人民黄河2025，Vol.47Issue(4)：141-144,151,5.DOI:10.3969/j.issn.1000-1379.2025.04.022
基于STOA-VMD和改进TCN模型的水泵机组振动趋势预测
Vibration Trend Prediction of Water Pump Units Based on STOA-VMD and Improved TCN Model
摘要
Abstract
Vibration trend prediction of water pumping units is an important initiative to ensure the normal operation of the units,while the complexity and nonlinearity of the vibration signals make the prediction difficult.Therefore,a vibration trend prediction model for water pump units based on STOA-VMD and improved Time Convolution Network(TCN)was proposed.Firstly,the Variable Modal Decomposition(VMD)parameters were optimized by using the Seagull Optimization Algorithm(STOA)to achieve the optimal adaptive decomposition of the vibration signal,and then each decomposed mode was predicted by using the improved TCN,and finally the final prediction result was ob-tained by superimposing all the results.Taking the pumping unit of a domestic rainwater pumping station as an example,the model validation was carried out based on the horizontal oscillation data of the water-guide bearing.The results show that the predicted values of the above combined model are basically consistent with the trend of the monitored values,and it has good predictive ability.Compared with the STOA-VMD-TCN,VMD-EnTCN,VMD-TCN,and TCN models,the proposed model has the smallest EMA、ERMS、EMAP,and the highest prediction accuracy.关键词
时间卷积网络/乌燕鸥算法/变分模态分解/振动信号/趋势预测/水泵机组Key words
Time Convolutional Network/Seagull Optimization Algorithm/Variational Mode Decomposition/vibration signal/trend predic-tion/water pump unit分类
建筑与水利引用本文复制引用
王伟生,张宁,邢磊,周保林,郭新帅,安东,高源,张孝远..基于STOA-VMD和改进TCN模型的水泵机组振动趋势预测[J].人民黄河,2025,47(4):141-144,151,5.基金项目
河南省科技研发计划联合基金资助项目(225200810038) （225200810038）
河南省自然科学基金资助项目(232300421207) （232300421207）
