期刊文献+
共找到2篇文章
< 1 >
每页显示 20 50 100
Auto-Tuning Parameters of Fractional PID Controller Design for Air-Conditioning Fan Coil Unit 被引量:2
1
作者 Ll Shaoyong WANG Duo +2 位作者 HAN Xilian CHENG Kang ZHChunrun 《Journal of Shanghai Jiaotong university(Science)》 EI 2021年第2期186-192,共7页
The traditional integer order PID controller manipulates the air-conditioning fan coil unit(FCU)that offers cooliug and heatins loads to each air-conditioning room in summer and winter,respectivelv.In order to maintai... The traditional integer order PID controller manipulates the air-conditioning fan coil unit(FCU)that offers cooliug and heatins loads to each air-conditioning room in summer and winter,respectivelv.In order to maintain a steady indoor temperature in summer and winter,the control quality cannot meet the related requirements of air-conditioning automation,such as large overshoot,large steady state error.long regulating time,etc.In view of these factors,this paper develops a fractional order PID controller to deal with such problem associated with FCU.Then,by varving mutation factor and crossover rate of basic differential evolution algorithmadaptivelv,a modified differential evolution algorithm(MDEA)is designed to tune the satisfactory values of five parameters of indoor temperature fractional order PID controller.This fractional order PID coutrol system is configured and the corresponding mumerical simulation is conducted by means of MATLAB software.The results indicate that the proposed fractional order PID control svstem and MDEA are reliable and the related control performance indexes meet with the related requirements of comfortable air-conditioning design and control criteria. 展开更多
关键词 air-conditioning fan coil unit(FCU) fractional order PID control modified differential evolution algorithm(MDEA) auto-tuning parameters of controller
原文传递
一种基于PF-RBF的ANNPID参数自整定方法
2
作者 苏岭东 赵成 马祥林 《通信技术》 2021年第3期658-663,共6页
针对实际控制过程中控制对象难以建立精确模型和由于非线性非高斯噪声干扰导致控制效果难以达到预期的问题,提出了一种基于粒子滤波和RBF神经网络辨识(RBF Neural Network and Particle Filter Algrothm,PF-RBF)的单神经元PID参数自整... 针对实际控制过程中控制对象难以建立精确模型和由于非线性非高斯噪声干扰导致控制效果难以达到预期的问题,提出了一种基于粒子滤波和RBF神经网络辨识(RBF Neural Network and Particle Filter Algrothm,PF-RBF)的单神经元PID参数自整定方法。通过PF和RBF系统辨识得到精确的系统Jacobian信息,解决神经网络PID控制由于Jacobian信息未知导致的近似计算不精确问题。仿真实例表明,相对于无PF滤波的RBF辨识控制系统,该方法能改善控制系统的性能指标和抗干扰能力,对实际控制过程具有一定的指导意义。 展开更多
关键词 粒子滤波 RBF神经网络辨识 信息 PID参数自整定
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部