Multi-loop model-free self-adaptation heading control method for ship

Abstract

The invention provides a multi-loop model-free self-adaptation heading control method for a ship. A control system adopts a combination mode of an outer ring navigation controller and an inner ring angular velocity controller, the outer ring navigation controller is used for calculating an expected heading turning angular velocity, the inner ring MFAC angular velocity controller finishes angular velocity control, and the purpose of yaw control is indirectly achieved. In practical application, an angular velocity sensor is loud in noise, and angular velocity prediction is carried out by utilizing historical input and output data of the control system; angular velocity data is filtered through a Kalman filter to serve as feedback input of the inner ring angular velocity controller, the noiseof the angular velocity sensor is inhibited, and a control effect in actual application is effectively improved. According to the method, an MFAC theory is introduced into the ship heading control field, the special advantages of self-adaptability and online data driving of the MFAC theory are used, and the multi-loop model-free self-adaptation heading control method for the ship has high self-adaptability.

Description

Technical field [0001] The invention relates to a multi-loop model-free adaptive heading control method for ships, which is used for ship heading control and belongs to the field of ship automatic motion control. The ship in the present invention refers to various underwater navigation in a broad sense. Equipment, such as surface ships, submersibles, submarines, underwater unmanned vehicles, surface unmanned boats, etc., collectively referred to as ships in the present invention, are all within the scope of application of the present invention. Background technique [0002] The heading control of the ship is very important to the ship system. Only by ensuring the stable heading of the ship can the desired track be effectively tracked. At present, in practical engineering applications, the course control of ships basically uses PID control algorithms and conventional control algorithms developed based on "model-oriented" design strategies. The PID controller is a data-driven cont...

AI Technical Summary

Problems solved by technology

The PID controller is a data-driven control algorithm based on offline data. However, when the ship is operating in the ocean environment, it is easily affected by model perturbation and ocean environment interference, which makes it difficult for the PID controller to maintain a consistent control effect. Only by adjusting the parameters can the system maintain good control performance or stability

However, the controller developed based on the "model-oriented" design strategy relies heavily on the mathematical model of the system. Because it is very difficult to obtain an accurate mathematical model, there are unmodeled dynamics, model perturbations, etc., which lead to poor self-adaptation of the system, and it is difficult to obtain an accurate mathematical model. Guarantee the robust performance of the system, so it is difficult to obtain application in engineering

[0003] Model-free adaptive control theory (MFAC) does not depend on precise mathematical models, and uses data-driven methods to adjust control parameters. It has been applied in many fields and achieved good results. For example, the publication number proposed by Cheng Qiming et al. It is an invention patent of "a gray model-free control method for a large time-delay system". The paper "Model-free Adaptive Control of Large Ship Integrated Anti-rolling System" published by Ma Jie et al., but none of the disclosed methods are applicable to The heading control of the ship, because the heading system of the ship does not meet the requirements of the MFAC algorithm for the "quasilinear" assumption of the controlled system, that is, when the input of the controlled system increases, the output of the corresponding controlled system does not decrease

Images