Study On The Technology Of High Throughput Data Detection In The Fermentation Process

With the rapid development of modern biological engineering technology,the scale of production of fermentation industry is rapidly expanding,the production process is gradually strengthened,so the requirements for optimizing the fermentation process are very urgent.The optimal control of fermentation process to achieve the optimal production(the highest product quality,the lowest cost,the biggest production capacity)is one of the main problems in the field of fermentation engineering.So the research on the detection technology and the optimal control method of the biological fermentation process attract more and more attention.The detection of the fermentation parameters is the starting point of the control and optimization of the fermentation system.In order to control and optimize the fermentation system,it is necessary to build the foundation of accurately measuring the fermentation parameters.Most physical parameters and chemical parameters of fermentation process has been achieved real-time on-line measurement,such as pressure,temperature,pH value,dissolved oxygen concentration.But due to the lack of biosensor,there is no precise measurement method to measure some key parameters,such as substrate concentration,product concentration and biomass concentration,etc,generally the parameters of the laboratory can be obtained by chemical measurement or other instruments only after off-line sampling,so that not only reach the real-time control effect,and are particularly vulnerable to infection.Especially,there is no fast method for the measurement of the biomass of the fermentation process,which makes the advanced control and optimization of the fermentation process more difficult.The measurement method of mid infrared spectrometer in this paper is widely used in many fields because of its fast,accuracy and non-pollution,and at present the research of infrared spectrum in quantitative analysis is a hot research.In this paper,the fermentation experiment of production of y-PGA by Bacillus subtilis was used as the typical model,to study the application of infrared spectroscopy to the fermentation process.For the substrate and the products which need to be measured,their pure solution are measured by the infrared spectra.The results of the ideal measurement are obtained by the model,and the reliability of the spectrum is verified.Then a large number of spectrum samples were collected by the fermentation process of Bacillus subtilis,and the mathematical model of the infrared spectrum of the fermentation liquid was established by using the multiple linear regression algorithm.In the process of modeling,the influence of the selection of sample set,spectral pretreatment and the selection of spectral wavelength range,and the spectral measurement of biomass were specialized studied.At last,the higher accuracy model is obtained,which can predict the substrate concentration,product concentration and biomass concentration,and have the value for popularization.