摘要
为预防风机齿轮箱严重故障的发生,提出了一种结合正态云模型和趋势状态分析法的齿轮箱状态评估模型.首先,基于数据采集与监控(SCADA)系统的历史数据,采用支持向量机方法建立齿轮箱运行温度预测模型,对齿轮箱不同状态下的趋势预测特征进行分析,求取正常和异常状态时预测值的相对误差序列.然后,采用改进无确定度逆向正态云发生器,利用所求取的相对误差序列提取正常云和异常云的数字特征,构建齿轮箱状态评估云模型.该模型能够根据风机SCADA系统的实测数据,求取齿轮箱当前状态对正常云和异常云的贴近度,并采用最大贴近度原则确定齿轮箱状态.最后,利用辽宁某风机齿轮箱的实测数据对所提模型进行验证.结果表明:该模型能够对齿轮箱的早期缺陷及时预警,达到实时监测的目的.
To prevent wind turbine gearbox severe faults occurrence, combined normal cloud model and trend condition analysis, this paper presents a novel condition assessment model. Firstly, based on supervisory control and data acquisi- tion （SCADA） historical data, the forecasting model of gearbox operating temperature is established by adapting SVM, and the relative errors sequence of the forecasting values under normal and abnormal conditions are calculated after analy- zing characteristics of trend forecasting. Then, inputting the relative errors sequence into improved backward normal cloud generator with the non-certainty degree, the digital features of normal and abnormal cloud model are obtained. Fur- thermore, the gearbox condition assessment cloud model is given, which is based on online SCADA data of wind turbine to calculate the closeness degree of normal and abnormal cloud model for gearbox current condition, and use the principle of maximum closeness degree to determine the gearbox condition. Finally, the proposed model is verified by online data of a wind turbine gearbox in Liaoning province, the results show that this model is capable of alarming early defects timely of a gearbox, achieving the aim of online condition assessment.
出处
《华侨大学学报（自然科学版）》
CAS
北大核心
2016年第1期32-37,共6页
Journal of Huaqiao University(Natural Science)
基金
国家自然科学基金资助项目(51177039)
福建省自然科学基金资助项目(2012J01223)
福建省厦门市重大科技创新平台项目(3502Z20111008)
