A Neural Network Modeling Method for Bridge Cranes with Object Age Feature Membrane Calculation

Abstract

The invention discloses a bridge crane neural network modeling method with object age characteristic film calculation. The steps are: 1) obtain the input and output sampling data of the overhead crane operation process, and establish the RBF neural network model of the overhead crane; The sum of squared differences is used as the objective function; 3) Inspired by the aging phenomenon of biological cell membranes and intracellular substances, a bridge crane neural network modeling method with object age characteristic membrane calculation is abstracted; 4) Setting the operating parameters of the algorithm; 5) Run the calculation optimization method with the object age characteristic membrane to estimate the unknown parameters in the RBF neural network model of the overhead crane, and obtain the estimated value of the unknown parameters in the model by minimizing the objective function, and bring the estimated value into the RBF of the overhead crane In a neural network model, a nonlinear model is formed. The modeling method of the invention has the characteristics of anti-premature, high local search precision and fast convergence.

Description

technical field [0001] The invention relates to a bridge crane neural network modeling method with object age characteristic film calculation. Background technique [0002] The bridge crane is a widely used assembly transportation tool, the key to its work is to realize the accurate, fast and non-sway-free delivery of goods. However, since the dimension of the control quantity of the overhead crane is less than the degree of freedom of the controlled quantity, it belongs to a nonlinear underactuated system. When the trolley is moving, it will be disturbed by friction and wind, which will cause the goods to swing, and there is an operation Low efficiency, poor positioning accuracy, low safety factor and other shortcomings. Therefore, the overhead crane must be controlled safely and effectively. To achieve this goal, the key lies in the establishment of a high-precision overhead crane model. [0003] Researchers have achieved a lot of results by using the mechanism modeling ...

AI Technical Summary

Problems solved by technology

[0005] For engineering optimization problems with characteristics such as nonlinearity, constraints, and complexity, traditional optimization methods tend to fall into local optimum or even fail to obtain the optimal value, which makes the biologically inspired intelligent optimization method get people's attention

Images