摘要
在传统直接转矩控制(DTC)的基础上引入模糊逻辑和神经网络,形成一种新型的模糊神经网络DTC策略,并将其应用到风能模拟系统中。利用神经网络建立DTC系统的定子磁链观测器,并引入多群体协同进化粒子群(MCPSO)算法对其进行网络训练。实验结果表明,与传统DTC相比,模糊神经网络DTC更能减小低速运行时极易出现的转矩和磁链脉动,在稳定性、动态响应、鲁棒性和实时性方面都有明显的改善。
On the basis of conventional direct torque control（DTC）,the fuzzy logic and the neural network are intro-duced to form a new fuzzy neural network strategy which is applied to wind simulation system.With neural network,stator flux observer of the DTC system is established and the multi-swarm cooperative particle swarm optimizer（MCP-SO） algorithm is introduced to do the network training.The Experimental results show that compared with the con-ventional DTC,it＇s much easier for fuzzy neural network DTC to reduce the torque and flux linkage ripple when run-ning at a low speed.The stability,dynamic response,robustness and real-time have got obvious improvements.
出处
《电力电子技术》
CSCD
北大核心
2011年第9期49-51,共3页
Power Electronics
关键词
风力机模拟
模糊逻辑
神经网络
直接转矩控制
wind turbine imitation
fuzzy logic
neural network
direct torque control
作者简介
王新新（1987-），女，江苏盐城人，硕士，研究方向为新能源发电与控制技术。
