Method for dynamic system modeling based on state observer

Abstract

The invention discloses a method for dynamic system modeling based on a state observer. The method comprises a step (1) of collecting input and output data of a controlled object on site, wherein the collected data comprises a data segment of the controlled object transiting from some dynamic process to the steady-state process in response to a curve; a step (2) of selecting a segment of continuous data transiting from the dynamic state to the steady state in the dynamic process, setting a prediction model of the controlled object, a parameter and a state observer pole, setting an initial state of the observer as 0, and applying the state observer to observe an observation state of the controlled object; a step (3) of applying the prediction model of the controlled object to simulate a remaining data segment after the selection in the step (2) with the observation state as the initial state, solving error sum of squares actually output by the controlled object and a simulation curve, and taking the error sum of squares as an objective function to apply an intelligent optimization algorithm to perform hypothetical model parameter optimization. The special modeling testing does not need to be performed on the controlled object, and the method is more practical and more convenient.

Description

technical field [0001] The invention relates to a dynamic system modeling method based on a state observer, in particular to a dynamic closed-loop data in the running process as the original data for modeling, using a state observer to observe the state of the system dynamic process, and using the observed The state is the initial state of the system, and the modeling method of applying an intelligent optimization algorithm to optimize the estimated model parameters belongs to the field of modeling methods. Background technique [0002] The modeling process of traditional industrial controlled objects often needs to add step disturbance to the system to model when the system is in a steady state. This modeling method requires the system to be in a steady state, which is difficult to achieve in real industrial processes. Moreover, when a step disturbance is added to the system, the stability of the system will decrease, and even cause accidents, which is not conducive to the...

AI Technical Summary

Problems solved by technology

This modeling method requires the system to be in a steady state, which is difficult to achieve in real industrial processes

Moreover, when a step disturbance is added to the system, it will degrade the stability of the system and even cause accidents, which is not conducive to the safe and economical operation of the system.

Images