Multi-modal bridge crane swing suppression control method based on artificial neural network

Abstract

The invention discloses a multi-modal bridge crane swing suppression control method based on an artificial neural network. The method comprises the steps of constructing a bridge crane multi-modal dynamic model with a distributed mass load; calculating a first optimal continuous pulse sequence under different parameters, and constructing an optimal continuous pulse data set; calculating a weight and a bias parameter of the artificial neural network, and constructing an artificial neural network identification related parameter; and calculating a second optimal continuous pulse sequence under the current real-time measured parameters by using the artificial neural network, and obtaining a final acceleration signal. According to the multi-modal bridge crane swing suppression control method, a PD controller tracks a final acceleration signal quadratic integral curve to realize positioning and anti-swing, a continuous pulse sequence is optimized through a particle swarm optimization algorithm, the bad performance of a traditional open-loop controller for a nonlinear system is solved, an optimal parameter set is learned, tedious offline iteration steps are avoided, the required optimal pulse sequence is generated in real time, and the efficiency of the controller is improved.

Description

technical field [0001] The invention relates to the technical field of anti-sway motion control of bridge cranes, in particular to a multi-mode bridge crane swing suppression control method based on artificial neural network. Background technique [0002] With the continuous improvement of modern industrial intelligence, the application of cranes is becoming more and more extensive. The traditional control methods for cranes mostly rely on the operator's experience, so the efficiency and stability of the control will be lacking. Moreover, due to the underactuation characteristics of the crane system, the angle of the hook and the load is regarded as the underactuation amount, which is difficult to control directly. [0003] Considering economy and convenience, bridge cranes are widely used in various industrial places, and most of the supporting control algorithms are open-loop control algorithms. Commonly used open-loop control algorithms include trajectory planning, input...

AI Technical Summary

Problems solved by technology

The traditional control methods for cranes mostly rely on the operator's experience, so the efficiency and stability of the control will be lacking

Moreover, due to the underactuation characteristics of the crane system, the angle of the hook and the load is used as the underactuation amount, which is difficult to directly control

[0003] Considering economy and convenience, bridge cranes are widely used in various industrial places, and most of the supporting control algorithms are open-loop control algorithms. Commonly used open-loop control algorithms include trajectory planning, input shaping, etc. , the trajectory planning method is to plan a curve that can suppress system oscillation by analyzing the input and output characteristics of the system, and let the car track this curve to achieve swing elimination. It is characterized by linearizing the model and solving a series of constraint equations , the input shaping also needs to linearize the system first, which analyzes each modal frequency of the system, solves a series of pulse signals that can suppress the system oscillation, and performs convolution operation with the initial system input signal to generate the final input signal, however, in practical applications, due to the non-negligible rope length between the hook and the load, the crane system usually exhibits double pendulum characteristics, so in order to fully suppress the oscillation of each mode of the crane, when designing for the input shaper , it is necessary to consider each modal frequency. Traditional controllers such as ZVDD, DZV, etc., are essentially discrete pulse sequences. When they are convolved with the input signal, the generated control signal is not smooth, which is lead to a reduction in the life of the control system

Images