一种基于SCADA参数关系的风电机组运行状态识别方法
|
摘要
提出一种基于SCADA参数关系的风电机组运行状态识别方法。首先,从风电机组运行特性出发,深入分析风电机组运行状态SCADA数据输入/输出参数关系;基于时间的滑动窗口模型,采用多项式回归拟合方法,构建风电机组运行状态输入/输出参数关系数学模型;然后,基于风电机组正常运行输入/输出参数关系数学模型,提出描述各个时刻风电机组运行状态异常程度的指标计算公式;对风电机组正常运行阶段的状态指标进行统计分析,获取其分布函数规律;最后,根据小概率事件假设,确定识别风电机组运行状态出现异常的阈值,据此对风电机组运行状态出现异常进行预警。以同风场同型号两台2 MW直驱式风电机组SCADA数据为例进行分析,结果表明:①基于SCADA数据的风电机组运行状态识别方法,可以实现对风电机组运行的异常状态识别和早期预警,该方法的特点是状态识别完全基于正常运行SCADA数据分析而无需异常运行SCADA数据进行挖掘训练和相关物理机制与故障模式方面的先验知识;②基于风电机组SCADA数据的运行状态识别方法,依据风电机组及其部件的运行状态输入/输出参数关系的层次结构,可以获得发生异常状态的相关部件信息,这对风电机组运行状态预警和维护决策具有重要意义。
An operating condition recognition method of wind turbine is proposed based on the parameter relations of SCADA data.Starting from the operation characteristics of wind turbine, the SCADA data input/output parameters relations of wind turbine operating condition are analyzed in-depth. By using a time sliding window model, the polynomial regression fitting method is adopted to build mathematic model of input/output parameters of wind turbine operating status. Then, based on the mathematic model of input/output parameters of wind turbine normal operation, an index formula for describing the abnormality of wind turbine operating status at each moment is put forward. The statistical analysis of the wind turbine state index is carried out in normal operation phase to obtain the distribution function. Finally, according to the hypothesis of small probability event, the threshold value of the state index of the wind turbine abnormal state is determined and the operation condition of wind turbine is forewarned accordingly. Taking the SCADA data of two 2 MW direct-driven wind turbines of the same wind farm as an example, the results show that: ① The proposed wind turbine operating condition recognition method that based on SCADA data can realize the abnormal status identification and early warning of wind turbine operation condition. The feature of this method is that the condition recognition is entirely based on normal operation SCADA data analysis which means mining training and prior knowledge of related physical mechanism and failure mode of abnormal operation SCADA data will not be required. ② Based on the proposed operating condition recognition method, according to the hierarchical structure of the relationship between input/output parameters of wind turbine and its components, the relevant parts information of the abnormal state can be obtained, which is of great significance for operating status early warning and maintenance decision-making of wind turbine.
An operating condition recognition method of wind turbine is proposed based on the parameter relations of SCADA data.Starting from the operation characteristics of wind turbine, the SCADA data input/output parameters relations of wind turbine operating condition are analyzed in-depth. By using a time sliding window model, the polynomial regression fitting method is adopted to build mathematic model of input/output parameters of wind turbine operating status. Then, based on the mathematic model of input/output parameters of wind turbine normal operation, an index formula for describing the abnormality of wind turbine operating status at each moment is put forward. The statistical analysis of the wind turbine state index is carried out in normal operation phase to obtain the distribution function. Finally, according to the hypothesis of small probability event, the threshold value of the state index of the wind turbine abnormal state is determined and the operation condition of wind turbine is forewarned accordingly. Taking the SCADA data of two 2 MW direct-driven wind turbines of the same wind farm as an example, the results show that: ① The proposed wind turbine operating condition recognition method that based on SCADA data can realize the abnormal status identification and early warning of wind turbine operation condition. The feature of this method is that the condition recognition is entirely based on normal operation SCADA data analysis which means mining training and prior knowledge of related physical mechanism and failure mode of abnormal operation SCADA data will not be required. ② Based on the proposed operating condition recognition method, according to the hierarchical structure of the relationship between input/output parameters of wind turbine and its components, the relevant parts information of the abnormal state can be obtained, which is of great significance for operating status early warning and maintenance decision-making of wind turbine.
引文
[1]ADRIAN W,TOBAIS R,JAVIER E,et al.Renewable capacity statistics 2017[R].International Renewable Energy Agency,2017.
[2]YANG Wenxian,TAVNER P J,CRABTREE C J,et al.Wind turbine condition monitoring:Technical and commercial challenges[J].Wind Energy,2014,17(5):673-693.
[3]ZHAI Yongjie,WANG Dongfeng,ZHANG Junying,et al.Research on early fault diagnostic method of wind turbines[J].Telkomnika Indonesian Journal of Electrical Engineering,2013,11(5):2330-2341.
[4]YANG Wenxian,COURT R,JIANG Jiesheng.Wind turbine condition monitoring by the approach of SCADAdata analysis[J].Renewable Energy,2013,53:365-376.
[5]GILL S,STEPHEN B,GALLOWAY S.Wind turbine condition assessment through power curve copula modeling[J].IEEE Transactions on Sustainable Energy,2012,3(1):94-101.
[6]FENG Yanhui,QIU Yingning,CRABTREE C J,et al.Use of SCADA and CMS signals for failure detection and diagnosis of a wind turbine gearbox[C]//European Wind Energy Conference and Exhibition 2011,Sheffield,2011:17-19.
[7]KIM K,PARTHASARATHY G,ULUYOL O,et al.Use of SCADA data for failure detection in wind turbines[C]//Proceedings of 2011 energy sustainability conference and fuel cell conference,Washington DC,2011:2071-2079.
[8]ZAHER A,MCARTHUR S,INFIELD D,et al.Online wind turbine fault detection through automated scada data analysis[J].Wind Energy,2009,12(6):574-593.
[9]LI Jian,LEI Xiao,LI Hui,et al.Normal behavior models for the condition assessment of wind turbine generator systems[J].Electric Power Components and Systems,2014,42(11):1201-1212.
[10]YAN Yonglong,LI Jian,GAO Wenzhong.Condition parameter modeling for anomaly detection in wind turbines[J].Energies,2014,7(5):3104-3120.
[11]KUSIAK A,LI Wenyan.The prediction and diagnosis of wind turbine faults[J].Renewable Energy,2011,36(1):16-23.
[12]ZHANG Zijun,KUSIAK A.Monitoring wind turbine vibration based on SCADA data[J].ASME Journal of Solar Energy Engineering,2012,134(2):021004.
[13]KUSIAK A,VERMA A.Monitoring wind farms with performance curves[J].IEEE Transactions on Sustainable Energy,2013,4(1):192-199.
[14]SCHLECHTINGEN M,SANTOS I F,ACHICHE S.Wind turbine condition monitoring based on SCADA data using normal behavior models.Part 1:System description[J].Applied Soft Computing,2013,13(1):259-270.
[15]DAI Juchuan,TAN Yayi,YANG Wenxian,et al.Investigation of wind resource characteristics in mountain wind farm using multiple-unit SCADA data in Chenzhou:A case study[J].Energy Conversion and Management,2017,148:378-393.
[16]DAI Juchuan,LIU Deshun,WEN Li,et al.Research on power coefficient of wind turbines based on SCADAdata[J].Renewable Energy,2016,86:206-215.
[2]YANG Wenxian,TAVNER P J,CRABTREE C J,et al.Wind turbine condition monitoring:Technical and commercial challenges[J].Wind Energy,2014,17(5):673-693.
[3]ZHAI Yongjie,WANG Dongfeng,ZHANG Junying,et al.Research on early fault diagnostic method of wind turbines[J].Telkomnika Indonesian Journal of Electrical Engineering,2013,11(5):2330-2341.
[4]YANG Wenxian,COURT R,JIANG Jiesheng.Wind turbine condition monitoring by the approach of SCADAdata analysis[J].Renewable Energy,2013,53:365-376.
[5]GILL S,STEPHEN B,GALLOWAY S.Wind turbine condition assessment through power curve copula modeling[J].IEEE Transactions on Sustainable Energy,2012,3(1):94-101.
[6]FENG Yanhui,QIU Yingning,CRABTREE C J,et al.Use of SCADA and CMS signals for failure detection and diagnosis of a wind turbine gearbox[C]//European Wind Energy Conference and Exhibition 2011,Sheffield,2011:17-19.
[7]KIM K,PARTHASARATHY G,ULUYOL O,et al.Use of SCADA data for failure detection in wind turbines[C]//Proceedings of 2011 energy sustainability conference and fuel cell conference,Washington DC,2011:2071-2079.
[8]ZAHER A,MCARTHUR S,INFIELD D,et al.Online wind turbine fault detection through automated scada data analysis[J].Wind Energy,2009,12(6):574-593.
[9]LI Jian,LEI Xiao,LI Hui,et al.Normal behavior models for the condition assessment of wind turbine generator systems[J].Electric Power Components and Systems,2014,42(11):1201-1212.
[10]YAN Yonglong,LI Jian,GAO Wenzhong.Condition parameter modeling for anomaly detection in wind turbines[J].Energies,2014,7(5):3104-3120.
[11]KUSIAK A,LI Wenyan.The prediction and diagnosis of wind turbine faults[J].Renewable Energy,2011,36(1):16-23.
[12]ZHANG Zijun,KUSIAK A.Monitoring wind turbine vibration based on SCADA data[J].ASME Journal of Solar Energy Engineering,2012,134(2):021004.
[13]KUSIAK A,VERMA A.Monitoring wind farms with performance curves[J].IEEE Transactions on Sustainable Energy,2013,4(1):192-199.
[14]SCHLECHTINGEN M,SANTOS I F,ACHICHE S.Wind turbine condition monitoring based on SCADA data using normal behavior models.Part 1:System description[J].Applied Soft Computing,2013,13(1):259-270.
[15]DAI Juchuan,TAN Yayi,YANG Wenxian,et al.Investigation of wind resource characteristics in mountain wind farm using multiple-unit SCADA data in Chenzhou:A case study[J].Energy Conversion and Management,2017,148:378-393.
[16]DAI Juchuan,LIU Deshun,WEN Li,et al.Research on power coefficient of wind turbines based on SCADAdata[J].Renewable Energy,2016,86:206-215.