基于时滞特性的超空泡航行体预测控制  

Predictive control of supercavitating vehicle based on time delay characteristics

作　　者：韩云涛[1] 许振 白涛[1] 张骁 HAN Yuntao;XU Zhen;BAI Tao;ZHANG Xiao(College of Automation,Harbin Engineering University,Harbin 150001,China)

机构地区：[1]哈尔滨工程大学自动化学院,黑龙江哈尔滨150001

出　　处：《华中科技大学学报（自然科学版）》2020年第7期52-58,共7页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(51209049);黑龙江省自然科学基金资助项目(E2017015)。

摘　　要：针对航行体在运动条件下存在的时滞问题,提出了一种状态空间预测控制方法.首先,考虑航行体动力学模型的时滞特性,将模型中非线性滑行力分解为时滞与非时滞两部分,通过相应变换推导出航行体的线性变参数(LPV)时滞模型.然后,针对航行体的参数摄动,利用多胞形理论得到LPV系统的等价模型,减小在线实时计算量.最后,将实时可测参数信息与预测模型相结合设计状态空间模型预测控制器,研究系统的动态性能.仿真结果表明,该控制器对具有时滞特性的超空泡航行体有良好的控制效果及抗扰动能力.The time-delay characteristics of underwater supercavitating bodies have great influence on the control efficiency and stability of the vehicle.A state space predictive control method was proposed for the time lag of the vehicle under moving conditions.First,considering the time-delay characteristics of the vehicle dynamics model,the nonlinear planing force in the model was decomposed into time delay and non-time delay part.The linear parameter-varying(LPV)systerm time-delay model of the vehicle was derived by corresponding transformation.Then,for the time-varying and uncontrollable cavitation radius of the vehicle,an equivalent model of the LPV system was obtained using the multi-cell theory,which could reduce the amount of online real-time calculation.Finally,the predictive model was combined with the real-time measurable parameter information to design the state space model predictive controller,and the dynamic performance of the system was studied.Simulation results show that the controller has excellent control effect and anti-disturbance capability for supercavitating vehicle with time-delay characteristics.

关 键 词：超空泡航行体 滑行力 时滞 预测控制 线性变参数(LPV)系统 鲁棒性 

分 类 号：U674[交通运输工程—船舶及航道工程]