| 基于忆阻神经网络的光伏面板清扫机器人PID控制器研究 |
| |
| 引用本文： | 张劲,应樱,宋鹏,陈子琪,王绍安,高明煜,董哲康. 基于忆阻神经网络的光伏面板清扫机器人PID控制器研究[J]. 可再生能源, 2021, 39(1): 37-44 |
| |
| 作者姓名： | 张劲 应樱 宋鹏 陈子琪 王绍安 高明煜 董哲康 |
| |
| 作者单位： | 国网浙江省电力有限公司电力科学研究院, 浙江 杭州 310014;浙江大学 电气工程学院, 浙江 杭州310027;杭州电子科技大学 电子信息学院, 浙江 杭州 310018;浙江省装备电子研究重点实验室, 浙江杭州 310018;浙江大学 电气工程学院, 浙江 杭州310027;杭州电子科技大学 电子信息学院, 浙江 杭州 310018;浙江省装备电子研究重点实验室, 浙江杭州 310018 |
| |
| 基金项目： | 浙江省电力有限公司科研项目(5211ds16001w);国家自然科学基金(62001149,61571394);浙江省自然科学基金(LQ21F010009);国家自然科学基金联合基金项目(U1609216)。 |
| |
| 摘 要： | 机器人清扫是光伏面板高效、便捷的清扫方式。文章研究了光伏面板清扫机器人的PID控制器的优化策略。文章基于忆阻器独特的纳米级电路特性,设计了一种新型的突触电路;然后,基于新型的突触电路,构建对应的神经元和神经网络电路,并将忆阻神经网络和光伏面板清扫机器人中的PID控制器相结合,提出一种新型PID控制方法,该方法在一定程度上解决了传统PID控制器体积大、功耗高和相关参数不可调的问题;最后,基于仿真结果验证了文章提出方案的可行性和有效性。
|
| 关 键 词： | 忆阻器 神经网络 光伏面板清扫机器人 PID控制器 |
A study of memristive neural network based photovoltaic panel cleaning robot PID controller |
| |
| Zhang Jing,Ying Ying,Song Peng,Chen Ziqi,Wang Shaoan,Gao Mingyu,Dong Zhekang. A study of memristive neural network based photovoltaic panel cleaning robot PID controller[J]. Renewable Energy(China), 2021, 39(1): 37-44 |
| |
| Authors: | Zhang Jing Ying Ying Song Peng Chen Ziqi Wang Shaoan Gao Mingyu Dong Zhekang |
| |
| Affiliation: | (State Grid Zhejiang Electric Power Research Institute,Hangzhou 310014,China;College of Electrical Engineering,Zhejiang University,Hangzhou 310027,China;Electronic Information College,Hangzhou Dianzi University,Hangzhou 310018,China;Zhejiang Provincial Key Lab of Equipment Electronics,Hangzhou 310018,China) |
| |
| Abstract: | Among all the PV panel cleaning approaches,the robot-based PV panel cleaning method is the most effective and convenient one.In this paper,we mainly focus on the design of PID control strategy in the PV panel cleaning robot system and present a novel memristive neural network based PID controller.Specifically,according to the unique properties of the nanoscale memristor,we design a new synaptic circuit.Meanwhile,the corresponding neuron circuit and the neural network circuit are designed respectively.Furthermore,the memristive neural network is combined with the PID control of the PV panel cleaning robot,which can be used to deal with some inherent issues occurring in the traditional PID controllers,such as big volume,high power consumption,and non-adjustable parameters.Finally,a series of computer experiments are conducted to verify the validity and effectiveness of the entire scheme. |
| |
| Keywords: | memristor neural network photovoltaic panel cleaning robot PID controller |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
