基于模糊时间序列的给水泵指标预测  

Index prediction of feed pump based on fuzzy time series

作　　者：周东阳[1] 万松森[1] 曹军[1] 毕胜山[2] ZHOU Dongyang;WAN Songsen;CAO Jun;BI Shengshan(Xi’an Thermal Engineering Research Institute Co.,Ltd.,Xi’an 710054,China;School of Energy and Power Engineering,Xi’an Jiaotong University,Xi’an 710049,China)

机构地区：[1]西安热工研究院有限公司,陕西西安710054 [2]西安交通大学能源与动力工程学院,陕西西安710049

出　　处：《电子设计工程》2023年第10期16-19,24,共5页Electronic Design Engineering

基　　金：中国华电科技项目(CHDKJ20-04-110)。

摘　　要：对火电厂给水泵系统设备的指标数据进行有效预测,有助于保障设备安全、可靠运行,提高指标数据预测的准确性具有重要意义。文中提出基于模糊时间序列算法来预测给水泵系统设备的指标数据,同时采用该算法在给水泵的真实运行数据集上进行实验,分别对给水泵指标出口压力及温度进行预测。结果显示,给水泵出口压力预测模型平均绝对百分比误差为0.009 5,均方根误差为0.250 5,而出口温度的模型平均绝对百分比误差为0.001 4,均方根误差为0.366 2,证实了模糊时间序列预测算法的预测误差较低,且能够应用于给水泵系统设备的指标预测中。Effective prediction of index data of feed pump system equipment in thermal power plant is of great significance to ensure the safe and reliable operation of equipment and improve the accuracy of index data prediction.In this paper,a fuzzy time series prediction algorithm is proposed to predict the equipment index data of feed pump system.Based on the real operation data set of feed pump,the fuzzy time series algorithm is used to predict the index outlet pressure and outlet temperature of feed pump respectively.The results show that the prediction model MAPE for the outlet pressure of the feedwater pump is 0.0095,with a root mean square error is 0.2505,while the model MAPE for the outlet temperature is 0.0014,and the root mean square error is 0.3662.It is verified that the prediction error of the fuzzy time series prediction algorithm proposed in this paper is low,and can be applied to the index prediction of feed pump system equipment.

关 键 词：预测 模糊时间序列 给水泵 给水泵出口压力 给水泵出口温度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TN99[自动化与计算机技术—计算机科学与技术]