摘要
为了确保工业复杂系统运行过程中的安全性和可靠性,对生产过程中的非线性数据进行预测分析成为一种有效手段。为了提高时间序列数据预测准确性,提出基于非线性归一化最小均方算法(LMS)的三阶Volterra自适应滤波器预测算法。首先针对时间序列数据的预测问题,利用有限项记忆单元的三阶Volterra级数对复杂系统运行数据进行预测。针对权重初始值会严重影响预测效果的问题,采用LMS自适应滤波算法对滤波器系数进行在线更新,对未来时刻的数据进行预测。最后利用联合循环发电厂数据对该预测算法进行实验,火电厂运行数据的预测值和实际观测值之间的误差很小,说明基于LMS的三阶Volterra自适应预测算法具有较好的预测效果,能够为实际的预测及控制提供有利的依据。
In order to ensure the safety and reliability of the industrial complex system in the process of operation,it has become an effective means to predict and analyze the nonlinear data in the production process.In order to improve the prediction accuracy of time series data,a third-order Volterra adaptive filter prediction algorithm based on nonlinear normalized least mean square(LMS)algorithm is proposed.Firstly,aiming at the prediction of time series data,the third-order Volterra series of finite term memory cell is used to predict the operation data of complex system.Aiming at the problem that the initial value of the weight will seriously affect the prediction effect,the LMS adaptive filter algorithm is used to update the filter coefficients online and predict the data at the future time.Finally,the prediction algorithm is tested with the data of thermal power plant.The error between the predicted value of the operation data of thermal power plant and the actual observation value is very small,which shows that the third-order Volterra adaptive prediction algorithm based on LMS has a good prediction effect and can provide a favorable basis for the actual prediction and control.
作者
张小洁
白蕾
ZHANG Xiaojie;BAI Lei(Western Innovation Research Institute;Shaanxi Polytechnic Institute,Shaanxi Xianyang 712000,China)
出处
《工业仪表与自动化装置》
2023年第1期108-111,115,共5页
Industrial Instrumentation & Automation
基金
2022年陕西省自然科学基础研究计划(青年人才项目)(2022JQ-609)
陕西省教育厅自然科学专项(20JK0802)
陕西省自然科学基础研究计划(2022JM-388)
陕西工业职业技术学院校级重点项目(2022YKZD-002)。
