
基于EEMD-GA-BP的水电机组状态趋势预测
陆丹, 肖志怀, 刘东, 胡晓, 邓涛
基于EEMD-GA-BP的水电机组状态趋势预测
A State Tendency Measurement for a Hydro-turbine Generating Unit Based on Ensemble Empirical Mode Decomposition and GA-BP Neural Network Method
电厂、电网的安全稳定与水电机组的运行状态息息相关。机组状态趋势预测弥补了故障诊断作为事后决策的不足,通过预测提前发现故障征兆,可以避免事故发生。本文结合EEMD和神经网络理论,提出了一种水电机组状态趋势预测模型。以国内某两电站的机组振动状态趋势预测为例,首先对机组振动信号进行EEMD分解,其次利用GA-BP预测模型预测各IMF分量运行趋势,最终预测信号是各分量的预测结果累加得到。实验结果表明,该模型能实现机组振动状态趋势的有效预测,相较于其他方法精度更高。
The safety and stability of power plants and power grids are closely related to the operating status of hydropower units. The forecast of the unit status trend makes up for the insufficiency of fault diagnosis as an after-the-fact decision, and the occurrence of accidents can be avoided by detecting fault signs in advance through predictions. Based on EEMD and neural network theory, this paper proposes a state trend prediction model for hydropower units. Taking the vibration state trend prediction of a domestic two power plants as an example, the vibration signal of the unit is firstly decomposed by EEMD, and then the GA-BP prediction model is used to predict the operation trend of each IMF component. The final prediction signal is the accumulation of the prediction results of each component. The experiment results show that the model can effectively predict the vibration state trend of the unit, which is more accurate than other methods.
水电机组振动信号 / 集合经验模态分解 / 趋势预测 / GA-BP神经网络 {{custom_keyword}} /
vibration signal of hydropower unit / ensemble empirical mode decomposition / trend prediction / GA-BP neural network {{custom_keyword}} /
表1 不同EEMD分解参数效果对比Tab.1 Comparison of the effect of different EEMD decomposition parameters |
集合次数/噪声等级 | RMSE/µm | MAPE/% |
---|---|---|
100/0.2 | 0.243 0 | 0.82 |
100/0.3 | 0.248 1 | 0.83 |
100/0.4 | 0.202 9 | 0.67 |
200/0.2 | 0.240 9 | 0.79 |
200/0.3 | 0.229 9 | 0.73 |
200/0.4 | 0.206 7 | 0.69 |
表2 各预测方法指标对比Tab.2 Comparison of indicators of various forecasting methods |
性能指标 | GA-BP | EMD-GA-BP | EEMD-GA-BP |
---|---|---|---|
RMSE/μm | 0.584 7 | 0.506 2 | 0.206 7 |
MAPE/% | 1.94 | 1.44 | 0.69 |
表3 各预测方法指标对比Tab.3 Comparison of indicators of various forecasting methods |
性能指标 | GA-BP | EMD-GA-BP | EEMD-GA-BP |
---|---|---|---|
RMSE/μm | 64.381 8 | 38.638 1 | 27.085 8 |
MAPE/% | 8.26 | 5.09 | 3.30 |
1 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
2 |
王玲花 .水轮发电机组振动及分析[M].黄河水利出版社,2011.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
3 |
王海龙,夏筱筠,孙维堂. 基于EMD与卷积神经网络的滚动轴承故障诊断[J]. 组合机床与自动化加工
{{custom_citation.content}}
{{custom_citation.annotation}}
|
4 |
姚玲峰,宋超,肖世德. 基于EMD及FFT的滚珠丝杠副振动信号分析[J]. 传动技术, 2019,33(1):40-43.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
5 |
付大鹏,翟勇,于青民. 基于EMD和支持向量机的滚动轴承故障诊断研究[J]. 机床与液压,2017,45(11):184-187.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
6 |
蒋文君,胡晓,张培,等. 基于EEMD近似熵的水电机组振动信号特征提取[J]. 水力发电学报, 2020,39(6):18-27.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
7 |
杨皓,黄东胜.AR模型应用于振动信号趋势预测的研究[J].南京工程学院学报(自然科学版),2007(2):45-49.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
8 |
周建民,张臣臣,王发令,等.基于ARMA的滚动轴承振动数据预测[J].华东交通大学学报,2018,35(5):99-103.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
9 |
付文龙,周建中,张勇传,等.基于OVMD与SVR的水电机组振动趋势预测[J].振动与冲击,2016,35(8):36-40.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
10 |
张飞,潘罗平.基于人工神经网络的水轮发电机组振动预测研究[J].人民长江,2011,42(13):48-50,106.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
11 |
付文龙. 水电机组振动信号分析与智能故障诊断方法研究[D].武汉:华中科技大学,2016.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
12 |
杨晓红,杨晓静,朱霄珣.基于小波变换的支持向量回归机振动数据短期预测[J].汽轮机技术,2011,53(1):77-80.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
13 |
戈英杰. 混合风速预测模型的建立与评估[D].东北财经大学,2018.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
14 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
15 |
刘春艳,凌建春,寇林元,等.GA-BP神经网络与BP神经网络性能比较[J].中国卫生统计,2013,30(2):173-176,181.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
16 |
彭利鸿,宋媛,刘冬,等.基于GA-BP神经网络的水轮机非线性建模方法研究[J].中国农村水利水电,2017(4):184-188,193.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
17 |
吕金虎,陆君安,陈士华. 混沌时间序列分析及其应用[M]. 武汉:武汉大学出版社,2002.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
18 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
19 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
20 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
21 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
22 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
/
〈 |
|
〉 |