].计算机集成制造系统,2009,15(5):990-997Lai Xuzhi,Li Aiping,Wu Min,et al. Optimization control based on the multi-objective genetic algorithm for coking plant production process[J]. Computer Integrated Manufacturing Systems,2009,15(5):990-997
4 王介生,高宪文,刘琳.基于减法聚类的焦炉集气管压力操作模式提取及迁移重构[J].化工学报,2013,64(12):4468-4473Wang Jiesheng,Gao Xianwen,Liu Lin. Subtractive clustering algorithm based operation pattern extraction and migration reconfiguration of coke oven collector pressure[J]. CIESC Journal,2013,64(12):4468-4473
5 刘昕明,高宪文,翁永鹏.数据驱动闭环子空间预测控制方法研究与应用[J].控制与决策,2014,29(5):913-918Liu Xinming,Gao Xianwen,Weng Yongpeng. Research and application of closed-loop subspace predictive control based on data driven[J]. Control and Decision,2014,29(5):913-918
6 Li K,Li D W,Xi Y G,et al. Model predictive control with feed forward strategy for gas collectors of coke ovens[J]. Chinese Journal of Chemical Engineering,2014,22(7)
7 田沛,温兴贤.热工过程建模中的数据处理方法研究[J].计算机仿真,2013,30(8):147-150Tian Pei,Wen Xingxian. Research on data processing method in thermal process modeling[J]. Computer Simulation,2013,30(8):147-150
8 刘方园,王水花,张煜东.深度置信网络模型及应用研究综述[J].计算机工程与应用,2018,54(1):11-18,47Liu Fangyuan,Wang Shuihua,Zhang Yudong. Survey of deep belief network model and its applications[J]. Computer Engineering and Applications,2018,54(1):11-18,47
9 杨佳玲,赵涓涓,强彦,等.基于深度信念网络的肺结节良恶性分类[J].科学技术与工程,2016,16(32):69-74Yang Jialing,Zhao Juanjuan,Qiang Yan,et al. A classification method of pulmonary nodules based on deep belief network[J]. Science Technology and Engineering,2016,16(32):69-74
10 姚腾辉,李峰.基于深度信念网络的建筑物用水流量预测[J].软件导刊,2018,17(10):36-40Yao Tenghui,Li Feng. Building water flow prediction based on deep belief network[J]. Software Guide,2018,17(10):36-40
11 谢彬,刘利华,张召峰.基于RBF的交通流预测中加速OLS寻优[J].科学技术与工程,2009,9(2):280-282,288Xie Bin,Liu Lihua,Zhang Zhaofeng. Speeding up optimization searching of the OLS algorithm with in traffic flow prediction based on the RBF network[J]. Science Technology and Engineering,2009,9(2):280-282,288
12 胡晨昊,杨瑞峰,郭晨霞,等.引入粒子群算法加速因子和个体认知的改进蝙蝠算法[J].科学技术与工程,2017,17(35):277-282Hu Chenhao,Yang Ruifeng,Guo Chenxia,et al. Improved bat algorithm with particle swarm optimization acceleration factor and individual cognition[J]. Science Technology and Engineering,2017,17(35):277-282
13 万东.差分进化算法研究及其应用[J].科学技术与工程,2009,9(22):6673-6676Wan Dong. Research and application based on differential evolution algorithm[J]. Science Technology and Engineering,2009,9(22):6673-6676
2 吕太,刘振华.基于随机孔模型的单颗粒多孔焦炭富氧燃烧特性[J].科学技术与工程,2018,18(7):12-17LüTai,Liu Zhenhua. Characteristics of single porous char particle oxy-fuel combustion with random pore model[J]. Science Technology and Engineering,2018,18(7):12-17
3 赖旭芝,李爱平,吴敏,等.基于多目标遗传算法的炼焦生产过程优化控制[J].计算机集成制造系统,2009,15(5):990-997Lai Xuzhi,Li Aiping,Wu Min,et al. Optimization control based on the multi-objective genetic algorithm for coking plant production process[J]. Computer Integrated Manufacturing Systems,2009,15(5):990-997
4 王介生,高宪文,刘琳.基于减法聚类的焦炉集气管压力操作模式提取及迁移重构[J].化工学报,2013,64(12):4468-4473Wang Jiesheng,Gao Xianwen,Liu Lin. Subtractive clustering algorithm based operation pattern extraction and migration reconfiguration of coke oven collector pressure[J]. CIESC Journal,2013,64(12):4468-4473
5 刘昕明,高宪文,翁永鹏.数据驱动闭环子空间预测控制方法研究与应用[J].控制与决策,2014,29(5):913-918Liu Xinming,Gao Xianwen,Weng Yongpeng. Research and application of closed-loop subspace predictive control based on data driven[J]. Control and Decision,2014,29(5):913-918
6 Li K,Li D W,Xi Y G,et al. Model predictive control with feed forward strategy for gas collectors of coke ovens[J]. Chinese Journal of Chemical Engineering,2014,22(7)
7 田沛,温兴贤.热工过程建模中的数据处理方法研究[J].计算机仿真,2013,30(8):147-150Tian Pei,Wen Xingxian. Research on data processing method in thermal process modeling[J]. Computer Simulation,2013,30(8):147-150
8 刘方园,王水花,张煜东.深度置信网络模型及应用研究综述[J].计算机工程与应用,2018,54(1):11-18,47Liu Fangyuan,Wang Shuihua,Zhang Yudong. Survey of deep belief network model and its applications[J]. Computer Engineering and Applications,2018,54(1):11-18,47
9 杨佳玲,赵涓涓,强彦,等.基于深度信念网络的肺结节良恶性分类[J].科学技术与工程,2016,16(32):69-74Yang Jialing,Zhao Juanjuan,Qiang Yan,et al. A classification method of pulmonary nodules based on deep belief network[J]. Science Technology and Engineering,2016,16(32):69-74
10 姚腾辉,李峰.基于深度信念网络的建筑物用水流量预测[J].软件导刊,2018,17(10):36-40Yao Tenghui,Li Feng. Building water flow prediction based on deep belief network[J]. Software Guide,2018,17(10):36-40
11 谢彬,刘利华,张召峰.基于RBF的交通流预测中加速OLS寻优[J].科学技术与工程,2009,9(2):280-282,288Xie Bin,Liu Lihua,Zhang Zhaofeng. Speeding up optimization searching of the OLS algorithm with in traffic flow prediction based on the RBF network[J]. Science Technology and Engineering,2009,9(2):280-282,288
12 胡晨昊,杨瑞峰,郭晨霞,等.引入粒子群算法加速因子和个体认知的改进蝙蝠算法[J].科学技术与工程,2017,17(35):277-282Hu Chenhao,Yang Ruifeng,Guo Chenxia,et al. Improved bat algorithm with particle swarm optimization acceleration factor and individual cognition[J]. Science Technology and Engineering,2017,17(35):277-282
13 万东.差分进化算法研究及其应用[J].科学技术与工程,2009,9(22):6673-6676Wan Dong. Research and application based on differential evolution algorithm[J]. Science Technology and Engineering,2009,9(22):6673-6676