ARMA (Autoregressive Moving Average) model order determination method based on steady state graph

Abstract

The invention discloses an ARMA (Autoregressive Moving Average) model order determination method based on a steady state graph, belonging to the technical field of modal analysis of an ARMA model, which comprises the following steps: acquiring a response signal by an accumulation sensor, setting up the ARMA model for the response signal, carrying out the modal analysis, identifying a modal parameter, making the model steady state graph, determining the model order from the steady state graph and rejecting false modals; calculating modal frequency average values of orders after rejecting the false modals, calculating the relative difference of the calculated modal frequency of the orders and a frequency average value under different orders, and simultaneously setting the modal frequency tolerance of the orders; searching the model orders from high to low, judging the relative difference of the calculated modal frequency and the frequency average value relative difference and the frequency tolerance, if the relative difference is less then the tolerance, stopping searching, and finally recording the current model order which is the minimum model order; and among the model orders from low to high by beginning from the minimum order, calculating the Akaike standard value under different orders, wherein the model order corresponding to the minimum Akaike is the optimal model order.

Description

technical field [0001] The invention relates to the technical field of ARMA model modal analysis, in particular to a method for determining the order of an ARMA model based on a steady state diagram. Background technique [0002] ARMA (Autoregressive Moving Average) model is a common form of time series model, suitable for modal analysis based on response signals. In structural dynamics analysis, the input signal is often difficult to obtain, while the output signal is easy to obtain. At this time, the response signal is usually used to establish an ARMA model, estimate model parameters, conduct modal analysis, and identify modal parameters. The parameter estimation of the model assumes that the order of the model is known, but the actual order of the model is unknown. Therefore, reasonably determining the model order is the key to accurate modeling, and the determination of the model order has always been a difficult point in research. Among the methods for determining t...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide a method for determining the order of the ARMA model by using the steady-state diagram, which solves the problem that the order of the ARMA model does not match the actual situation, so that the order of the ARMA model is more consistent with the actual situation. Conformity, improve the identification accuracy of modal parameters of ARMA model in modal analysis

Images