基于神经网络PID的气垫船转艏角速度保持控制研究

Yawing Angular Velocity Keeping Control of Air Cushion Vehicle Based on PID-Neural Network Controller

为解决全垫升气垫船受风和惯性力影响较大,容易产生侧漂,其转艏角速度较常规船舶难于控制的问题,建立了全垫升气垫船的平面运动模型。设计了神经网络PID控制器,利用BP神经网络具有的任意非线性表达能力,通过对系统性能的学习,实现具有最佳组合的PID控制。对气垫船的转艏角速度控制进行仿真,并与常规PID控制效果进行对比,结果表明,在全垫升气垫船受风力干扰的情况下,神经网络PID控制器相比常规PID控制器在克服风力干扰方面控制效果更好。

For solving the problem that the air cushion vehicle is easy to be affected by wind and inertia force, and it is easy to have large side drift, so the keeping of its angular velocity is difficult to control. In order to improve the heading maneuver of Air Cushion Vehicle, the Yawing Angular Velocity Keeping Control must be used frequently, established the air cushion vehicle motion model, and the design of PID-neural network controller based on BP neural network with arbitrary nonlinear expression ability, the system can achieve the best combination of learning. The control of the angular velocity of air cushion vehicle is simulated by the model, and the results are compared with those of the conventional PID control. The simulation results show that the air cushion vehicle in the case of wind disturbance, PID-neural network controller of the yawing angular velocity control, control effect in overcoming the wind disturbance compared with the conventional PID controller has better effect.