电力系统保护与控制
電力繫統保護與控製
전력계통보호여공제
Power System Protection and Control
2015年
17期
29-37
,共9页
柔性直流输电%PID%自适应神经元模糊推理系统%三重合作粒子群算法%直接功率控制
柔性直流輸電%PID%自適應神經元模糊推理繫統%三重閤作粒子群算法%直接功率控製
유성직류수전%PID%자괄응신경원모호추리계통%삼중합작입자군산법%직접공솔공제
VSC-HVDC%PID%ANFIS%TCPSO%DPC
针对柔性直流输电系统(Voltage Source Converter based High-Voltage Direct-Current, VSC-HVDC)双闭环控制中PI控制存在参数整定困难及控制器数量过多等问题,提出一种具有PID功能的自适应神经元模糊推理系统(Adaptive Neuro-Fuzzy Inference System with PID function, PID-ANFIS)控制器用于该系统控制。其中,PID-ANFIS控制器兼有神经网络控制、二阶模糊控制及PID功能;同时提出的基于三重合作粒子群算法(Treble Cooperative Particle Swarm Optimization, TCPSO)用于优化该控制器中神经网络参数。TCPSO采用由降维合作、分组合作与记忆合作组成的三重合作策略,极大程度上提升了神经网络参数优化的精度。深入研究了TCPSO优化PID-ANFIS控制器参数的步骤。基于TCPSO优化的PID-ANFIS控制器能够实现VSC-HVDC系统的直接功率控制效果。仿真结果表明该控制器具有控制速度快、超调量小、抗干扰能力强等优点,是VSC-HVDC控制系统的一个可行方案。
針對柔性直流輸電繫統(Voltage Source Converter based High-Voltage Direct-Current, VSC-HVDC)雙閉環控製中PI控製存在參數整定睏難及控製器數量過多等問題,提齣一種具有PID功能的自適應神經元模糊推理繫統(Adaptive Neuro-Fuzzy Inference System with PID function, PID-ANFIS)控製器用于該繫統控製。其中,PID-ANFIS控製器兼有神經網絡控製、二階模糊控製及PID功能;同時提齣的基于三重閤作粒子群算法(Treble Cooperative Particle Swarm Optimization, TCPSO)用于優化該控製器中神經網絡參數。TCPSO採用由降維閤作、分組閤作與記憶閤作組成的三重閤作策略,極大程度上提升瞭神經網絡參數優化的精度。深入研究瞭TCPSO優化PID-ANFIS控製器參數的步驟。基于TCPSO優化的PID-ANFIS控製器能夠實現VSC-HVDC繫統的直接功率控製效果。倣真結果錶明該控製器具有控製速度快、超調量小、抗榦擾能力彊等優點,是VSC-HVDC控製繫統的一箇可行方案。
침대유성직류수전계통(Voltage Source Converter based High-Voltage Direct-Current, VSC-HVDC)쌍폐배공제중PI공제존재삼수정정곤난급공제기수량과다등문제,제출일충구유PID공능적자괄응신경원모호추리계통(Adaptive Neuro-Fuzzy Inference System with PID function, PID-ANFIS)공제기용우해계통공제。기중,PID-ANFIS공제기겸유신경망락공제、이계모호공제급PID공능;동시제출적기우삼중합작입자군산법(Treble Cooperative Particle Swarm Optimization, TCPSO)용우우화해공제기중신경망락삼수。TCPSO채용유강유합작、분조합작여기억합작조성적삼중합작책략,겁대정도상제승료신경망락삼수우화적정도。심입연구료TCPSO우화PID-ANFIS공제기삼수적보취。기우TCPSO우화적PID-ANFIS공제기능구실현VSC-HVDC계통적직접공솔공제효과。방진결과표명해공제기구유공제속도쾌、초조량소、항간우능력강등우점,시VSC-HVDC공제계통적일개가행방안。
Due to the PI control system for VSC-HVDC has problems of parameters difficult to set, too many control users, and so on, a novel controller composed of PID function and multiple-output ANFIS (PID-ANFIS) is presented, which is made up of neural network, two order fuzzy control and PID control. A treble cooperative PSO (TCPSO) is also presented to optimize PID-ANFIS controller’s neural parameters. TCPSO is forged by harmonizing the grouping cooperation, the dimension-reduced cooperation and memory cooperation, which is able to improve the precision of optimizing neural networks. This paper provides the process of PID-ANFIS parameters training by TCPSO. Then, the TCPSO based PID-ANFIS controller performs the function of direct power control. The simulation results show that the controller presented has significant advantages of faster speed, smaller overshoot and better robustness by comparing to PI and it is a viable choice for VSC-HVDC control system.