基于TrueTime的BP神经网络控制系统的仿真平台构建  

Construction of simulation platform for BP neural network control system based on TrueTime

作　　者：高月芳[1] 叶建锋[1] 张薇 Gao Yuefang;Ye Jianfeng;Zhang Wei(School of Computer Science,Shenzhen Institute of Information Technology,Shenzhen,Guangdong,China 518172;Financial Department,Shenzhen Institute of Information Technology,Shenzhen,Guangdong,China 518172)

机构地区：[1]深圳信息职业技术学院计算机学院,广东深圳518172 [2]深圳信息职业技术学院计财处,广东深圳518172

出　　处：《深圳信息职业技术学院学报》2018年第5期15-20,共6页Journal of Shenzhen Institute of Information Technology

基　　金：广东省自然科学基金资助项目(2015A030313592;2018A0303130055);深圳市科技计划资助项目(JCYJ20160307101532282;JCY20160307100530069)

摘　　要：控制系统中最广泛使用的是PID控制算法,但传统PID控制算法不能有效处理系统中的干扰、丢包和时延等问题,为此本文引入了基于BP神经网络的PID控制算法。该算法具有学习功能,可以自动调整PID参数,减小干扰节点引起的资源争夺与系统延时等方面的影响,优化系统指标。另外,为了模拟真实的网络环境,对系统的性能和设计方案进行评估,本文利用TrueTime工具箱搭建了网络控制系统仿真平台。仿真结果表明,BP神经网络算法的性能明显优越于传统PID控制算法。PID control algorithm is the most widely used in the control system,but the traditional PID control algorithm cannot effectively deal with the interference,packet loss and delay in the system.Therefore,a PID control algorithm based on BP neural network is introduced in this paper.The algorithm has a learning function,which can automatically adjust PID parameters,reduce the impact of resource contention caused by interfering nodes and system delay,and optimize system indicators.In addition,in order to simulate the real network environment and evaluate the performance and design of the system,this paper uses the TrueTime toolbox to build a network control system simulation platform.The simulation results show that the performance of BP neural network algorithm is superior to the traditional PID control algorithm.

关 键 词：网络控制系统 TRUETIME 神经网络 仿真平台 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]