摘要
基于风力机和离心水泵的特点,提出了一种风力辅助提水机结构及该机的控制系统。该机是一个耦合的两输入两输出时变系统,系统存在的响应较慢,负荷的随机变化及参数快时变的特性。固定参数PID控制难以适应此系统控制要求,因此,提出一种基于回归神经网络(DRNN)的两输入两输出PID控制器结构,给出了DRNN神经网络参数学习算法和PID控制器参数自整定算法。使该系统能在自然界的风速随机变化的情况下使风力机最大可能利用风能,同时与离心水泵输出功率匹配.计算机仿真结果验证了该控制策略可行性,这为以后进一步研究奠定了基础。
Based on the characteristics of wind turbines and water pumps,a scheme for wind power aided pumping water machines and its control system are proposed.The system is a coupling two-input and two-output and time_variable system.The system has the problem of a slow response and the property of fast parameter variance with the stochastic load.Conventional PID controller which is tuned at typical operating point can hardly work well at different loads,so a two-input and two-output PID controller structure based on diagonal recurrent neural network（DRNN） is proposed.Besides,the learning algorithms of the parameters of DRNN and PID controller are proposed.The control strategy can make the wind turbine work at its best,and match output power of water pump simultaneously.Finally,the validity of the proposed control strategy is revealed via computer simulations and it has laid an excellent foundation for a further research.
出处
《中国农村水利水电》
北大核心
2011年第10期93-95,105,共4页
China Rural Water and Hydropower
基金
四川省科技支撑计划(2011GZ0102)
关键词
风力辅助提水机
耦合
PID控制
回归神经网络(DRNN)
wind power aided pumping water machine
coupling
PID control
d iagonal recurrent neural network（DRNN）
