【作者】 侯鹏飞； 胡磊； 王琦； 白建云；

【Author】 Hou Peng-fei;Hu Lei;Wang Qi;Bai Jian-yun;School of Automation and Software,Shanxi University;

【机构】 山西大学自动化与软件学院；

【摘要】 超临界机组的给水流量预测具有重要意义,可以为给水的预判和实时调整提供决策依据。为了提高超临界机组给水流量的预测精度,本文以山西某电厂350MW超临界CFB机组的给水控制系统为研究对象,以采集到的该电厂2020年8月1日的某段时间内机组运行历史数据对ELM算法进行训练,使用遗传算法对ELM算法的参数进行寻优,建立GAELM算法预测模型。仿真结果表明,GA-ELM算法的预测性能优于ELM算法。更多还原

【Abstract】 The prediction of feed water flow of supercritical units is of great significance,which can provide decision-making basis for the prediction and real-time adjustment of feed water.In order to improve the prediction accuracy of supercritical unit feed water flow,this paper takes the 350 mw supercritical CFB unit of a certain power plant in Shanxi Province of the feed water control system as the research object,to collected the plant on August 1,2020,a period of time the unit running history data of ELM algorithm for training,using the genetic algorithm for parameters optimization of ELM algorithm,GA-ELM algorithm prediction model was established.Simulation results show that the prediction performance of GA-ELM algorithm is better than that of ELM algorithm.更多还原