Ship autopilot composite neural network PID control method

Abstract

The invention relates to a ship autopilot composite neural network PID control method. The method comprises the following steps: providing a composite neural network four-layer structure which comprises an error sequence input layer, a network algorithm hidden layer, a network output layer and a network identification layer; determining the number of neurons and parameter values of each layer through open-loop test training learning, wherein the output layer is composed of three adjustable PID control parameters with non-negative values larger than zero; taking a performance index function ofthe error quadratic sum to set a gradient steepest descent method of a weighting coefficient, and adding an inertia term to prevent local convergence; increasing jacobian information of the output course of the ship to the input steering angle, improving the learning capacity and sensitivity to control characteristics, and achieving PID control parameter online self-adaptive adjustment. Accordingto the control method disclosed in the invention, the course keeping control performance of the ship uncertain motion can be improved; a neural network algorithm improved by Jacobian information identification based on a gradient steepest descent method and ship course keeping and embedding of a radial basis Gaussian function into a back propagation hyperbolic function can be used to solve the problems of large course deviation amplitude and high course reciprocating crossing frequency, and energy conservation and consumption reduction are realized.

Description

technical field [0001] The invention relates to ship motion control and navigation technology, in particular to a ship autopilot compound neural network PID control method. Background technique [0002] The course keeping control of ocean-going ships is the key technology of ship navigation. Most of the ship heading control (Ship Heading Control1) based on PID means to control the heading of the ship as the course direction. Under the influence of marine environment, the bow deviates from the given course direction, and the autopilot adjusts the deflection of the rudder blades to generate additional torque to make the bow return to the given course. However, the ship has large inertia, large delay, uncertainty, and large Due to the characteristics of complex scale and shape, and large changes in loading capacity, the control parameters must adapt to the characteristics of the ship and can be automatically adjusted accordingly. This is the difficulty of maintaining the contro...

AI Technical Summary

Problems solved by technology

Under the influence of marine environment, the bow deviates from the given course direction, and the autopilot adjusts the deflection of the rudder blades to generate additional torque to make the bow return to the given course. However, the ship has large inertia, large delay, uncertainty, and large Due to the characteristics of complex scale and shape, and large changes in loading capacity, the control parameters should adapt to the characteristics of the ship and can be automatically adjusted accordingly. This is the difficulty of maintaining the control of the ship's course. Frequent adjustment of steering gear, reciprocating course changes, increased navigation resistance and high fuel consumption have always been the focus of research and solutions for ship autopilots.

[0005] figure 2 Deficiencies in the course control of ocean-going ships shown: figure 2 (a) and figure 2 (b) The ship swings on both sides of a given course for a long time, and its average course can still be around the positive course, figure 2 (a) is convergent and feasible, and the practical application pursues such an effect; figure 2 (b) is divergent and unusable; figure 2 (c) The ship swings unevenly from side to side in the yaw dead zone, and the average course deviates from the positive course side, and it lasts for a long time, which is also divergent and unusable; figure 2 (d) shows that the average course has a large deviation. After the correction of the PID course keeping controller, the ship returns to the positive course, which cannot meet the practical application

Based on the analysis of relevant literature on ship autopilots, the current four types of ship autopilot technologies are summarized: PID autopilot; PID-based control parameter optimization autopilot, in which there are many methods to optimize PID c

Images