基于生产间歇改进Elman的转炉煤气发生量预测

doi:10.16182/j.issn1004731x.joss.23-0022

系统仿真学报 ›› 2024, Vol. 36 ›› Issue (5): 1179-1188.doi: 10.16182/j.issn1004731x.joss.23-0022

基于生产间歇改进Elman的转炉煤气发生量预测

费佳杰¹(), 吴定会¹(), 范俊岩¹, 汪晶²

^1.江南大学物联网工程学院，江苏无锡 214122
^2.上海宝信软件股份有限公司，上海 201203

收稿日期:2023-01-06 修回日期:2023-02-13 出版日期:2024-05-15 发布日期:2024-05-21
通讯作者: 吴定会 E-mail:404623746@qq.com;wdh123@jiangnan.edu.cn
第一作者简介:费佳杰(1999-)，男，硕士生，研究方向为能源预测。E-mail：404623746@qq.com
基金资助:
国家重点研发计划(2020YFB1711102)

Prediction of Converter Gas Generation Based on Intermission Production Improved Elman

Fei Jiajie¹(), Wu Dinghui¹(), Fan Junyan¹, Wang Jing²

^1.College of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China
^2.Shanghai Baosight Software Co. Ltd, Shanghai 201203, China

Received:2023-01-06 Revised:2023-02-13 Online:2024-05-15 Published:2024-05-21
Contact: Wu Dinghui E-mail:404623746@qq.com;wdh123@jiangnan.edu.cn

摘要/Abstract

摘要：

针对钢铁企业转炉煤气发生量间歇时长波动大，预测精度低的问题，基于生产间歇特征分类，提出基于混沌映射粒子群算法（CPSO）优化Elman神经网络的转炉煤气发生量预测模型（CPSO-Elman）。提取转炉煤气发生量时间序列中生产间歇特征，并根据间歇时长进行分类；引入经混沌扰动改进的PSO算法优化ENN的初始权值和阈值，利用非线性更新的惯性权重以平衡全局搜索与局部搜索能力，并在粒子初始化中添加了混沌映射；构建CPSO-Elman转炉煤气发生量组合预测模型；在预测未来时间内间歇时长基础上，预测转炉煤气发生量。仿真结果表明：所提方法在预测精度上比未经过优化而预测的方法提高了5%左右。

关键词: 转炉煤气, 发生量预测, PSO算法, 混沌扰动, Elman神经网络, 间歇分类

Abstract:

Aiming at large fluctuations of intermission and low prediction accuracy in iron and steel industry, based on the classification of intermission characteristics, a converter gas generation predicting model(CPSO-Elman) based on Elman neural network(ENN) optimized by chaoticPSO(CPSO) algorithm is proposed. The intermittent characteristics of converter gas generation time series are extracted and raw data is classified according to intermittent duration. The PSO algorithm improved by chaotic disturbance is introduced to optimize the initial weight and threshold of ENN and inertia weight of nonlinear updating is designed to balance global search ability and local search ability. Construct the combined prediction model of CPSO-Elman converter gas generation. Converter gas generation is predicted on the basis of predicting the intermission in the future time. Simulation results show that prediction accuracy of the proposed method is about 5% higher than that of the method without optimization.

Key words: converter gas, generation prediction, PSO algorithm, chaotic perturbation, Elman neural network, intermission classification

中图分类号:

TP391.9

费佳杰,吴定会,范俊岩等 . 基于生产间歇改进Elman的转炉煤气发生量预测[J]. 系统仿真学报, 2024, 36(5): 1179-1188.

Fei Jiajie,Wu Dinghui,Fan Junyan,et al . Prediction of Converter Gas Generation Based on Intermission Production Improved Elman[J]. Journal of System Simulation, 2024, 36(5): 1179-1188.

图/表 16

图1

图2

表1

图3

图4

表2

表3

算法参数实验结果

种群规模	$c 1 、 c 2$	MAE/ $(k m 3 / h)$	MAPE/%
30	2.00	8.46	3.35
30	1.49	7.89	2.46
40	2.00	7.44	1.98
40	1.49	7.49	2.42
50	2.00	10.10	4.54
50	1.49	9.49	1.99

表3

表4

各类别间歇下算法参数设置

参数	短	中	长
种群规模	30	40	50
$c 1$ 、 $c 2$	1.49	2.00	1.49

表4

图5

表5

图6

图7

表6

间歇时长未变动下LDG发生量预测误差

模型	MAE/( $k m 3 / h)$	MAPE/%
Elman	7.66	7.28
PSO-Elman	5.45	5.32
CPSO-Elman	4.28	4.13

表6

图8

图9

表7

间歇时长变动下LDG发生量预测误差

模型	MAE/( $k m 3 / h$ )	MAPE/%
Elman	10.01	9.16
PSO-Elman	7.55	6.79
CPSO-Elman	6.47	5.14

表7

参考文献 16

1	Ishaq H, Dincer I, Naterer G F. Exergy-based Thermal Management of a Steelmaking Process Linked with a Multi-generation Power and Desalination System[J]. Energy, 2018, 159: 1206-1217.
2	Wen Zongguo, Wang Yihan, Zhang Chenkai, et al. Uncertainty Analysis of Industrial Energy Conservation Management in China's Iron and Steel Industry[J]. Journal of Environmental Management, 2018, 225: 205-214.
3	An Runying, Yu Biying, Li Ru, et al. Potential of Energy Savings and CO₂ Emission Reduction in China's Iron and Steel Industry[J]. Applied Energy, 2018, 226: 862-880.
4	Jin Feng, Zhao Jun, Liu Ying, et al. A Scheduling Approach with Uncertainties in Generation and Consumption for Converter Gas System in Steel Industry[J]. Information Sciences, 2021, 546: 312-328.
5	徐化岩, 马家琳. 基于数据驱动的高炉煤气复合预测模型[J]. 中国冶金, 2019, 29(7): 56-60.
	Xu Huayan, Ma Jialin. Composite Prediction Model of Blast Furnace Gas Based on Data Driven[J]. China Metallurgy, 2019, 29(7): 56-60.
6	Zhang Qi, Li Hui, Ma Jialin, et al. Dynamic Forecasting and Optimal Scheduling of By-product Gases in Integrated Iron and Steel Works[J]. Journal of Iron and Steel Research International, 2019, 26(5): 529-546.
7	Zhao Jun, Wang Wei, Liu Ying, et al. A Two-stage Online Prediction Method for a Blast Furnace Gas System and Its Application[J]. IEEE Transactions on Control Systems Technology, 2011, 19(3): 507-520.
8	Li Lei, Li Hongjuan. Forecasting and Optimal Probabilistic Scheduling of Surplus Gas Systems in Iron and Steel Industry[J]. Journal of Central South University, 2015, 22(4): 1437-1447.
9	Zhao Jun, Wang Wei, Pedrycz Witold, et al. Online Parameter Optimization-based Prediction for Converter Gas System by Parallel Strategies[J]. IEEE Transactions on Control Systems Technology, 2012, 20(3): 835-845.
10	Han Zhongyang, Liu Ying, Zhao Jun, et al. Real Time Prediction for Converter Gas Tank Levels Based on Multi-output Least Square Support Vector Regressor[J]. Control Engineering Practice, 2012, 20(12): 1400-1409.
11	Wang Jibin. A Deep Learning Approach for Atrial Fibrillation Signals Classification Based on Convolutional and Modified Elman Neural Network[J]. Future Generation Computer Systems, 2020, 102: 670-679.
12	Zhang Ying, Wang Xiaoping, Tang Huiming. An Improved Elman Neural Network with Piecewise Weighted Gradient for Time Series Prediction[J]. Neurocomputing, 2019, 359: 199-208.
13	Li Guoqing, Wang Wanliang, Zhang Weiwei, et al. Grid Search Based Multi-population Particle Swarm Optimization Algorithm for Multimodal Multi-objective Optimization[J]. Swarm and Evolutionary Computation, 2021, 62: 100843.
14	Zeng Nianyin, Wang Zidong, Liu Weibo, et al. A Dynamic Neighborhood-based Switching Particle Swarm Optimization Algorithm[J]. IEEE Transactions on Cybernetics, 2022, 52(9): 9290-9301.
15	曾艳阳, 冯云霞, 赵文涛. 基于logistic映射的自适应变尺度混沌粒子群算法[J]. 系统仿真学报, 2017, 29(10): 2241-2246.
	Zeng Yanyang, Feng Yunxia, Zhao Wentao. Adaptive Mutative Scale Chaos Particles Swarm Optimization Based on Logistic Mapping[J]. Journal of System Simulation, 2017, 29(10): 2241-2246.
16	Frans van den Bergh. An Analysis of Particle Swarm Optimizers[D]. Pretoria: University of Pretoria, 2006.

间歇类别	个数	占比/%	持续时间/min
总间歇	1 390	100	20~100
短间歇	6	0.4	<25
中间歇	971	69.9	25~45
长间歇	413	29.7	>45

优化算法	迭代次数	适应度值
PSO	26	464.68
CPSO	23	451.55

[1]	牛侃, 李冰, 付强. 基于混沌扰动机制粒子群算法的战场频率分配方法[J]. 系统仿真学报, 2021, 33(8): 1905-1913.
[2]	李昌华, 毕成功, 李智杰. 基于改进PSO算法的人群疏散模型[J]. 系统仿真学报, 2020, 32(6): 1000-1008.
[3]	李照, 靳雁霞, 秦志鹏, 任超. 基于优化Snake模型的变形物体碰撞检测算法研究[J]. 系统仿真学报, 2018, 30(1): 62-68.
[4]	柯钢. 基于合作对策论的网络安全态势组合预测模型[J]. 系统仿真学报, 2017, 29(5): 1153-1159.
[5]	张宏立, 李瑞国, 范文慧. 基于相空间重构的Bernstein神经网络混沌序列预测[J]. 系统仿真学报, 2016, 28(4): 880-889.
[6]	李小喜, 陈浩光, 李大喜, 陈疆萍. 基于Elman神经网络的作战效能预测模型研究[J]. 系统仿真学报, 2015, 27(1): 43-49.

基于生产间歇改进Elman的转炉煤气发生量预测

Prediction of Converter Gas Generation Based on Intermission Production Improved Elman

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 16

参考文献 16

相关文章 6

编辑推荐

Metrics

本文评价

参数	短	中	长
输入层节点	35	45	55
隐含层节点	7	9	12
粒子维数	260	504	685