基于神经滑模的某顶置火炮自动装填调炮控制  

Self-loading Adjustment of a Top-mounted Gun Based on a Novel Neural Slide Mode Control

作　　者：高强[1] 庞雷 杨国来[1] 王力[1] 侯远龙[1] 

机构地区：[1]南京理工大学机械工程学院,江苏南京210014 [2]总装工兵军代局驻武汉地区军代室,湖北武汉430073

出　　处：《机床与液压》2013年第11期35-38,63,共5页Machine Tool & Hydraulics

基　　金：国家重点基础研究发展计划973资助项目(61311603)

摘　　要：设计了一种神经滑模控制器,并将其应用于某顶置火炮的调炮控制,以提高非线性环节影响下调炮系统的控制品质。该控制策略采用了一种新型切换函数,基于RBF神经网络对切换增益进行动态调节,以获得动态最优控制性能并抑制SMC固有的陡振现象。数值仿真和样机试验结果表明:所提出之控制策略对参数摄动及负载扰动具有很好的鲁棒性,能够满足自动装填调炮技术指标要求,实现预期的快速、平稳、精准调炮,具有重要的实际工程应用价值。To solve the nonlinearities and enhance the performances of a gun control system （GCS）,a novel neural slide mode control （NSMC） strategy was proposed and implemented on the elevation and decline control of a certain top-mounted GCS.A novel switch function was employed and its corresponding gain was dynamically adjusted by the RBF neural network.Taking advantage of this strategy,optimum control performances and suppression of the inherent chatter phenomenon can be well achieved.Both numerical simulations and experiments on a semi-physics simulation platform were conducted to investigate the control performance of the GCS.The results demonstrate that the proposed NSMC is of high robustness to the parameter perturbation and load disturbance of the GCS.The proposed NSMC can satisfy the requirements of gun adjustments,and the fast,smooth and accurate adjustments of the gun can be well achieved.

关 键 词：调炮控制 交流伺服系统 滑模控制器 RBF神经网络 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]