摘要
针对分布式光伏组件清洁难度大、人工清洁效率低等问题,设计一种新型的基于STM32的光伏组件清洁装置。在装置运行时,由于双电机存在差速,装置会向别的方向偏移,提出以摄像头为主、陀螺仪与编码器结合修正方向,从而达到自主运行的目的;采用神经网络PID算法对多输入系统进行软件编程,并与传统PID算法比较。此设计经过了多次实验验证,得到了大量的数据,对智能光伏清洁、装置规划路径和分布式光伏清洁的发展都有着一定的意义。
A new STM32⁃based photovoltaic module cleaning device is designed to solve the high cleaning difficulty and low manual cleaning efficiency of distributed photovoltaic modules.The device can offset in other directions caused by the differential speed between the two motors while it running,with which the direction correction in combination of the gyro and encoder with the camera as the principal thing is proposed to achieve the goal of autonomous operation.The neural network PID algorithm is applied to program the software of the multi⁃input system,and compared with the traditional PID algorithm.This design has been verified by many experiments,and a large number of data has been obtained,which has certain significance for the development of intelligent photovoltaic cleaning,device planning path and distributed photovoltaic cleaning.
作者
卫芃毅
任一峰
陈昌鑫
郭睿
栾天
WEI Pengyi;REN Yifeng;CHEN Changxin;GUO Rui;LUAN Tian(School of Electrical and Control Engineering,North University of China,Taiyuan 030051,China)
出处
《现代电子技术》
北大核心
2020年第2期41-44,共4页
Modern Electronics Technique
基金
中北大学科技项目立项（20181571）
山西省科技成果转化引导专项项目（201804D131038）
国家自然科学基金资助项目（61774138）
山西省重点研发计划项目(201903D421032)
关键词
光伏清洁
姿态控制
STM32
神经网络PID
方向修正
实验验证
photovoltaic cleaning
attitude control
STM32
neural network PID
direction correction
experimental verification
