摘要
针对已有水电机组振动趋势预测模型的局限性,提出了一种基于最优变分模态分解(OVMD)、时变滤波器经验模态分解(TVFEMD)、猎人猎物优化算法(HPO)和极限学习机(ELM)的水电机组振动趋势预测方法。该方法先通过OVMD对原始水电机组振动信号进行自适应分解,进一步采用TVFEND对分解后的残差进行二次分解。然后建立各子序列的HPO-ELM振动趋势预测模型;叠加重构所有子序列预测结果获得最终的预测振动信号。研究结果表明,该方法预测效果明显优于传统方法,有效提高了水电机组振动趋势预测精度,具有较好的工程应用价值。
In order to address the limitations of the existing vibration trend prediction model for hydroelectric units,a vibration trend prediction method for hydroelectric units based on optimal variational mode decomposition(OVMD),time-varying filter empirical mode decomposition(TVFEMD),hunter-prey optimization algorithm(HPO),and extreme learning machine(ELM) is proposed.This method first applies OVMD to adaptively decompose the original vibration signal of the hydroelectric unit,and then further employs TVFEMD to perform a secondary decomposition of the residuals obtained from the first decomposition.Subsequently,vibration trend prediction models HPO-ELM are established for each subsequence.The final predicted vibration signal is obtained by aggregating and reconstructing the prediction results of all the sub-sequences.The research results demonstrate that this method outperforms traditional methods in terms of prediction accuracy for the vibration trend of hydroelectric units,and it has good engineering application value.
作者
张楠
朱永奇
孙娜
赖昕杰
李超顺
ZHANG Nan;ZHU Yong-qi;SUN Na;LAI Xin-jie;LI Chao-shun(Jiangsu Key Laboratory of Advanced Manufacturing Technology,Huaiyin Institute of Technology,Huaian 223299,China;Faculty of Automation,Huaiyin Institute of Technology,Huaian 223299,China;Power China Huadong Engineering Corporation Limited,Hangzhou 311122,China;School of Civil and Hydraulic Engineering,Huazhong University of Science and Technology,Wuhan 430074,China)
出处
《水电能源科学》
北大核心
2023年第10期204-207,199,共5页
Water Resources and Power
基金
江苏省自然科学基金项目(BK20201069)
江苏省高校自然科学基金面上项目(20KJD480003)
江苏省双创计划(JSSCBS(2020)31038)。
作者简介
张楠(1991-),男,讲师、硕导,研究方向为水电、风电等清洁能源优化运行与控制、人工智能应用,E-mail:zhangnanhust@163.com;通讯作者:孙娜(1992-),女,讲师、硕导,研究方向为水文水资源、人工智能应用,E-mail:sunna1347@126.com。
