| 41 | 0 | 4 |
| 下载次数 | 被引频次 | 阅读次数 |
受水激振动等多种因素影响，闸门的启闭运行呈现多场耦合、复杂非线性动力学特征，造成设备安全性态辨识困难。闸门启闭运行的测试数据表明，人工神经网络算法可辨识水激振动病害特征和准确预测水激振动病害趋势。为此，通过BP神经网络和GA-BP神经网络，构建闸门水激振动病害辨识和预测模型，对卷筒振动有效值进行辨识与预测，并通过相对误差(RRE)、平均绝对误差率(MMAPE)、均方根误差(RRMSE)等指标评价模型辨识性能。结果表明，相对于BP神经网络的辨识模型，GA-BP神经网络模型的相对误差减少了20.77%,平均绝对误差率减少了4.74%,均方根误差减少了6.27%,GA-BP闸门水激振动病害辨识技术更好贴合实测样本集，且随预测时间增大表现更好稳定性，可为工程减害运行和防范重大险病提供关键技术支撑。
Abstract：Affected by hydrodynamic excitation and other factors, the opening-closing operation of hydraulic gates exhibits multi-field coupling effects and complex nonlinear dynamic characteristics, leading to difficulties in identifying equipment safety states. Test data of gate operation demonstrate that artificial neural network algorithms can identify hydrodynamic excitation disease features and accurately predict its development trends. To address this, BP and GA-BP neural networks were employed to construct identification and prediction models for hydrodynamic excitation disease. These models were applied to identify and forecast the effective values of reel vibration, with model performance evaluated using metrics including Relative Error(RRE), Mean Absolute Percentage Error(MMAPE), and Root Mean Square Error(RRMSE). Compared to the BP model, the results indicate that the GA-BP model achieves reductions of 20.77% in RRE, 4.74% in MMAPE, and 6.27% in RRMSE, demonstrating superior fitting to measured samples and enhanced stability with extended prediction durations, thus providing critical technical support for engineering risk mitigation and hazard prevention.
[1] 马斌，郭乙良.水工闸门振动研究现状及发展趋势[J].水利水运工程学报，2019(2):55-64.
[2] XU C,LIU J L,ZHAO C L,et al.Dynamic failures of water controlling radial gates of hydro-power plants:Advancements and future perspectives[J].Engineering failure analysis,2023,148:107168.
[3] TIELEMAN O C,HOFLAND B,TSOUVALAS A,et al.A fluid-structure interaction model for assessing the safety of flood gate vibrations due to wave impacts[J].Coastal engineering,2021,170:104007.
[4] YE F,HU Z J.Robust load frequency control of interconnected power systems with back propagation neural network-proportional-integral-derivative-controlled wind power integration[J].Sustainability,2024,16(18):8062.
[5] CHEN Y.Application of the BP neural network algorithm to the ultimate bearing capacity test of building structures[J].Journal of computational methods in sciences and engineering,2024,24(3):1759-1770.
[6] WAN X Y,TANG Y J,LIU S.Application and research of mine safety management strategy based on the BP network model[J].International journal of high speed electronics and systems,2024:2540038.
[7] 周良，王华伟，许珊珊，等.基于GA优化BP算法的滑油状态监测[J].航空发动机，2022,48(5):137-142.
[8] WANG Y S,XIN B,LI J T,et al.Surface roughness prediction model and surface topography analysis of 2.5D-Cf/SiC in two-dimensional ultrasonic assisted grinding based on GA-BP neural network[J].Tribology international,2025,201:110272.
[9] KIRANKUMAR K R,KUMAR G N,KAMATH N,et al.Experimental investigation and optimization of performance,emission,and vibro-acoustic parameters of SI engine fueled with n-propanol and gasoline blends using ANN-GA coupled with NSGA3-modified TOPSIS hybrid approach[J].Energy,2024,306:132521.
[10] LI Y P,JIA M,CHANG Y C,et al.Parametric study and optimization of a RCCI (reactivity controlled compression ignition) engine fueled with methanol and diesel[J].Energy,2014,65:319-332.
基本信息:
DOI：10.20040/j.cnki.1000-7709.2025.20242180
中图分类号:TB53;TP18;TV698.18
引用信息:
[1]高建伟,朱佳,黎军杰等.闸门水激振动病害辨识技术研究[J].水电能源科学,2025,43(09):110-113.DOI:10.20040/j.cnki.1000-7709.2025.20242180.
基金信息:
国网新源控股有限公司科技项目(SGXYKJ-2023-0155)