军事知识服务平台
SSA-PSO-DBN的风电机组故障预警研究
Research on early warning of wind turbine fault based on SSA-PSO-DBN
摘要 [目的]针对风电机组工作环境复杂多变带来运行挑战的问题,为实现风电机组故障的早期预警以保障机组运行效率,提出一种基于麻雀搜索算法联合粒子群算法优化深度置信网络(SSA-PSO-DBN)的风电机组故障预警算法.[方法]首先使用 KNN-箱型图对采集的 SCADA 数据进行预处理并划分数据集;然后将预处理后的训练数据输入 SSA-PSO-DBN 模型,利用联合优化算法对 DBN的超参数进行寻优并完成模型训练;最后使用最优参数训练好的模型对测试数据进行验证,以评估模型的有效性.[结果]最终实验结果表明,所提算法能够有效实现对风电机组故障的早期预警.[结论]所提方法为风电机组在复杂工况下的安全稳定运行提供了可靠的技术保障,具有一定的工程应用前景.
Abstract:
[Objective]The working environment of wind turbines is complex and variable,posing many challenges to their normal and stable operation.To achieve early fault warning of wind turbines and thereby ensure operational efficiency,this paper proposes a fault warning algorithm based on a deep belief network optimized by the sparrow search algorithm combined with particle swarm optimization(SSA-PSO-DBN).[Methods]First,the collected SCADA data are preprocessed using the KNN-box plot method,and the dataset is divided accordingly.Then,the preprocessed training data are fed into the SSA-PSO-DBN model for hyperparameter optimization and model training.Finally,the model trained with optimal parameters is used to process test data to verify the model's effectiveness.[Results]Experimental results demonstrate that the proposed algorithm can effectively achieve early warning of wind turbine faults.[Conclusion]The proposed method provides a reliable technical safeguard for the safe and stable operation of wind turbines under complex working conditions,demonstrating promising engineering application prospects.
刊 名 中国测试 ISTICPKU
年,卷(期): 2026, 52(9)
分类号: TM315 TP277 TP181 TB9
机标分类号: TP391 TM715 TN911.22
在线出版日期: 2026年10月14日
基金项目: CXY2023009:河北省高等学校科学技术研究项目2394G001:保定市科技计划项目,