直流电弧炉无功超短期滚动预测及控制  

DC EAF Reactive Power Ultra-short-term Rolling Prediction and Control

作　　者：周文 苏灿 高敏 朱明星 李铁成[1] 孟良 ZHOU Wen;SU Can;GAO Min;ZHU Mingxing;LI Tiecheng;MENG Liang(State Grid Hebei Electric Power Research Institute,Shijiazhuang 050021,China;Anhui University Green Industry Innovation Research Institute,Hefei 230601,China;School of Electrical Engineering and Automation,Anhui University,Hefei 230039,China)

机构地区：[1]国网河北省电力有限公司电力科学研究院,石家庄050021 [2]安徽大学绿色产业创新研究院,合肥230601 [3]安徽大学电气工程与自动化学院,合肥230039

出　　处：《电力系统及其自动化学报》2023年第1期96-102,共7页Proceedings of the CSU-EPSA

基　　金：国网河北能源技术服务有限公司科技项目(TSS2020-14)。

摘　　要：为提升直流电弧炉系统SVC装置动态补偿性能,提出了一种直流电弧炉无功超短期滚动预测及控制方法。首先,分析并揭示了直流电弧炉的控制原理及无功特性;其次,建立了直流电弧炉超短期无功滚动预测模型,基于AIC信息准则和历史冶炼数据实现了预测模型的定阶;然后,基于预测模型提出了直流电弧炉无功超短期滚动预测控制方法;最后,对预测及控制效果进行了仿真验证。与传统控制方法相比,采用预测控制算法后,瞬时闪变降低了31.16%。所提预测控制算法对直流电弧炉系统SVC装置的闪变抑制效果改善显著。To improve the dynamic compensation performance of SVC in a direct current electric arc furnace(DC EAF),a DC EAF reactive power ultra-short-term rolling prediction and control method is proposed. First,the control principle for the DC EAF and its reactive power characteristics are analyzed and revealed. Second,a DC EAF ultrashort-term reactive power rolling prediction model is established,and its fixed order is realized based on the Akaike information criterion(AIC)and historical smelting data. Third,the DC EAF reactive power ultra-short-term rolling prediction and control method is proposed based on the prediction model. Finally,the prediction and control effects are verified by simulations. Compared with that under the conventional control method,the instantaneous flicker is reduced by31.16% after the use of the proposed prediction and control method,indicating that the novel method can significantly improve the flicker suppression effect on the SVC device in the DC EAF system.

关 键 词：直流电弧炉 预测模型 超短期 滚动预测 预测控制 

分 类 号：TM711[电气工程—电力系统及自动化]