论文部分内容阅读
水下及水面浮动平台是海洋开发研究中常见的支撑平台,它的平稳性对平台上设备的正常工作有极其重要的意义,由于水下环境复杂,受水流等不确定因素的影响,平台难以保持平稳状态,这就要求控制系统具有较强的抗干扰性和较高的控制精度。基于某水下试验平台的姿态控制系统,采用液压马达驱动绞车控制张力腿,分析并建立液压系统模型,结合神经网络控制和PID控制的优点,设计单神经元PID控制器,利用输入输出的误差,通过带监督的Hebb学习法则对权值进行实时修正,这种控制器不仅结构简单,且具有较好的自适应能力和鲁棒性。利用计算机仿真技术,在Malab中用simulink仿真平台,结合S函数进行建模仿真,得出仿真效果曲线图,结果表明单神经元PID控制器具有良好的控制效果和适应性。
Underwater and surface floating platform is a common support platform for marine development research. Its stability is extremely important to the normal operation of the equipment on the platform. Due to the complex underwater environment and the uncertainties such as water flow, the platform is difficult Maintain a steady state, which requires the control system has strong anti-interference and high control accuracy. Based on the attitude control system of an underwater test platform, a hydraulic motor is used to drive the tension leg to control the tension leg. The hydraulic system model is analyzed and established. Combining the advantages of neural network control and PID control, a single neuron PID controller is designed to utilize the error of input and output , Through real-time correction of the weights by supervised Hebb learning rules, the controller is not only simple in structure, but also has better self-adaptability and robustness. Using computer simulation technology, using simulink simulation platform in Malab, combined with S function modeling and simulation, the simulation results show that the single neuron PID controller has good control effect and adaptability.