基于单神经元自整定PID的稳定平台调平控制  

Leveling control of an inertial platform based on single neuron self-tuning PID

作　　者：赵芃沛 孟卫锋 史永杰 李联涛 刘明雍[3] ZHAO Pengpei;MENG Weifeng;SHI Yongjie;LI Liantao;LIU Mingyong(Jiaxing Key Laboratory of Aero-Engine Manufacturing Technology of Key Components,Jiaxing Vocational and Technical College,Jiaxing 314036,China;Shaanxi Aeronautics and Astronautics Propulsion Co.,Ltd.,Xi’an 710103,China;School of Marine Science and Technology,Northwestern Polytechnical University,Xi’an 710072,China)

机构地区：[1]嘉兴职业技术学院嘉兴市航空发动机关键零部件制造重点实验室,浙江嘉兴314036 [2]陕西空天动力装备科技有限公司,西安710103 [3]西北工业大学航海学院,西安710072

出　　处：《兵器装备工程学报》2023年第1期183-187,247,共6页Journal of Ordnance Equipment Engineering

基　　金：国家自然基金项目(51879219)。

摘　　要：稳定平台调平回路大多采用PID进行控制,但这种算法具有抗干扰性能不高的问题,利用神经网络具有自学习、自组织、联想记忆和并行处理的功能和优势,设计了单神经元自整定PID控制算法。该控制算法不仅结构简单,而且适应性强,鲁棒性强。在设计中,采用了一种改良的Hebb学习算法对稳定平台调平回路进行控制。最后的仿真结果表明,单神经元自整定PID控制算法比传统PID控制算法在很多指标上都要好。尤其超调量、干扰抑制能力、过渡时间等动态指标非常优秀,是一种较为理想的控制算法,可以推广应用于各类稳定平台系统。The leveling loop of an inertial platform is mostly controlled by PID,but this algorithm has the problem of a low anti-interference performance.A single neuron self-tuning PID control algorithm is designed based on the advantages of neural network which has self-study,self-organization,associative memory and concurrent processing.The control algorithm is not only simple in structure,but also has strong acclimatization and robustness.An improved Hebb learning algorithm is applied to the design of controlling the leveling loop of the inertial platform.Finally,as the simulation results indicate,the single neuron self-tuning PID control algorithm is better than the traditional PID control algorithm in many indicators,especially in dynamic ones like overshoot,capacity of anti-interference and transition time.Thus,it is an ideal control algorithm and can be applied to all kinds of inertial platform systems.

关 键 词：稳定平台 调平回路 单神经元 自整定PID控制 

分 类 号：TP27[自动化与计算机技术—检测技术与自动化装置]