船舶吊舱推进电机无模型自适应矢量控制策略研究

发布时间：2018-02-05 02:25

本文关键词： 船舶电力推进吊舱推进电机矢量控制无模型自适应控制扰动抑制　出处：《大连海事大学》2015年博士论文　论文类型：学位论文

【摘要】：远洋运输是高能耗高排放的行业,实现船舶节能减排是海运业发展的重大需求。研究船舶吊舱推进系统的相关控制问题,不仅对解决推进电机的运行、优化与管理有重要的指导作用,更对实现海运业节能减排有重要意义。针对船舶吊舱推进控制系统的大惯性、强耦合、模型参数不确定以及航行中受到的风、浪、流等测量扰动及负载扰动的问题,本论文研究了基于无模型自适应矢量控制的船舶吊舱推进系统控制策略,具体内容如下：针对常规船舶吊舱推进矢量控制系统中由于不确定性动态而存在的模型摄动问题,将无模型自适应控制算法与矢量控制方法相融合,提出一种全新的无模型自适应矢量控制算法,推导出SSP推进电机的动态线性化方程,基于速度跟踪误差设计了无模型自适应矢量控制器,在线调整伪偏导数,保证了推进电机控制系统跟踪误差一致且有界。并通过严格的收敛性证明和仿真实验论证了该方法应用于船舶吊舱推进控制系统的有效性。针对无模型自适应矢量控制方法仅利用过去一步采样时刻信息的问题,在高阶学习律的基础上提出了一种高阶无模型自适应矢量控制算法,通过设计一个新的控制输入准则函数,使得更多船舶航行过程中的控制信息得到充分利用,从而使得控制器的设计更加灵活,通过严格的数学证明,论证了伪偏导数估计值的有界性、系统稳定收敛性以及控制输入的有界性,仿真实验表明所提出方法进一步提高了船舶吊舱推进控制系统的收敛性能。针对船舶吊舱推进系统由于受到外部海况等未知负载扰动,导致控制量的频繁动作等问题,设计一种带死区扰动补偿的无模型自适应矢量控制器,首先在无模型自适应矢量控制器控制律的更新算法中引入一个控制死区,并设计了基于扩张状态观测器的扰动估计器,最后进行了收敛性证明,仿真实验表明该算法可以抑制未知负载扰动的作用,避免控制量的频繁动作对SSP推进控制系统执行机构的齿轮传动部件的较大磨损,提高控制系统的鲁棒性能。针对船舶吊舱推进系统的复杂动态特性及恶劣工作环境而带来的测量扰动,进行了一些实际应用的推广,研究了无模型自适应矢量控制的主动抗扰能力,提出了一种基于小波分析的无模型自适应矢量控制器,该控制器应用小波分析对输出信号进行滤波处理,然后将处理后的信号应用于无模型自适应矢量控制算法构造控制律,最后理论证明了当系统存在有界测量扰动时改进算法的稳定性和收敛性,仿真实验论证了该算法对于抑制测量扰动的有效性。
[Abstract]:Ocean transportation is an industry with high energy consumption and high emission. It is an important demand for marine transportation industry to realize energy saving and emission reduction. To study the control problems of ship podded propulsion system is not only to solve the operation of propulsion motor. Optimization and management play an important role in guiding the marine transportation industry to achieve energy saving and emission reduction. Aiming at the large inertia, strong coupling, model parameters uncertainty and wind in navigation of the propulsion control system of ship pods. In this paper, the control strategy of ship podded propulsion system based on modelless adaptive vector control is studied. The main contents are as follows: aiming at the model perturbation problem existing in the conventional ship podded propulsion vector control system due to the uncertain dynamics, the model-free adaptive control algorithm is combined with the vector control method. A novel model-free adaptive vector control algorithm is proposed, and the dynamic linearization equation of SSP propulsion motor is derived. The model-free adaptive vector controller is designed based on the velocity tracking error. Adjust the pseudo-partial derivative online. The tracking error of propulsion motor control system is consistent and bounded. The validity of this method applied to ship pods propulsion control system is proved by strict convergence proof and simulation experiment. The method of quantity control only uses the information of sampling time in the past step. On the basis of higher order learning law, an adaptive vector control algorithm with high order modeless is proposed. By designing a new control input criterion function, more and more control information during ship navigation are fully utilized. Through strict mathematical proof, the boundedness of pseudo-partial derivative estimator, the stability convergence of system and the boundedness of control input are proved. The simulation results show that the proposed method can further improve the convergence performance of the ship pods propulsion control system. The ship pods propulsion system is disturbed by unknown loads such as external sea conditions. In this paper, a model-free adaptive vector controller with dead-zone disturbance compensation is designed. Firstly, a dead-time control algorithm is introduced to update the control law of the model-free adaptive vector controller. A disturbance estimator based on the extended state observer is designed. Finally, the convergence is proved. The simulation results show that the proposed algorithm can suppress the unknown load disturbance. Avoid the large wear of the gear transmission parts of the actuators of the SSP propulsion control system by avoiding the frequent movement of the control quantities. Improving the robust performance of the control system. Aiming at the complex dynamic characteristics of the ship podded propulsion system and the measurement disturbance caused by the bad working environment, some practical applications are extended. The active immunity ability of model-free adaptive vector control is studied. A model free adaptive vector controller based on wavelet analysis is proposed. Wavelet analysis is used to filter the output signal. Then the processed signal is applied to the model-free adaptive vector control algorithm to construct the control law. Finally, the stability and convergence of the improved algorithm are proved when the system has bounded measurement disturbance. Simulation results demonstrate the effectiveness of the proposed algorithm in suppressing measurement disturbances.
【学位授予单位】：大连海事大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：U664.3

【参考文献】