Modeling method and model of linear first-order inverted pendulum system based on cnn-arx model

Abstract

The invention discloses a linear first-order inverted pendulum system modeling method and model based on a CNN-ARX model. The inverted pendulum is a relatively complex system with strong nonlinearity and instability, and an accurate mechanism mathematical model of the system is difficult to obtain. The CNN‑ARX model can be used to describe the dynamic characteristics of such systems. The present invention uses the convolutional neural network (CNN) technology, the local linearization method and the state-dependent ARX model to construct the CNN-ARX model structure, optimizes the parameters of the convolutional neural network through the RMSProp algorithm, and adopts the random deactivation technology (dropout) The output of the hidden layer of the integrated neural network is randomly reset to zero, which reduces the interdependence between nodes and reduces the risk of model overfitting. The invention can improve the prediction accuracy and robustness of the identification model of the inverted pendulum system, and has high practical value and application prospect.

Description

technical field [0001] The invention relates to the field of engineering design and optimization, and can utilize the input and output discrete time series data of a linear first-order inverted pendulum system, in particular to a linear first-order inverted pendulum system modeling method based on a CNN-ARX model. Background technique [0002] The linear first-order inverted pendulum is a kind of strong nonlinear, unstable and relatively complex system. It can be modeled in a data-driven way, and the relationship model of the system input and output variables can be constructed to describe the dynamic characteristics of the inverted pendulum system. As a nonlinear modeling method, the state-dependent ARX model has the advantages of describing the nonlinear dynamic characteristics of the state-dependent function coefficients and easy application of the autoregressive structure to control. Linear time series modeling has a wide range of applications. One of the core problems i...

AI Technical Summary

Problems solved by technology

As a nonlinear modeling method, the state-dependent ARX model has the advantages of the ability to describe the nonlinear dynamic characteristics of the coefficients of the state-dependent function and the advantages that the autoregressive structure is easy to apply to control. Linear time series modeling is widely used in the field of linear time series modeling. One of the core problems of using the state-dependent ARX model to model the inverted pendulum system is to find the appropriate structure of the state-dependent function coefficients. Using the neural network to approximate the coefficients of the state-dependent ARX model can be Obtain higher-performance identification models, such as RBF-ARX, DBN-ARX, etc. In theory, the neural network can fit any complex nonlinear relationship when the network level is deep enough. In practical applications, the neural network may There are problems such as underfitting, overfitting and gradient disappearance

Images