基于融合注意力机制的多智能体磁浮谐波预测算法

王宗艳; 毛中亚; 黄世泽; 郭其一

doi:10.13890/j.issn.1000-128X.2023.06.014

您当前的位置：

首页 >

文章列表页 >

基于融合注意力机制的多智能体磁浮谐波预测算法

智能化技术 | 更新时间：2024-08-02

- 基于融合注意力机制的多智能体磁浮谐波预测算法
- Multi-agent maglev harmonic prediction algorithm based on integrated attention mechanism
- 机车电传动 2023年第6期页码：114-121
- 作者机构：
  
  1.同济大学电子与信息工程学院，上海　　201804
  2.同济大学交通运输工程学院，上海;201804
- 作者简介：
  
  王宗艳（1999—），女，硕士研究生，主要从事深度学习与地铁牵引系统谐波治理方面的研究；E-mail: 2133078@tongji.edu.com
- 基金信息：
  
  国家重点研发计划项目(2017YFB1200901-11)
- DOI：10.13890/j.issn.1000-128X.2023.06.014
  中图分类号： U237
- 纸质出版日期：2023-11-10，
  
  收稿日期：2023-07-12，
  
  修回日期：2023-08-25，
- 稿件说明：
扫描看全文
王宗艳, 毛中亚, 黄世泽, 等. 基于融合注意力机制的多智能体磁浮谐波预测算法[J]. 机车电传动, 2023(6): 114-121.

WANG Zongyan, MAO Zhongya, HUANG Shize, et al. Multi-agent maglev harmonic prediction algorithm based on integrated attention mechanism[J]. Electric drive for locomotives,2023(6): 114-121.
王宗艳, 毛中亚, 黄世泽, 等. 基于融合注意力机制的多智能体磁浮谐波预测算法[J]. 机车电传动, 2023(6): 114-121. DOI： 10.13890/j.issn.1000-128X.2023.06.014.

WANG Zongyan, MAO Zhongya, HUANG Shize, et al. Multi-agent maglev harmonic prediction algorithm based on integrated attention mechanism[J]. Electric drive for locomotives,2023(6): 114-121. DOI： 10.13890/j.issn.1000-128X.2023.06.014.

摘要

为减小高速磁浮系统谐波对牵引供电网产生的影响，解决谐波治理存在的时滞性问题，通常需要对谐波电流进行预测。对此，采用融合深度学习算法的组合模型相比于传统算法的表现更加出色。文章提出一种新的融合注意力机制的多智能体磁浮谐波预测算法，该算法通过麻雀搜索算法（SSA）优化变分模态分解（VMD）的相关参数，并利用该优化参数将原始电流信号分解为多个不同中心频率的谐波分量，将各个分量分别输入融合注意力机制的长短时记忆网络（LSTM）中进行时序预测，形成多个独立预测智能体，再对各个智能体预测结果进行重构，从而实现对高速磁浮谐波电流预测。在此基础上引入误差修正机制，进一步提高模型的预测精度。对上海高速磁浮牵引系统进行理论分析与仿真，采集网侧电流数据，并用所提出的算法模型对此数据进行试验和分析。结果表明，与其他模型相比，所提的预测模型在磁浮谐波电流预测方面具有较好的性能，并可使时滞性得到进一步改善。

Abstract

To mitigate the impact of high-speed maglev harmonics on the traction power supply network and address the time lag in harmonic control

predicting harmonic current is commonly necessary. In this context

a superior alternative to traditional algorithms is a combined model that integrates deep learning algorithms. In this paper

a new multi-agent maglev harmonic prediction algorithm with integrated attention mechanism was proposed. The algorithm optimized the parameters of variational mode decomposition (VMD) by sparrow search algorithm (SSA)

and used these optimized parameters to decompose the original current signal into harmonic components with different center frequencies. Subsequently

all components were input into the long short-term memory (LSTM) network of the integrated attention mechanism for time series prediction

and multiple independent prediction agents were formed

followed by the reconstruction of each agent's prediction results

so as to predict the harmonic current of the high-speed maglev. On this basis

the error correction mechanism was introduced to enhance the prediction accuracy of the model. Theoretical analysis and simulations of Shanghai high-speed maglev traction system were conducted

with the grid-side current data collected for experiment and analysis using the proposed algorithm model. The results indicate that

compared with other models

the proposed prediction model demonstrated superior performance in predicting maglev harmonic current

with the potential for further reducing the time lag.

关键词

谐波预测麻雀搜索算法变分模态分解长短时记忆神经网络注意力机制

Keywords

harmonic predictionsparrow search algorithmvariational mode decompositionlong short-term memory neural networkattention mechanism

references

谭保华, 张文宇, 黄程旭, 等. 基于全相位FFT三谱线校正的电网谐波与间谐波检测算法[J]. 华中师范大学学报(自然科学版), 2021, 55(6): 1044-1050.

TAN Baohua, ZHANG Wenyu, HUANG Chengxu, et al. Harmonic and interharmonic detection algorithm based on all-phase FFT triple spectrum line correction[J]. Journal of central China normal university (natural sciences), 2021, 55(6): 1044-1050.

CZARNECKI L S. Instantaneous reactive power p-q theory and power properties of three-phase systems[J]. IEEE transactions on power delivery, 2006, 21(1): 362-367.

周建萍, 郑应平, 王志萍. 基于小波包变换的电力系统谐波分析[J]. 中国电力, 2009, 42(7): 6-9.

ZHOU Jianping, ZHENG Yingping, WANG Zhiping. Harmonic analysis of power system using wavelet packet transform[J]. Electric power, 2009, 42(7): 6-9.

WANG Rongkun, HUANG Wenjie, HU Bingtao, et al. Harmonic detection for active power filter based on two-step improved EEMD[J]. IEEE transactions on instrumentation and measurement, 2022, 71: 1-10.

YUE Qianqian, HU Rui, ZHANG Xiaoling. A power system harmonic problem based on the BP neural network learning algorithm[J]. Computational intelligence and neuroscience, 2022, 2022: 7247881.

LIU Nixuan, FEI Juntao. Adaptive fractional sliding mode control of active power filter based on dual RBF neural networks[J]. IEEE access, 2017, 5: 27590-27598.

KUYUMANI E M, HASAN A N, SHONGWE T. A hybrid model based on CNN-LSTM to detect and forecast harmonics: a case study of an Eskom substation in South Africa[J]. Electric power components and systems, 2023, 51(8): 746-760.

ZHOU J Z, RAJAPAKSE A, GOLE A M. Effects of control systems time delay on the performance of direct harmonics elimination （power network active filters）[C]//IEEE. Canadian Conference on Electrical and Computer Engineering 2004. Niagara Falls: IEEE, 2004: 609-612.

LE ROUX W, VAN WYK J D. The effect of signal measurement and processing delay on the compensation of harmonics by PWM converters[J]. IEEE transactions on industrial electronics, 2000, 47(2): 297-304.

李圣清, 罗晓东, 李永安, 等. 基于前向线性预测理论的混合电力滤波器谐波电流预测方法[J]. 中国电机工程学报, 2011, 31(34): 100-104.

LI Shengqing, LUO Xiaodong, LI Yongan, et al. A harmonic current forecasting method for hybrid active power filters based on the forward linear prediction theory[J]. Proceedings of the CSEE, 2011, 31(34): 100-104.

MARKS J H, GREEN T C. Predictive transient-following control of shunt and series active power filters[J]. IEEE transactions on power electronics, 2002, 17(4): 574-584.

翟晓寒, 丁惜瀛, 姚润宇, 等. 基于径向基神经网络的UPQC并联侧谐波电流预测与补偿[J]. 电测与仪表, 2021, 58(8): 139-145.

ZHAI Xiaohan, DING Xiying, YAO Runyu, et al. Prediction and compensation of harmonic current in parallel side of unified power quality controller based on radial basis neural network[J]. Electrical measurement & instrumentation, 2021, 58(8): 139-145.

ZHOU Min, WANG Bo, GUO Shudong, et al. Multi-objective prediction intervals for wind power forecast based on deep neural networks[J]. Information sciences, 2021, 550: 207-220.

HAN Li, ZHANG Rongchang, WANG Xuesong, et al. Multi-step wind power forecast based on VMD-LSTM[J]. IET renewable power generation, 2019, 13(10): 1690-1700.

刘杰, 金勇杰, 田明. 基于VMD和TCN的多尺度短期电力负荷预测[J]. 电子科技大学学报, 2022, 51(4): 550-557.

LIU Jie, JIN Yongjie, TIAN Ming. Multi-scale short-term load forecasting based on VMD and TCN[J]. Journal of university of electronic science and technology of China, 2022, 51(4): 550-557.

XUE Jiankai, SHEN Bo. A novel swarm intelligence optimization approach: sparrow search algorithm[J]. Systems science & control engineering, 2020, 8(1): 22-34.

DRAGOMIRETSKIY K, ZOSSO D. Variational mode decomposition[J]. IEEE transactions on signal processing, 2014, 62(3): 531-544.

ZHONG Jun, BI Xiaowen, SHU Qin, et al. Partial discharge signal denoising based on singular value decomposition and empirical wavelet transform[J]. IEEE transactions on instrumentation and measurement, 2020, 69(11): 8866-8873.

姜飞, 林政阳, 王文烨, 等. 考虑最小平均包络熵负荷分解的最优Bagging集成超短期多元负荷预测[J/OL]. 中国电机工程学报: 1-17. (2023-04-27) [2023-08-22]. https://doi.org/10.13334/j.0258-8013.pcsee.223470https://doi.org/10.13334/j.0258-8013.pcsee.223470.

JIANG Fei, LIN Zhengyang, WANG Wenye, et al. Optimal bagging ensemble ultra short term multi-energy load forecasting considering least average envelope entropy load decomposition[J/OL]. Proceedings of the CSEE: 1-17. (2023-04-27) [2023-08-22]. https://doi.org/10.13334/j.0258-8013.pcsee.223470https://doi.org/10.13334/j.0258-8013.pcsee.223470.

HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural computation, 1997, 9(8): 1735-1780.

陈海涵, 吴国栋, 李景霞, 等. 基于注意力机制的深度学习推荐研究进展[J]. 计算机工程与科学, 2021, 43(2): 370-380.

CHEN Haihan, WU Guodong, LI Jingxia, et al. Research advances on deep learning recommendation based on attention mechanism[J]. Computer engineering and science, 2021, 43(2): 370-380.

LIAO Wenlong, YANG Zhe, CHEN Xinxin, et al. WindGMMN: scenario forecasting for wind power using generative moment matching networks[J]. IEEE transactions on artificial intelligence, 2022, 3(5): 843-850.

浏览量

下载量

CSCD

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于仿真试验和BOA-VMD的轴箱轴承故障特征提取算法研究

优化参数VMD和MED在列车齿轮箱滚动轴承故障诊断中的应用