基于DE-SARIMA方法的机组产出量预测
Unit Output Prediction Based on DE-SARIMA Method
黄俊婷 1王忠鑫 1滕用秋 2宋波 1曾祥玉 1白仁喜3
作者信息
- 1. 中煤科工集团沈阳设计研究院有限公司,辽宁 沈阳 110015
- 2. 国网辽宁省电力有限公司沈阳供电公司,辽宁 沈阳 110052
- 3. 国家能源集团煤炭运输部,北京 100010
- 折叠
摘要
获得高精度机组产出量预测是维持钢铁企业库存量合理稳定的前提.受生产环境复杂多变和市场需求不确定性的影响,机组产出量表现出一定波动.为此,提出基于差分进化——季节性自回归整合移动平均(DE-SARIMA)的机组产出量预测方法.为提高预测精度,设计带有分段迭代自适应变异策略和DBSCAN选择策略的DE算法对SARIMA模型的参数进行寻优.采用实际生产数据进行测试,验证了所提方法的有效性.
Abstract
Obtaining high-precision unit output prediction is a prerequisite for steel enterprises to maintain reasonable and stable inventory levels.Due to the complex and ever-changing production environment and the uncertainty of market demand,the unit output shows certain fluctuations.To this end,a unit output prediction method based on differential evolution and seasonal autoregressive integrated moving average(DE-SARIMA)is proposed.To improve prediction accuracy,a DE algorithm integrating the segmented iterative adaptive mutation strategy and DBSCAN selection strategy is designed to optimize the parameters of the SARIMA model.Testing is carried out using actual production data,and the effectiveness of the proposed method is verified.
关键词
SARIMA/DBSCAN/DE/产出量/预测Key words
SARIMA/DBSCAN/DE/output/prediction引用本文复制引用
基金项目
沈阳市科技人才专项(RC230427)
沈阳市中青年科技创新人才支持计划(RC220436)
出版年
2024
国家科技期刊平台
NSTL
万方数据