Method for predicting throughput of large-scale production reactor and production process

Abstract

The invention relates to the technical field of cell culture, in particular to a method for predicting throughput of a large-scale production reactor and a production process. The method comprises the following steps: setting variable parameters; constructing a mathematical relationship model of the mass transfer coefficient Kla in the reactor and the variable parameters; operating data of the corresponding reactors are collected, and constant values of K, alpha and beta of the corresponding reactors are solved; according to historical data of the basic reactor, predicting the daily throughput in the culture period of the expanded reactor. The method has the beneficial effects that on the premise that the dissolved oxygen value in a culture system is kept constant, the ventilation value at any rotating speed can be predicted by measuring the mass transfer correlation coefficient of the reactor and on the premise that the structure of the tank body is not changed, so that the shear force damage of stirring and ventilation is balanced, and the success rate of amplified production is improved.

Description

Technical Field

[0001] The invention relates to the technical field of cell culture, and in particular to a method and a production process for predicting the ventilation volume of a large-scale production reactor. Background Art

[0002] The scale-up of bioreactors is an important part of the process development of cell culture. There are many factors that need to be examined in the process of studying the scale-up of mammalian cell culture processes, such as mixing time, oxygen transfer and carbon dioxide removal. At present, the process development of cell culture processes is usually carried out in small-scale bioreactors first. After the process is locked, it is scaled up step by step until it is used in larger bioreactors for commercial production. Compared with small reactors, the reduction in mixing and mass transfer effects in large reactors will cause problems such as poor carbon dioxide escape effect and longer mixing time, which will lead to cell growth not being as expected when the culture process established in the small-scale reactor is scaled up to the large-scale reactor for production, which is manifested as a series of problems such as reduced cell density, shortened culture cycle, significant decrease in yield or difficulty in controlling product quality. Therefore, it is of great significance to establish appropriate strategies and models for scale-up of cell culture processes for stable large-scale production.

[0003] The parameters in the cell culture process can be divided into two categories: parameters that do not change with the size of the reactor, such as pH, DO, temperature, feeding time, etc. These parameters can be kept consistent before and after the scale-up process. Parameters that change with the size of the reactor, such as aeration rate, stirring speed, etc.

[0004] Regarding the stirring speed, on the one hand, a too low stirring speed will lead to a too long mixing time, resulting in poor mass transfer and thus affecting cell growth; on the other hand, in order to maintain a constant DO value, a too low stirring speed will lead to an increase in oxygen ventilation, thereby causing damage to the cells; and although a too high stirring speed will result in better mixing conditions, the shear force caused by the stirring may also cause irreversible damage to the cells.

[0005] The size of the gas ventilation rate will change according to the size and method of the reactor ventilation aperture, as well as the size and form of the stirring paddle. Too high aeration rate will cause the gas at the outlet to instantly displace the liquid at a too high flow rate, thus causing damage to the cells. On the other hand, too high aeration rate will increase the number of bubbles in the liquid. When the bubbles reach the top of the liquid and burst, the shear force will cause damage to the cells adsorbed around them.

[0006] In summary, the shear force in the reactor mainly comes from the following three sources: 1. Bubble bursting at the liquid surface 2. Shear caused by the rotation of the stirring paddle 3. The gas flow rate at the outlet of the gas distributor is too high, causing shear on the liquid. Therefore, the influence of the three factors should be comprehensively considered during the scale-up of the bioreactor. It is necessary to consider both providing sufficient mixing capacity for the reactor and avoiding unnecessary cell damage caused by shear force under high speed and high ventilation conditions. These problems are the main reasons for the scale-up effect being lower than expected or the scale-up failure.

[0007] Prior art, application number: CN202411130287X, a universal model and method for predicting the growth rate of microorganisms in a photobioreactor, in which ventilation / agitation power is used as an important part of the prediction, and this prediction is limited to the growth rate in the same reactor. If the ventilation rate and agitation power in reactors of different sizes are different, therefore, if a larger reactor culture prediction is required, the corresponding ventilation rate and agitation power need to be known, and adjusted in real time according to the needs of cell culture, otherwise the cells may be damaged, resulting in the failure of the entire experiment.

[0008] Therefore, it is necessary to design a large-scale production reactor ventilation volume prediction method and production process to solve the above problems. Summary of the invention

[0009] The purpose of the present invention is to provide a method and a production process for predicting the ventilation volume of a large-scale production reactor to overcome the above-mentioned shortcomings of the current prior art.

[0010] In order to achieve the above object, the present invention adopts the following technical solutions: A method for predicting the ventilation volume of a large-scale production reactor, characterized in that it comprises the following steps: S1. Variable parameter setting: The variable parameters include the volume of the cell culture medium V, the input power P of the stirring paddle of the reactor, and the apparent ventilation rate Vs; S2. Construct a mathematical relationship model between the mass transfer coefficient Kla in the reactor and the variable parameters: ; Where K, α and β are characteristic constants of the reactor; S3, collect the operating data of the basic reactor and the amplified reactor, and solve the characteristic constants (K 0 , α 0 , β 0 )、(K 1 , α 1 , β 1 ); The method for solving the characteristic constant includes: S31, adding a fixed volume of cell culture medium into the reactor, and calibrating DO according to the set process, wherein DO is the dissolved oxygen concentration; S32, close the air intake, open the nitrogen intake, blow the DO to below 5%, and close the nitrogen intake; S33, arbitrarily set n groups of stirring speeds N and ventilation rates Q to obtain a first data group (Ni, Qi), i=1, 2, ..., n; S34, setting the stirring speed and ventilation rate of the reactor in sequence according to the first data group; after each set of data, start parameter monitoring, record the change value of DO over time, and when DO is greater than 95%, end the experiment, then balance the volume of the cell culture medium and calibrate DO, and then set the next stirring speed and ventilation rate data group, and so on, until n groups of DO change values ​​over time under different stirring speeds N and ventilation rates Q are obtained; S35, performing n linear fittings according to the change values ​​of n groups of DO over time, respectively, to obtain n slope parameters, the slope parameter being the mass transfer coefficient Kla, and further obtaining n groups of mass transfer coefficients Kla; S36, converting n groups of the first data groups (Ni, Qi) into the impeller input power P and the apparent ventilation rate Vs by formula (1) and formula (2), to obtain n groups of second data groups (Pi, Vsi), i=1, 2, ..., n; The formula (1) is: ; Where Np is the power factor of the stirring paddle, N is the stirring speed, is the diameter of the stirring paddle, and ρ is the liquid density; The formula (2) is: ; Among them, Q is the ventilation rate, and D is the tank radius; S37, according to the n second data sets (Pi, Vsi) and the volume of the cell culture medium V, substitute into the mathematical relationship model Kla=K(P / V) α (Vs) β , and solve for the corresponding reactor's K, α and β, and obtain the relationship model between the mass transfer coefficient Kla and the impeller input power P and the apparent ventilation rate Vs; S4. Predict the daily ventilation volume Qgas of the expanded reactor during the culture period based on the historical data of the basic reactor.

[0011] Preferably, The prediction method in step S4 includes: S41, according to the mass transfer coefficient Kla of the basic reactor and the mass transfer coefficient Kla* of the expanded reactor being equal, That is, formula (3) is: K 0 (P 0 / V 0 ) α0 (Vs 0 ) β0 =K 1 (P 1 / V 1 ) α1 (Vs 1 ) β1 , where P 0 、V 0 , Vs 0 , are the impeller input power, cell culture medium volume, and superficial aeration rate of the basic reactor, respectively. 1 、V 1 , Vs 1 , are the impeller input power, cell culture medium volume, and superficial aeration rate of the expanded reactor, respectively; S42, setting the stirring speed N of the enlarged reactor; S43, obtain the historical data of normal production of the basic reactor, and obtain the parameter data P of each day during the culture cycle. 0 、V 0 , Vs 0 , the cell culture medium volume V in the expanded reactor 1 , stirring speed N and stirring paddle input power P 1 If known, the apparent ventilation rate Vs of the expanded reactor per day during the culture period can be calculated according to formula (3): 1 , and then calculate the daily ventilation volume Qgas through formula (4); The formula (4) is: ; Where Qgas is the ventilation volume and Ar is the surface area of ​​the culture medium in the reactor.

[0012] Preferably, in step S42, the stirring speed N of the expanded reactor is set according to the culture cycle of historical data, and the stirring speed of the expanded reactor is set specifically according to the maximum ventilation volume of the predicted model scale so that it does not exceed the maximum ventilation value, or the ventilation volume is controlled within the optimal range for cell culture, or it is adjusted according to the set ventilation rate Q and the corresponding stirring speed N is calculated.

[0013] Preferably, in step S4, the basic reactor calculates Kla corresponding to different days according to the culture cycle, that is, the 1st day, the 2nd day, ..., the Nth day, and obtains Kla1, Kla2, ..., KlaN accordingly. The Kla value of the basic reactor corresponding to the daily ventilation volume and the set stirring speed N of the expanded reactor are calculated as needed to calculate the apparent ventilation rate Vs of the expanded reactor corresponding to the number of days. 1 .

[0014] A production process for a large-scale production reactor comprises the following steps: predicting the ventilation volume Qgas of each day in the culture cycle after expansion according to the ventilation volume prediction method of the large-scale production reactor described in steps S41-S43, judging whether the ventilation volume Qgas meets the process requirements, and if so, indicating that the stirring paddle input power P currently set is 1 , Apparent Ventilation Rate Vs 1 Under the condition of , the expanded reactor has the conditions for normal cell growth; if the ventilation volume Qgas does not meet the process requirements, according to formula (3): K 0 (P 0 / V 0 ) α0 (Vs 0 ) β0 =K 1 (P 1 / V 1 ) α1 (Vs 1 ) β1 Adjustment parameters: stirring paddle input power P 1 Recalculate the apparent ventilation rate Vs 1 , and then recalculate the ventilation volume Qgas so that the recalculated ventilation volume Qgas is within the process requirements.

[0015] The beneficial effects of the present invention are as follows: the technical solution is based on the premise of maintaining a constant dissolved oxygen value in the culture system. By measuring the mass transfer correlation coefficient of the reactor, the ventilation value at any rotation speed can be predicted under the premise that the tank structure remains unchanged, thereby balancing the shear force damage of stirring and ventilation. The ventilation level after scale-up can be predicted based on the oxygen consumption of the pilot-scale reactor. At the same time, a comprehensive consideration is made between the shear forces brought by stirring and ventilation, thereby improving the success rate of scale-up production. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a line graph of Kla fitting of a 10L reactor of a large-scale production reactor ventilation prediction method and production process of the present invention; Figure 2 It is a line comparison diagram of the predicted ventilation volume and the actual ventilation volume of the present invention. DETAILED DESCRIPTION

[0017] The mathematical symbols and definitions involved in this scheme and the specification can be found in Table (1): Table (1) Definitions and explanations of the meanings of all mathematical letters and symbols:

[0018] A method for predicting the ventilation rate of large-scale production reactors is based on the analysis of oxygen consumption rate and the common points between reactors of different sizes, and it is concluded that the mass transfer coefficient Kla is the same among them; The specific analysis process includes the following steps: It can be seen that the oxygen consumption rate OUR = single cell oxygen consumption rate q (O 2 )*X cell density. When the production is scaled up, the cells and culture process are the same, so the oxygen consumption of a single cell and the cell density are also the same, so the oxygen consumption rate OUR in reactors of different sizes is equal; Since the oxygen consumption rate OUR = oxygen transfer rate OTR, the value of dissolved oxygen remains basically constant during the culture process. Therefore, the supply and demand of oxygen by cells is in a balanced relationship during the entire culture process, that is, the consumption of oxygen is the same as the transfer of oxygen. Under the conditions specified above, it can be obtained that the oxygen transfer rate OTR is equal in reactors of different sizes; According to the existing formula for culturing cells, the oxygen transfer rate OTR = mass transfer coefficient Kla* (saturated oxygen concentration in the air C* - actual oxygen concentration in the culture system C). According to actual operations, in the same environmental space and for the same type of cell culture, C* is the saturated oxygen concentration (a constant value) and C is the actual value of dissolved oxygen (a constant value). Therefore, when the oxygen transfer rate OTR is equal, the mass transfer coefficient Kla is equal, and it can be concluded that the mass transfer coefficient Kla in reactors of different sizes is equal.

[0019] The data parameters that change with the size of the reactor include aeration rate and stirring speed. The aeration rate mainly depends on the size and method of the reactor's ventilation aperture, as well as the size and form of the stirring paddle. The stirring speed is controlled according to the input power of the stirring paddle of reactors of different specifications. Therefore, to measure the Kla mass transfer coefficient at different speeds and aeration rates, it is necessary to calculate the value of the mass transfer coefficient Kla under different conditions based on the changes in dissolved oxygen over time at different speeds, different cell culture medium volumes, and different bottom air aeration rates during the cultivation process of reactors of different sizes. A mathematical relationship is established based on Kla, apparent ventilation rate Vs, and volume power ratio P / V; A mathematical model was further established, and several groups of data were measured by selecting different stirring speeds and air bottom ventilation rates, and a mathematical relationship between Kla and the apparent ventilation rate Vs and the volume power ratio P / V was established; The specific steps include: 1. Add a fixed volume of cell culture medium to the corresponding basic reactor or amplified reactor, and calibrate DO according to the set process, where DO is the dissolved oxygen concentration (the dissolved oxygen concentration in the fermentation broth (abbreviated as DO) is a crucial parameter in the aerobic microbial fermentation and cell culture process); 2. Close the air intake, open the nitrogen intake, blow the DO to below 5%, and close the nitrogen intake; 3. Arbitrarily set n groups of stirring speed N and ventilation rate Q to obtain the first data group (Ni, Qi), i=1,2,……,n; 4. The stirring speed and ventilation rate of the reactor are set in sequence according to the first data group; after each set of data is set, parameter monitoring is turned on to record the change value of DO over time, and when DO is greater than 95%, the experiment is ended, and then the volume of the cell culture medium is balanced and DO is calibrated, and then the next stirring speed and ventilation rate data group setting is performed, and so on, until n groups of DO change values ​​over time under different stirring speeds N and ventilation rates Q are obtained; 5. According to the change values ​​of n groups of DO over time, perform n linear fittings respectively to obtain n slope parameters, which are the mass transfer coefficients Kla, and then obtain n groups of mass transfer coefficients Kla; 6. Convert n groups of the first data sets (Ni, Qi) into the impeller input power P and the apparent ventilation rate Vs by using formula (1) and formula (2), and obtain n groups of the second data sets (Pi, Vsi), i=1,2,...,n; The formula (1) is: ; Where Np is the power factor of the stirring paddle, N is the stirring speed, is the diameter of the stirring paddle, and ρ is the liquid density; The formula (2) is: ; Where Vs is the apparent ventilation rate, Q is the ventilation rate, and D is the tank radius; 7. According to the n second data sets (Pi, Vsi) and the cell culture medium volume V, substitute into the mathematical relationship model Kla=K(P / V) α (Vs) β , and solve for the corresponding reactor's K, α and β, and obtain the relationship model between the mass transfer coefficient Kla and the impeller input power P and the apparent ventilation rate Vs; The specific operating principle is as follows. It can be seen that the DO inside the reactor is related to the internal ventilation rate and the input power of the stirring paddle. Through the above steps, after the volume V of the cell culture medium to be cultured is added, the initial state is detected, that is, the dissolved oxygen baseline values ​​of the water and air in the culture medium without any external force, and then the air entry is closed, and the internal dissolved oxygen baseline value is continuously reduced by inputting nitrogen to below 5%. After setting different stirring speeds and ventilation rates, oxygen input is re-provided, and the changes in the dissolved oxygen baseline over time are detected in real time until the dissolved oxygen baseline value is greater than 95%, and then the ventilation and stirring are terminated. This process is used to obtain the dissolved oxygen rate brought about by different stirring speeds and ventilation rates as time changes.

[0020] According to the above steps, the K, α and β values ​​corresponding to the basic reactor and the expanded reactor are calculated respectively. Specifically, the constant values ​​K, α and β of the corresponding reactor are calculated according to the reactors of different specifications; Finally, a prediction is made to predict the daily ventilation volume Qgas of the expanded reactor during the culture cycle based on the historical data of the basic reactor. The specific prediction methods include: According to the mass transfer coefficients of the basic reactor Kla and the expanded reactor Kla* being equal, the formula (3) is obtained as follows: K 0 (P 0 / V 0 ) α0 (Vs 0 ) β0 =K 1 (P 1 / V 1 ) α1 (Vs 1 ) β1 , where P 0 、V 0 , Vs 0 , are the impeller input power, cell culture medium volume, and superficial aeration rate of the basic reactor, respectively. 1 、V 1 , Vs 1 , are the impeller input power, cell culture medium volume, and superficial aeration rate of the expanded reactor, respectively; The stirring speed N of the expansion reactor is set according to historical data; the speed of the expansion reactor is set specifically according to the maximum ventilation volume of the predicted model scale, so that it does not exceed the maximum ventilation value of the cultured cells, or the ventilation volume is controlled within the optimal range of cell culture, or it is adjusted according to the set ventilation rate Q and the corresponding stirring speed N is calculated.

[0021] In a known scale-up reactor, the volume of cell culture medium required to culture cells is V 1When the scale-up reactor Kla is equal to the base reactor Kla, the apparent ventilation rate Vs of the scale-up reactor can be calculated according to formula (3): 1 , so that the daily ventilation volume Qgas can be calculated; Determine whether the corresponding daily ventilation volume Qgas meets the process requirements. If it does, it means that the input power P of the stirring paddle is 1 , Apparent Ventilation Rate Vs 1 Under the condition of , the expanded reactor has the conditions for normal cell growth; if the ventilation volume Qgas does not meet the process requirements, the parameter stirring paddle input power P is adjusted according to the formula (3) 1 Recalculate the apparent ventilation rate Vs 1 , so that the recalculated ventilation volume Qgas is within the process requirements. Specifically, the daily ventilation volume Qgas can be calculated by formula (4); The formula (4) is: ; Where Qgas is the ventilation volume and Ar is the surface area of ​​the culture medium in the reactor.

[0022] The specific implementation case is explained by taking the basic reactor 10L and the expanded reactor 200L as examples: According to the cells to be cultured, a 16-day process cycle is established. The steps described in 1-7 above are followed, and the different stirring speeds N and aeration rates Q are substituted into the formula: Kla=K(P / V) α (Vs) β Calculate and get the constant term K of 10L 0 , α 0 and β 0 The value of The specific calculation results of the 10L constant term are shown in Table (2): Table (2) shows the K of 10L reactor 0 , α 0 and β 0 The constant value of :

[0023] At the same time, according to the different stirring speeds and ventilation rates every day, the DO values ​​of the day are fitted with time, that is, the Kla values ​​of the day, as shown in the following table (3). Figure 1 , is a linear graph of one set of DO values ​​changing with time; Table (3) corresponds to the Kla value for each day:

[0024] The constant term K corresponding to the 200L scale-up reactor was calculated. 1 , α 1 and β 1 The value of The specific calculation results of the 200L constant term value are shown in Table (4); Table (4) shows the K of the 200L reactor. 1 , α 1 and β 1 The constant value of :

[0025] 10L basic reactor Kla=K 0 (P 0 / V 0 ) α0 (Vs 0 ) β0 , 200L expansion reactor Kla*=K 1 (P 1 / V 1 ) α1 (Vs 1 ) β1 , Since the mass transfer coefficients of the 10L basic reactor Kla and the 200L expanded reactor Kla* are equal, that is, Kla= Kla*, That is, K 0 (P 0 / V 0 ) α0 (Vs 0 ) β0 =K 1 (P 1 / V 1 ) α1 (Vs 1 ) β1 ; According to the historical data, the stirring speed N corresponding to the 200L expansion reactor is set (the corresponding stirring paddle input power P can be obtained according to Formula 1), and the Kla corresponding to all days is substituted into it. 1 , α 1 , β 1 、V 1 In the case of , the apparent ventilation rate Vs of the corresponding days can be calculated 1 , and then the apparent ventilation rate Vs is obtained 1 The daily ventilation volume Qgas is calculated, that is, the prediction result of the 200L expansion reactor is completed. Finally, the prediction result is judged to determine whether the corresponding daily ventilation volume Qgas meets the process requirements. If it does, it means that the input power P of the stirring paddle is 1, Apparent Ventilation Rate Vs 1 Under the condition of , the expanded reactor has the conditions for normal cell growth; if the ventilation volume Qgas does not meet the process requirements, the parameter stirring paddle input power P is adjusted according to the formula (3) 1 Recalculate the apparent ventilation rate Vs 1 , so that the recalculated ventilation volume Qgas is within the process requirements; The specific calculation results are as follows: the apparent ventilation rate Vs of the 200L scale-up reactor for the corresponding number of days 1 ; The aeration rate of the 200L reactor predicted by the 10L reactor is shown in Table (5); Table (5) predicts the ventilation results of the 200L reactor using the 10L reactor: ;

[0026] The stirring speed can be set in advance, and the preferred stirring speed is set according to the size of the cell volume and the size of the reactor. The volume of the culture medium for the cultured cells is known, so the input power P of the stirring paddle can be calculated.

[0027] Verification of the above-mentioned expanded ventilation volume: Based on the ventilation volume calculated based on the 10L basic reactor process and the assumption that the Kla of large and small scale bioreactors is equal, the ventilation volume prediction of the above 200L scale-up reactor is established and compared with the actual ventilation volume in the 200L production process to verify the reliability of the model. For details, please refer to the attached Figure 2 The line table of predicted ventilation volume and actual ventilation volume is compared, where blue is the actual ventilation volume of 200L and red is the predicted ventilation volume of 200L. The data obtained in this figure are the instantaneous ventilation volume at the same time. By comparing the instantaneous ventilation volume, it can be seen that the difference between the predicted ventilation volume and the actual ventilation volume is very small, and there is a high degree of consistency, which proves that the ventilation volume prediction is successful and the cells can be cultured normally in the expanded reactor.

[0028] The benefits of the present invention are that the technical solution is based on the premise of maintaining a constant dissolved oxygen value in the culture system. By measuring the mass transfer correlation coefficient of the reactor, the ventilation value at any rotation speed can be predicted under the premise that the tank structure remains unchanged, thereby balancing the shear force damage of stirring and ventilation. The ventilation level after scale-up can be predicted based on the oxygen consumption of the pilot-scale reactor, and a comprehensive consideration is made between the shear forces brought by stirring and ventilation, thereby improving the success rate of scale-up production.

[0029] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for predicting the ventilation rate of a large-scale production reactor, characterized in that: It includes the following steps: S1. Variable parameter setting: The variable parameters include the volume of the cell culture medium V, the input power P of the stirring paddle of the reactor, and the apparent ventilation rate Vs; S2. Construct a mathematical relationship model between the mass transfer coefficient Kla in the reactor and the variable parameters: ； Where K, α and β are characteristic constants of the reactor; S3, collecting the operating data of the basic reactor and the amplified reactor, and solving the characteristic constants (K0, α0, β0) and (K1, α1, β1) in the relationship model of the basic reactor and the amplified reactor respectively; The method for solving the characteristic constant includes: S31, adding a fixed volume of cell culture medium into the reactor, and calibrating DO according to the set process, wherein DO is the dissolved oxygen concentration; S32, close the air intake, open the nitrogen intake, blow the DO to below 5%, and close the nitrogen intake; S33, arbitrarily set n groups of stirring speeds N and ventilation rates Q to obtain a first data group (Ni, Qi), i=1, 2, ..., n; S34, setting the stirring speed and ventilation rate of the reactor in sequence according to the first data group; after each set of data, start parameter monitoring, record the change value of DO over time, and when DO is greater than 95%, end the experiment, then balance the volume of the cell culture medium and calibrate DO, and then set the next stirring speed and ventilation rate data group, and so on, until n groups of DO change values ​​over time under different stirring speeds N and ventilation rates Q are obtained; S35, performing n linear fittings according to the change values ​​of n groups of DO over time, respectively, to obtain n slope parameters, the slope parameter being the mass transfer coefficient Kla, and further obtaining n groups of mass transfer coefficients Kla; S36, converting n groups of the first data groups (Ni, Qi) into the impeller input power P and the apparent ventilation rate Vs by formula (1) and formula (2), to obtain n groups of second data groups (Pi, Vsi), i=1, 2, ..., n; The formula (1) is: ； Where Np is the power factor of the stirring paddle, N is the stirring speed, is the diameter of the stirring paddle, and ρ is the liquid density; The formula (2) is: ； Where Vs is the apparent ventilation rate, Q is the ventilation rate, and D is the tank radius; S37, according to the n second data sets (Pi, Vsi) and the volume of the cell culture medium V, substitute into the mathematical relationship model Kla=K(P / V) α (Vs) β , and solve for the corresponding reactor's K, α and β, and obtain the relationship model between the mass transfer coefficient Kla and the impeller input power P and the apparent ventilation rate Vs; S4. Predict the daily ventilation volume Qgas of the expanded reactor during the culture period based on the historical data of the basic reactor.

2. The method for predicting the ventilation volume of a large-scale production reactor according to claim 1, characterized in that: The prediction method in step S4 includes: S41, according to the mass transfer coefficient Kla of the basic reactor and the mass transfer coefficient Kla* of the expanded reactor being equal, That is, formula (3) is: K0(P0 / V0) α0 (Vs0) β0 =K1(P1 / V1) α1 (Vs1) β1 , where P0, V0, and Vs0 are the impeller input power, cell culture medium volume, and apparent aeration rate of the basic reactor, respectively; and P1, V1, and Vs1 are the impeller input power, cell culture medium volume, and apparent aeration rate of the expanded reactor, respectively; S42, setting the stirring speed N of the enlarged reactor; S43, obtaining historical data of normal production of the basic reactor, obtaining parameter data P0, V0, Vs0 for each day during the culture cycle, and calculating the apparent ventilation rate Vs1 of the expanded reactor during each day during the culture cycle according to formula (3) when the cell culture medium volume V1, stirring speed N and stirring paddle input power P1 of the expanded reactor are known, and then calculating the daily ventilation volume Qgas according to formula (4); The formula (4) is: ； Wherein, Qgas is the ventilation volume, and Ar is the surface area of ​​the culture medium in the reactor.

3. A method for predicting the ventilation volume of a large-scale production reactor according to claim 2, characterized in that: In step S42, the stirring speed N of the expanded reactor is set according to the culture cycle of historical data, and is specifically set according to the maximum ventilation volume of the predicted model scale so that it does not exceed the maximum ventilation value, or the ventilation volume is controlled within the optimal range for cell culture, or adjusted according to the set ventilation rate Q and the corresponding stirring speed N is calculated.

4. The method for predicting the ventilation volume of a large-scale production reactor according to claim 1, characterized in that: In step S4, the basic reactor calculates the Kla corresponding to different days according to the culture cycle, that is, the 1st day, the 2nd day, ..., the Nth day, and obtains Kla1, Kla2, ..., KlaN accordingly. The Kla value of the basic reactor corresponding to the daily ventilation volume and the set stirring speed N of the expanded reactor are calculated as needed to obtain the apparent ventilation rate Vs1 of the expanded reactor for the corresponding days.

5. A production process for large-scale production of reactors, characterized in that: The method comprises the following steps: predicting the ventilation volume Qgas of each day in the culture period after expansion according to the ventilation volume prediction method of large-scale production reactor as claimed in claim 2, judging whether the ventilation volume Qgas meets the process requirements, and if so, it means that under the conditions of the currently set stirring blade input power P1 and the apparent ventilation rate Vs1, the expanded reactor has the conditions for normal cell growth; if the ventilation volume Qgas does not meet the process requirements, then according to formula (3): K0(P0 / V0) α0 (Vs0) β0 =K1(P1 / V1) α1 (Vs1) β1 The parameter agitator input power P1 is adjusted to recalculate the apparent ventilation rate Vs1, and then the ventilation volume Qgas is recalculated, so that the recalculated ventilation volume Qgas is within the process requirement range.

Images