自适应蝙蝠算法优化PF的风力机桨距系统故障诊断方法
|
摘要
针对粒子滤波(PF)在变桨距系统故障诊断中存在的样本贫化现象导致故障诊断精度低的问题,提出一种蝙蝠算法自适应优化粒子滤波的故障诊断方法。通过改进的蝙蝠算法优化粒子滤波的采样过程,并结合最新的观测值定义粒子适应度函数,引导粒子整体向高似然区域移动;同时引入一个动态自适应惯性权重来设计新的粒子全局搜索位置更新机制,自适应调整粒子的全局搜索与局部搜索能力的有效协调,改善粒子贫化及陷入局部极值的问题,以期提高粒子群对故障状态的估计性能。通过对风力机桨距系统执行器和传感器故障诊断的仿真分析,表明该方法可有效提高故障诊断的准确性。
Aiming at the problem of low accuracy for fault diagnosis due to the sample impoverishment phenomenon exist in traditional particle filter( PF) in the fault diagnosis of the variable pitch system,a fault diagnosis method based on adaptive bat algorithm optimizing particle filter is proposed. Combined the sampling process of optimizing particle filter by the improved bat algorithm with the latest observation to define the fitness function to guide the particle to move to the high likelihood region. At the same time,the dynamic adaptive inertia weight design is introduced to design a new global search position update mechanism,the balance between global search and local search ability is adjusted adaptively. Alleviating the phenomenon of particle impoverishment and avoiding local extremal,and the estimation performance of particle swarm to fault state is improved. Through the simulation analysis of actuator and sensor fault diagnosis of wind turbine pitch system,it is shown that this method can effectively improve the accuracy of fault diagnosis.
Aiming at the problem of low accuracy for fault diagnosis due to the sample impoverishment phenomenon exist in traditional particle filter( PF) in the fault diagnosis of the variable pitch system,a fault diagnosis method based on adaptive bat algorithm optimizing particle filter is proposed. Combined the sampling process of optimizing particle filter by the improved bat algorithm with the latest observation to define the fitness function to guide the particle to move to the high likelihood region. At the same time,the dynamic adaptive inertia weight design is introduced to design a new global search position update mechanism,the balance between global search and local search ability is adjusted adaptively. Alleviating the phenomenon of particle impoverishment and avoiding local extremal,and the estimation performance of particle swarm to fault state is improved. Through the simulation analysis of actuator and sensor fault diagnosis of wind turbine pitch system,it is shown that this method can effectively improve the accuracy of fault diagnosis.
引文
[1]Perveen R,Kishor N,Mohanty S R.Off-shore wind farm development:Present status and challenges[J].Renewable and Sustainable Energy Reviews,2014,29(7):780-792.
[2]Odgaard P F,Johnson K E.Wind turbine fault detection and fault tolerant control-An enhanced benchmark challenge[C]//American Control Conference(ACC),17-19 June2013,Washington,DC,USA.IEEE,2013.
[3]Simani S,Farsoni S,Castaldi P.Fault Diagnosis of a Wind Turbine Benchmark via Identified Fuzzy Models[J].IEEETransactions on Industrial Electronics,2015,62(6):3775-3782.
[4]Wu D,Song J,Shen Y.Variable forgetting factor identification algorithm for fault diagnosis of wind turbines[C]//Chinese Control and Decision Conference.2016:1895-1900.
[5]赵洪山,连莎莎,邵玲.基于模型的风电机组变桨距系统故障检测[J].电网技术,2015(2):440-444.
[6]Wu D,Zhai Y,Gao W,et al.Multi-innovation observer based fault detection for pitch system of wind turbines[C]//IEEEInternational Conference on Cyber Technology in Automation,Control,and Intelligent Systems.IEEE,2015:677-680.
[7]Esbensen T,Sloth C.Fault diagnosis and fault-tolerant control of wind turbines[D].Aalborg University,2009.
[8]张玲霞,刘志仓,王辉,等.非线性系统故障诊的粒子滤波方法[J].电子学报,2015,43(3):615-619.
[9]郭健彬,纪丁菲,王鑫,等.混杂系统粒子滤波混合状态估计及故障诊断算法[J].系统工程与电子技术,2015,37(8):1936-1942.
[10]黄金泉,陈煜,周浩文,等.基于QPSO粒子滤波的航空发动机突变故障诊断[J].南京航空航天大学学报,2014,46(4):494-500.
[11]Liu X,Gao Z.Takagi-Sugeno fuzzy modelling and robust fault reconstruction for wind turbine systems[C]//IEEE,International Conference on Industrial Informatics.IEEE,2017:492-495.
[12]朱志宇.粒子滤波算法及其应用[M].科学出版社,2010.
[13]李煜,马良.新型全局优化蝙蝠算法[J].计算机科学,2013,40(9):225-229.
[14]Gandomi A H,Yang X S,Alavi A H,et al.Bat algorithm for constrained optimization tasks[J].Neural Computing&Applications,2013,22(6):1239-1255.
[15]陈志敏,田梦楚,吴盘龙,等.基于蝙蝠算法的粒子滤波法研究[J].物理学报,2017,66(5):41-50.
[16]朱奇光,肖亚昆,陈卫东,等.基于萤火虫算法改进移动机器人定位方法研究[J].仪器仪表学报,2016,37(2):323-329.
[2]Odgaard P F,Johnson K E.Wind turbine fault detection and fault tolerant control-An enhanced benchmark challenge[C]//American Control Conference(ACC),17-19 June2013,Washington,DC,USA.IEEE,2013.
[3]Simani S,Farsoni S,Castaldi P.Fault Diagnosis of a Wind Turbine Benchmark via Identified Fuzzy Models[J].IEEETransactions on Industrial Electronics,2015,62(6):3775-3782.
[4]Wu D,Song J,Shen Y.Variable forgetting factor identification algorithm for fault diagnosis of wind turbines[C]//Chinese Control and Decision Conference.2016:1895-1900.
[5]赵洪山,连莎莎,邵玲.基于模型的风电机组变桨距系统故障检测[J].电网技术,2015(2):440-444.
[6]Wu D,Zhai Y,Gao W,et al.Multi-innovation observer based fault detection for pitch system of wind turbines[C]//IEEEInternational Conference on Cyber Technology in Automation,Control,and Intelligent Systems.IEEE,2015:677-680.
[7]Esbensen T,Sloth C.Fault diagnosis and fault-tolerant control of wind turbines[D].Aalborg University,2009.
[8]张玲霞,刘志仓,王辉,等.非线性系统故障诊的粒子滤波方法[J].电子学报,2015,43(3):615-619.
[9]郭健彬,纪丁菲,王鑫,等.混杂系统粒子滤波混合状态估计及故障诊断算法[J].系统工程与电子技术,2015,37(8):1936-1942.
[10]黄金泉,陈煜,周浩文,等.基于QPSO粒子滤波的航空发动机突变故障诊断[J].南京航空航天大学学报,2014,46(4):494-500.
[11]Liu X,Gao Z.Takagi-Sugeno fuzzy modelling and robust fault reconstruction for wind turbine systems[C]//IEEE,International Conference on Industrial Informatics.IEEE,2017:492-495.
[12]朱志宇.粒子滤波算法及其应用[M].科学出版社,2010.
[13]李煜,马良.新型全局优化蝙蝠算法[J].计算机科学,2013,40(9):225-229.
[14]Gandomi A H,Yang X S,Alavi A H,et al.Bat algorithm for constrained optimization tasks[J].Neural Computing&Applications,2013,22(6):1239-1255.
[15]陈志敏,田梦楚,吴盘龙,等.基于蝙蝠算法的粒子滤波法研究[J].物理学报,2017,66(5):41-50.
[16]朱奇光,肖亚昆,陈卫东,等.基于萤火虫算法改进移动机器人定位方法研究[J].仪器仪表学报,2016,37(2):323-329.