Research On Steady State Detection Method Of Chemical Process Based On Statistical Analysis

Whether the system is in a steady state is the most basic and common assumption when researching a process system.Before performing data correction calculations for a set of chemical process data,it is necessary to figure out whether the system is in a steady state or dynamic when the set of data is obtained,which will determine Which calibration method is used directly affects the selection of the method for subsequent control and optimization of the system.The application of measurement data is multifaceted.For example,the data required for intermediate tests and process design are always expected to be obtained under steady-state conditions;when the equipment is used for material and heat balance calculations,steady-state process data should also be used.Therefore,it is necessary to check whether the chemical process is in a steady state.It has already become an indispensable link in the process of data coordination,model correction,system optimization,and advanced process implementation.From a practical industrial point of view,this paper studies the steady-state detection method based on the statistical concept and combined with the traditional steady-state detection method.On the basis of studying the steady state recognition algorithms of the predecessors,the detection characteristics of several commonly used process steady state detection methods are compared,and the mathematical description of the ideal and actual state of the steady state in the chemical process is given.The more practical steady-state detection methods in the software have verified the characteristics and effects of their respective detections through actual data.Aiming at the variable characteristics of the process,a steady-state detection method based on weights is proposed on the basis of a priori method.The method of calculating the weights of linear and non-linear function weights is introduced,and the effectiveness and practicability of several steady-state detection methods involved in the article are verified by using the actual operating data of the atmospheric and vacuum unit.