Method for estimating initial conditions of media during continuous fermentation

Abstract

The invention discloses a method for estimating the initial conditions of a culture medium in a continuous fermentation process. Discrete state space model; introduce auxiliary variables on the basis of Kalman filter to obtain the predicted distribution obeying the student distribution; use variational Bayesian theory to represent the system state and auxiliary variables; predict the system state and auxiliary variables, and update the system state and auxiliary variables Until the set number of iterations is reached, the system state and auxiliary variables at this time are output; until the set number of running steps is reached, the update is ended, and the estimated value of the key variables in the medium at the initial moment is obtained. By establishing a three-variable mathematical model of the continuous fermentation process and introducing student distribution, the invention can quickly and accurately estimate the initial conditions of key variables in the culture medium, thereby ensuring the safety of the fermentation process and the fermentation quality.

Description

technical field [0001] The invention relates to the technical field of estimation of initial conditions of fermentation medium, in particular to a method for estimating initial conditions of medium in a continuous fermentation process. Background technique [0002] In the existing continuous fermentation process, due to imperfect sensor equipment or limitation of process conditions, most operators use offline analysis methods to analyze the values ​​of key variables in the culture medium. However, the method of offline analysis is intermittent, which further increases the time delay characteristic in the intermittent process, which is not conducive to real-time monitoring of fermentation quality or real-time diagnosis of faults in the fermentation process. The estimation of the initial condition of the medium is a key part of the analysis of the medium, and the inaccurate estimation of the initial condition of the medium will affect the quality of the entire fermentation pro...

AI Technical Summary

Problems solved by technology

[0002] In the existing continuous fermentation process, due to the imperfection of sensor equipment or the limitation of process conditions, operators mostly use off-line analysis methods to analyze the values ​​of key variables in the medium

However, the off-line analysis method is intermittent, and the time-delay characteristics are further increased in the intermittent process, which is not conducive to real-time monitoring of fermentation quality or real-time diagnosis of faults in the fermentation process

Estimation of the initial condition of the medium is a key part of the analysis of the medium, and inaccurate estimation of the initial condition of the medium will affect the quality of the entire fermentation process

Although there are currently methods for estimating the initialization conditions of the medium, there are problems such as large amount of calculation and uncertain performance indicators, which limit the application of these estimation methods in the continuous fermentation process.

Especially for the key variables in the continuous fermentation process, there are currently no good estimation methods to estimate the initial conditions of

Images