中国电机工程学报
中國電機工程學報
중국전궤공정학보
ZHONGGUO DIANJI GONGCHENG XUEBAO
2014年
z1期
160-165
,共6页
尹诗%余忠源%孟凯峰%李闯%王其乐
尹詩%餘忠源%孟凱峰%李闖%王其樂
윤시%여충원%맹개봉%리틈%왕기악
主故障%非线性状态状态估计%数据挖掘%风电机组
主故障%非線性狀態狀態估計%數據挖掘%風電機組
주고장%비선성상태상태고계%수거알굴%풍전궤조
main fault%nonlinear state estimation%data mining%wind turbine
变桨系统是风力发电机组中控制算法复杂,设备故障率较高的子系统。当风机出现故障停机时,数据采集与监视控制(supervisory control and data acquisition,SCADA)系统在故障出现会显示出具体的故障信息,该具体故障信息往往包含多个,给风机迅速定位故障类别、确认检修方式、及时有效地恢复生产造成了不利影响。因此,基于变桨运行数据的风力发电机组主故障识别对于提升机组可利用率及机组发电性能有着积极重要的意义。文中采用非线性状态估计技术作为数据挖掘方法,在某风电场机组 SCADA 数据基础上,分析机组变桨系统运行趋势及故障类型建立机组变桨控制系统主故障模型,并对该模型进行验证。研究结果表明,基于非线性状态估计的风电机组变桨控制系统主故障识别能够在多个故障信息中识别出主故障,次故障等,从而指导风场检修人员确定检修顺序,并为后续开展风机性能分析及评价提供了新的思路。
變槳繫統是風力髮電機組中控製算法複雜,設備故障率較高的子繫統。噹風機齣現故障停機時,數據採集與鑑視控製(supervisory control and data acquisition,SCADA)繫統在故障齣現會顯示齣具體的故障信息,該具體故障信息往往包含多箇,給風機迅速定位故障類彆、確認檢脩方式、及時有效地恢複生產造成瞭不利影響。因此,基于變槳運行數據的風力髮電機組主故障識彆對于提升機組可利用率及機組髮電性能有著積極重要的意義。文中採用非線性狀態估計技術作為數據挖掘方法,在某風電場機組 SCADA 數據基礎上,分析機組變槳繫統運行趨勢及故障類型建立機組變槳控製繫統主故障模型,併對該模型進行驗證。研究結果錶明,基于非線性狀態估計的風電機組變槳控製繫統主故障識彆能夠在多箇故障信息中識彆齣主故障,次故障等,從而指導風場檢脩人員確定檢脩順序,併為後續開展風機性能分析及評價提供瞭新的思路。
변장계통시풍력발전궤조중공제산법복잡,설비고장솔교고적자계통。당풍궤출현고장정궤시,수거채집여감시공제(supervisory control and data acquisition,SCADA)계통재고장출현회현시출구체적고장신식,해구체고장신식왕왕포함다개,급풍궤신속정위고장유별、학인검수방식、급시유효지회복생산조성료불리영향。인차,기우변장운행수거적풍력발전궤조주고장식별대우제승궤조가이용솔급궤조발전성능유착적겁중요적의의。문중채용비선성상태고계기술작위수거알굴방법,재모풍전장궤조 SCADA 수거기출상,분석궤조변장계통운행추세급고장류형건립궤조변장공제계통주고장모형,병대해모형진행험증。연구결과표명,기우비선성상태고계적풍전궤조변장공제계통주고장식별능구재다개고장신식중식별출주고장,차고장등,종이지도풍장검수인원학정검수순서,병위후속개전풍궤성능분석급평개제공료신적사로。
The pitch system of a wind turbine is a subsystem with complex control algorithm and high rate of equipment failure. When the wind machine suffers downtime, the supervisory control and data acquisition system will display the fault information, which contains multiple contents, causing an adverse effect on the quick fault category location of fans, selection of maintenance method timely and effective recovery. Therefore, the main fault identification based on pitch operating data of wind turbine has a positive significance to the increase of utilization rate and power generation. The nonlinear state estimation technique (NSET) ,as a data mining method, can be used to analyze the trend and fault type to confirm the pitch control model system for the main fault based on the SCADA data in a wind turbine. The results show that using the NSET estimating fault of pitch control system can identify the main fault of the wind turbine in multiple fault information, fault and so on, so as to guide the wind farm maintainer to determine the maintenance sequence, and provide a new idea for further analysis and evaluation of performance of the fan.