Air separation system argon fraction variable multi-step prediction method based on iterative multi-output-Markov chain

Abstract

The invention discloses an air separation system argon fraction variable multi-step prediction method based on an iterative multi-output-Markov chain, and the method comprises the following steps: S1, collecting data, constructing a data set, carrying out the normalization processing of the data set, and dividing the data set into a training set, a verification set and a test set; S2, constructing an iterative multi-output prediction model; s3, carrying out error correction by adopting a Markov chain. According to the method, the iterative multi-output method is combined with the Markov chain, on one hand, the optimal input and output dimension of the prediction model is determined in a parameter optimization mode, iteration error accumulation of the model is effectively reduced, and the applicability and flexibility of the model are enhanced; and on the other hand, the Markov chain can determine the correction range and direction of the prediction value by calculating the error state of the prediction value, so that the deviation degree of the prediction value and the real value is smaller, the prediction precision is higher, and the method has better applicability to variable time sequence multi-step prediction of the air separation system.

Description

technical field [0001] The invention relates to the technical field of automation, in particular to a multi-step prediction method for argon fraction variables in an air separation system based on iterative multi-output-Markov chains. Background technique [0002] Air separation refers to the process of separating air into nitrogen, oxygen and argon under low temperature conditions through physical or chemical reactions such as expansion and rectification. The air separation process is a process industry, the production process is continuous and the mechanism is complex, and some key index variable data are important manifestations of equipment and process conditions. Industrial field personnel usually adjust the operation of related equipment according to the changing trend of key indicators, so as to ensure the safe and stable operation of the production process. However, in the actual production site, some important indicators are difficult to detect in real time or comp...

AI Technical Summary

Problems solved by technology

[0004] The present invention mainly solves the problem of large uncertainty and randomness in the prediction results of air separation in the existing technology; it provides a multi-step prediction method for the argon fraction variable of the air separation system based on iterative multi-output-Markov chain, reducing The randomness of the prediction results improves the prediction accuracy and has good applicability

Images