基于bp神经网络的工业缝纫机控制系统的研究-电机与电器专业论文.docxVIP

浙江大学硕士学位论文 浙江大学硕士学位论文 摘要 摘要 中国是纺织大国，同时也是世界上工业缝纫机的主要生产基地。现在先进的 工业缝纫机控制系统以DSP为控制核心，集光、电、磁等检测和控制技术于一 体，逐渐取代了传统的离合式异步电机缝纫机。本文根据平缝机启动快，调速范 围广、停转快以及定位精确等要求，设计和研制了实验样机，经过多次改版调试， 已经可以进行大批量的产业化生产。本论文主要完成了以下几方面工作： (1)考虑到工业平缝机控制系统是一个滞后、时变、非线性的控制系统，在 分析无刷直流电机结构和数学模型的基础上，提出了BP神经网络整定参数的PID 控制算法，并进行了MATLAB仿真和实验。 (2)完成了控制系统的软硬件设计。硬件包括DSP主控电路及周边电路，电 源、驱动电路、电机信号检测电路以及脚踏板电路等。然后针对工业平缝机的多 项功能，在程序的设计中对各项子功能进行了模块化设计。 (3)测试了控制系统的调速性能和停针位置，将采用神经网络整定参数的 PID控制器与传统的PID控制器调速效果进行了对比。测试了所研制的平缝机控 制器的综合性能，满足了实际的应用需求。 关键词：工业平缝机；无刷直流电机；BP神经网络；．MATLAB仿真 浙江大学硕士学位论文 浙江大学硕士学位论文 Abstract Abstract China is a country witll prosperous sew industry,and also the main a manufacture base of industrial sewing machine in the world．Nowadays the control of industry sewing machine is based on DSP．and this system is usually integrated witll optical，electrical and magnetic as the essential monitoring and controlling technology． In order to meet the sewing machine performance requirements，which is fast starting and stopping．、析de range of speed changing and stopping at a precise position． Having experienced revision and commissioning many times，it has basically meet the demand of massive industrial production．The main work of this paper is as follows． (1)Based on the introduction of structure and mathematical model of BLDCM， the paper comes up winl speed control strategy accordingly．Because industry sewing machine system is hysteretic，time-variant and nonlinear，BP neural algorithm is added to the conventional PID control algorithm，whose self-learning and self-adjusting capabilities help realize the thoughts that parameters of a PID controller can regulate themselves·as the system is varying，and this thought will also be emulated by MATLAB and used in experimental coding． (2)The hardware and software for Industrial Sewing Machine is designed， including control circuit of DSP and peripheral circuit,power cir