Bioreactor process optimization system and method thereof

Abstract

The invention discloses a bioreactor process optimization system and method, and relates to the technical field of bioreactors. Comprising the following steps: collecting historical reaction process data of a bioreactor, and obtaining real-time reaction process data of the bioreactor; analyzing the historical reaction process data to obtain different influence results of different adjustment modes on the biological reaction process; establishing a cell culture model according to a corresponding relation obtained by different influence results of different adjustment modes on the biological reaction process; the real-time reaction process data is obtained, the biological culture process is influenced by adjusting the bioreactor, the specific adjustment mode is obtained through the real-time reaction process data and the required adjustment result, the required adjustment result can be obtained after the reaction in the adjustment mode obtained through backstepping, hysteresis is avoided, materials are saved, and the efficiency is improved. And the problem of poor subsequent biological reaction caused by excessive material addition, excessive design parameter adjustment and poor subsequent biological reaction due to hysteresis is also avoided.

Description

Technical Field

[0001] The present application relates to the technical field of bioreactors, and in particular to a bioreactor process optimization system and method thereof. Background Art

[0002] Bioreactors are core process equipment that support the development and production of biological products. They utilize the biological functions of organisms, including microorganisms and cells, to obtain target products through biochemical reactions or the growth and metabolism of the organisms themselves in vitro. Bioreactors are widely used in many fields such as biomedicine, food, and modern agriculture. Especially in the field of biomedicine, they have become an indispensable tool for the production of high-end biological preparations such as antibodies, vaccines, and recombinant proteins. With the rapid development of high-tech technologies such as global biomedicine, bio-based new materials, and biomass energy, the importance of bioreactors has become increasingly prominent, and market demand has shown a blowout growth trend.

[0003] As the technical requirements for bioreactors become increasingly higher, the dynamic adjustment of current bioreactors in large-scale biological reaction processes is prone to lag. The larger the culture scale, the more obvious the lag, resulting in inaccurate material addition in the later stage, material waste, and poor biological reaction. Therefore, there is an urgent need for a bioreactor process optimization system and method. According to the highly nonlinear, time-varying and uncertain characteristics of the biological culture process, the biological reaction process can be simulated, predicted and dynamically adjusted from multiple directions to improve the competitiveness of the entire industrial chain. Summary of the invention

[0004] In view of the problem that the dynamic adjustment of the current bioreactor in the biological reaction process is prone to lag, resulting in inaccurate material addition in the later stage, this application provides a bioreactor process optimization system and method, which uses physical models, online sensor acquisition, operation history and other data to build a simulation model to simulate, predict and dynamically adjust the biological reaction process. The specific technical solution is as follows: In a first aspect of the present application, a bioreactor process optimization system is provided, comprising: The data acquisition module collects the historical reaction process data of the bioreactor in the database, obtains the real-time reaction process data of the bioreactor through the sensor, and saves it; A data analysis module analyzes the historical reaction process data to obtain different effects of different adjustment methods on the biological reaction process; A simulation module, which establishes a cell culture model according to the corresponding relationship obtained by the data analysis module, and adds the biological type and growth stage as independent variables to the cell culture model; The dynamic adjustment module obtains the real-time reaction process data and affects the biological culture process by adjusting the bioreactor. The specific adjustment method is obtained by inputting the real-time reaction process data and the required adjustment results into the simulation module.

[0005] In one embodiment of the present application, the dynamic adjustment module includes a material adjustment submodule and an environment adjustment submodule. The material adjustment submodule adjusts the biological culture process by inputting materials into the bioreactor, and the materials include pH regulators, gases, biological nutrients, and catalytic enzymes; the environment adjustment submodule adjusts the biological culture process by adjusting the reaction environment of the bioreactor, and the environment includes temperature, pressure, and speed data in the bioreactor; In the data analysis module, the different adjustment methods include inputting different materials and / or adjusting environmental data, and the real-time reaction process data includes biological types and growth stages, and different growth stages of different biological types are obtained, and the impact results on the biological reaction process under different adjustment methods.

[0006] In one embodiment of the present application, a design adjustment module is also included for adjusting the design parameters of the bioreactor before the biological reaction, including the tank size, impeller type, impeller combination form, impeller size, ventilation method and baffle size. Different design parameters correspond to different fluid dynamic simulation results through the simulation model, and the design parameters are associated with cell culture performance indicators. The design parameters corresponding to the highest cell culture performance indicators are used to adjust the bioreactor.

[0007] In one embodiment of the present application, a parameter adjustment module is also included. After the dynamic adjustment module completes the adjustment of the current biological culture process, the real-time reaction process data is saved as the historical reaction process data, and the adjustment parameter indicators in the reaction process are learned and updated by inputting multiple batches of the historical reaction process data into the neural network model.

[0008] In one embodiment of the present application, the dynamic adjustment module also includes an early warning adjustment submodule, which obtains the type of reaction data collected by the sensor and sets corresponding thresholds for various reaction data of different biological types at different growth stages. The sensor obtains real-time reaction process data of the bioreactor and obtains a change curve of the reaction data. When the change curve shows an abnormal trend, an early warning operation is performed; when the reaction process data reaches the corresponding threshold, a corresponding adjustment operation is performed according to the threshold reached by the reaction process data.

[0009] In one embodiment of the present application, the biological types include single biological types and composite biological types. The single biological type refers to the presence of one organism in the biological reaction process, and the composite biological type refers to the presence of multiple organisms in the biological reaction process; the growth stages include the growth stages of non-reproductive organisms and reproductive organisms. The non-reproductive organisms refer to the organisms in the biological reaction process that do not reproduce each other, and the corresponding growth stages include early, middle and late stages; the reproductive organisms refer to the organisms in the biological reaction process that reproduce each other, and the corresponding growth stages include the early stage of the group, the group reproduction period and the group circulation period.

[0010] In one embodiment of the present application, when the biological type is a composite organism, the historical reaction process data is analyzed by the data analysis module to obtain the optimal proportion of each organism in the bioreactor, and the proportion of each organism is maintained by the dynamic adjustment module during the biological reaction process.

[0011] In one embodiment of the present application, the data acquisition module also includes an external data sub-module, which collects and identifies the operations of the staff through the sensor, screens and classifies standard operations and erroneous operations, and trains the recognition models respectively. The trained recognition models identify the operations of the staff, and when the operations of the staff are inconsistent with the standard operations, or are consistent with the erroneous operations, corresponding prompts and processing are performed.

[0012] In one embodiment of the present application, the external data submodule obtains the adjustment method issued by the dynamic adjustment module. When the staff adjusts the biological reaction process according to the adjustment method, the recognition model matches the corresponding standard operation according to the adjustment method, and judges whether the staff's adjustment operation is correct according to the standard operation.

[0013] In one embodiment of the present application, the recognition model further includes identifying the identity of the staff, establishing an operation database of all staff, marking operations that are inconsistent with the standard operation or consistent with the erroneous operation, and prompting the staff whose number of markings is greater than a threshold before performing subsequent operations; The erroneous operations that occurred multiple times by the staff are retrieved. If it is identified that the previous operation of the erroneous operation is the same operation, the staff will be prompted when the same operation occurs subsequently.

[0014] In a second aspect of the present application, a bioreactor process optimization method is provided, comprising: Collect historical reaction process data of the bioreactor in the database, obtain real-time reaction process data of the bioreactor through sensors, and save them; Analyze the historical reaction process data to obtain different effects of different adjustment methods on the biological reaction process; Establishing a cell culture model according to the corresponding relationship obtained from the different effects of different adjustment methods on the biological reaction process, and adding the biological type and growth stage as independent variables to the cell culture model; The real-time reaction process data is obtained, and the biological culture process is affected by adjusting the bioreactor. The specific adjustment method is obtained through the real-time reaction process data and the required adjustment result.

[0015] In one embodiment of the present application, the biological culture process is adjusted by feeding materials into the bioreactor, wherein the materials include a pH regulator, a gas, biological nutrients, and a catalytic enzyme; the biological culture process is adjusted by adjusting the reaction environment of the bioreactor, wherein the environment includes temperature, pressure, and rotation speed data in the bioreactor; The different adjustment methods include inputting different materials and / or adjusting environmental data. The real-time reaction process data includes biological types and growth stages, and different growth stages of different biological types are obtained, and the impact results on the biological reaction process under different adjustment methods.

[0016] In one embodiment of the present application, the design parameters of the bioreactor, including the tank size, impeller type, impeller combination, impeller size, ventilation method and baffle size, are adjusted before the biological reaction. Different design parameters correspond to different fluid dynamic simulation results through the simulation model, and the design parameters are associated with cell culture performance indicators. The design parameters corresponding to the highest cell culture performance indicators are used to adjust the bioreactor.

[0017] In one embodiment of the present application, after the current biological culture process is adjusted, the real-time reaction process data is saved as the historical reaction process data, and the adjustment parameter indicators in the reaction process are learned and updated by inputting multiple batches of the historical reaction process data into the neural network model.

[0018] In one embodiment of the present application, the type of reaction data collected by the sensor is obtained, and thresholds of various reaction data of different biological types at different growth stages are set accordingly. The sensor obtains real-time reaction process data of the bioreactor and obtains a change curve of the reaction data. When the change curve shows an abnormal trend, an early warning operation is performed; when the reaction process data reaches the corresponding threshold, a corresponding adjustment mode operation is performed according to the threshold reached by the reaction process data.

[0019] In one embodiment of the present application, the biological types include single biological types and composite biological types. The single biological type refers to the presence of one organism in the biological reaction process, and the composite biological type refers to the presence of multiple organisms in the biological reaction process; the growth stages include the growth stages of non-reproductive organisms and reproductive organisms. The non-reproductive organisms refer to the organisms in the biological reaction process that do not reproduce each other, and the corresponding growth stages include early, middle and late stages; the reproductive organisms refer to the organisms in the biological reaction process that reproduce each other, and the corresponding growth stages include the early stage of the group, the group reproduction period and the group circulation period.

[0020] In one embodiment of the present application, when the biological type is a composite organism, the historical reaction process data is analyzed to obtain the optimal proportion of each organism in the bioreactor, and the proportion of each organism is maintained through continuous adjustment during the biological reaction process.

[0021] In one embodiment of the present application, the sensor is used to collect and identify the staff's operations, and standard operations and erroneous operations are screened and classified. The recognition models are trained respectively, and the trained recognition models are used to identify the staff's operations. When the staff's operations are inconsistent with the standard operations, or are consistent with the erroneous operations, corresponding prompts and processing are performed.

[0022] In one embodiment of the present application, when a staff member performs an adjustment operation on a biological reaction process according to the adjustment method, the recognition model matches the corresponding standard operation according to the adjustment method, and determines whether the adjustment operation of the staff member is correct according to the standard operation.

[0023] In one embodiment of the present application, the recognition model further includes identifying the identity of the staff, establishing an operation database of all staff, marking operations that are inconsistent with the standard operation or consistent with the erroneous operation, and prompting the staff whose number of markings is greater than a threshold before performing subsequent operations; The erroneous operations that occurred multiple times by the staff are retrieved. If it is identified that the previous operation of the erroneous operation is the same operation, the staff will be prompted when the same operation occurs subsequently.

[0024] This application has the following beneficial effects: 1. By collecting and analyzing the historical reaction process data, the cell culture model is established, and the corresponding adjustment results obtained by different adjustment methods are obtained through simulation. Therefore, the real-time reaction process data and the required adjustment results can be input into the cell culture model to infer the specific adjustment method. The adjustment method obtained by inference can obtain the required adjustment result after the reaction, avoiding the hysteresis caused by the sensor data collection, the time required for the reaction, etc., saving materials, and also avoiding the problem of excessive material addition and excessive adjustment of design parameters due to the hysteresis of the adjustment method, resulting in poor subsequent biological reactions. Further, the simulation module realizes all-round real-time monitoring, dynamic simulation and accurate prediction of the system state through the digital twin by constructing a real-time virtual model of the physical system, that is, the cell culture model. Through real-time data collection by sensors, it can reflect the operating state of the physical system and provide an accurate data basis for subsequent analysis. The dynamic adjustment module is realized through artificial neural networks, with the real-time data of digital twins as input, to model, analyze and optimize the decision of complex systems. Through deep learning technology, it adaptively mines the system operation rules, predicts future trends, and proposes an optimization control scheme to realize intelligent management of the system. The main effects achieved after the combination include combining real-time data with learning capabilities, so that the system can predict possible problems or abnormalities in advance; and improving the operating efficiency and stability of the system through intelligent optimization.

[0025] 2. Consider adjusting the biological reaction process by adding materials, and also consider adjusting the biological reaction process by adjusting the design parameters. With the support of a large amount of data, different growth stages of different biological types are obtained. Under different adjustment methods, the impact on the biological reaction process, different adjustment methods include adjustment of materials and / or design parameters, which may be the adjustment of a single material, or the adjustment of a single design parameter, or the impact of both on the reaction process; further, through the growth and metabolism of multiple batches of cells, the amount of data is expanded. After each adjustment by the dynamic adjustment module, the actual adjustment result is compared with the adjustment result estimated by the simulation module. According to the difference between the two, the cell culture model learns and updates the adjustment parameter indicators in the reaction process. At the same time, each batch of data only accounts for a part of the adjustment parameters to avoid occasional parameter errors, gradually update the parameters of the current bioreactor, and predict more accurate adjustment methods for the subsequent simulation module.

[0026] 3. In the historical reaction process data, the data before the abnormality occurs are selected for model training. When the data identical to the data before the abnormality appears in the real-time reaction process data, it is regarded as a trend of abnormality. At this time, an early warning operation is performed, and the staff makes adjustments in advance to avoid damage to the biological activity when the abnormal state is reached. Further, if the data has reached the abnormal threshold, it means that the staff has failed to perform effective adjustment operations. The bioreactor is automatically set to perform corresponding adjustment operations according to the threshold reached by the reaction process data. For example, when the oxygen content exceeds the minimum amount required for biological survival, oxygen is correspondingly introduced into the bioreactor. Through the hierarchical setting of early warning followed by automatic operation, on the one hand, the number of abnormalities in the biological reaction process is effectively reduced to ensure biological activity. On the other hand, it also gives the staff time to react to the adjustment operation. At the same time, the staff's experience and actual situation are used to judge whether an abnormal situation will occur, thereby avoiding unnecessary adjustment operations and saving resources.

[0027] 4. The operation of the staff is identified through the recognition model. When the operation of the staff is inconsistent with the standard operation, or is consistent with the erroneous operation, corresponding prompts and processing are performed. For example, when depressurizing the bioreactor, the standard operation is to first check whether the reaction is completed, then stop stirring and gas supply, and finally slowly open the pressure relief valve. When the staff releases the pressure, they follow the standard operating procedures for comparison; if an erroneous operation occurs, a prompt is given, and the correct operation is informed, further including the position of each operating component, such as identifying the position of the pressure relief valve, determining whether the pressure relief valve is slowly opened by the staff, and effectively standardizing the staff's operation to avoid the negative impact of erroneous operation on the biological reaction; further, when the dynamic adjustment module gives the adjustment method, the standard operation corresponding to the adjustment method is called to compare whether the actual operation performed by the staff is correct. The comparison process is more targeted and the operation accuracy is higher.

[0028] 5. Establish an operation database for all staff members and record the operations of each staff member. When a staff member makes mistakes in performing the same operation multiple times, it means that the staff member is prone to make habitual mistakes in this operation. In this case, the staff member will be reminded in advance when performing the operation in the future. Furthermore, the staff member's operations will be analyzed to find out the operations that each staff member is prone to making mistakes in, and corresponding adjustments will be made to reduce erroneous operations. In addition, it will be analyzed whether the same operation occurred before the erroneous operation, thereby analyzing the pattern of the staff member's erroneous operations, and prompting when the same operation occurs in the future to prevent the staff from making erroneous operations in the future based on habit. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0030] Figure 1 A schematic diagram of the electronic device structure of the hardware operating environment involved in an embodiment of the present application.

[0031] Figure 2 It is a functional module schematic diagram of a bioreactor process optimization system provided in an embodiment of the present application.

[0032] Figure 3 This is a module construction diagram of a bioreactor process optimization system provided in an embodiment of the present application.

[0033] Figure 4 It is a step flow chart of a bioreactor process optimization method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0035] The solution of the present application is further described below in conjunction with the accompanying drawings.

[0036] like Figure 1 As shown, the electronic device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM) memory, or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0037] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0038] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module and a data storage module.

[0039] exist Figure 1 In the electronic device shown, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the electronic device of the present invention can be set in the electronic device, and the electronic device calls a bioreactor process optimization system stored in the data storage module in the memory 1005 through the processor 1001, and executes a bioreactor process optimization method provided in an embodiment of the present application.

[0040] Based on the aforementioned hardware operating environment and system architecture, in the first aspect of the present application, refer to Figure 2 and Figure 3 As shown, a bioreactor process optimization system is provided, comprising: The data acquisition module collects the historical reaction process data of the bioreactor in the database, obtains the real-time reaction process data of the bioreactor through the sensor, and saves it; It should be noted that the historical reaction process data is data of a bioreaction that has been carried out, and the database includes a local database and a cloud database. The local database stores the historical reaction process data of the local bioreactor, and further enriches the data volume by collecting the historical reaction process data of bioreactors in other regions in the cloud database; the sensors are set according to the data required by the current bioreactor, and specifically include a pH sensor for measuring the pH in the bioreactor, a gas sensor for measuring the content of various gases in the bioreactor, an electrochemical sensor for detecting nutrients and / or metabolites such as glucose, glutamine or lactic acid, as well as a temperature sensor, a pressure sensor, and a bio-enzyme sensor, which together collect the real-time reaction process data in the bioreactor; A data analysis module analyzes the historical reaction process data to obtain different effects of different adjustment methods on the biological reaction process; A simulation module, which establishes a cell culture model according to the corresponding relationship obtained by the data analysis module, and adds the biological type and growth stage as independent variables to the cell culture model; It should be noted that the specific analysis methods include parameter association, analyzing the relationship between parameters and biological reaction processes, and statistical analysis, comparing the differences in parameters under different adjustment strategies through two-sample tests, evaluating whether the differences are statistically significant, and finally establishing a mathematical model based on historical data, that is, a cell culture model, to simulate the effects of different adjustment methods on the biological reaction process; The dynamic adjustment module obtains the real-time reaction process data and affects the biological culture process by adjusting the bioreactor. The specific adjustment method is obtained by inputting the real-time reaction process data and the required adjustment results into the simulation module.

[0041] In this embodiment, the cell culture model is established by collecting and analyzing the historical reaction process data, and the corresponding adjustment results obtained by different adjustment methods are obtained through simulation. Therefore, the real-time reaction process data and the required adjustment results can be input into the cell culture model to infer the specific adjustment method required. The adjustment method obtained by reverse inference can obtain the required adjustment result after reaction, avoiding the lag caused by reasons such as sensor data collection and reaction consumption time, saving materials, and also avoiding the problem of excessive material addition and excessive adjustment of design parameters due to the lag of the adjustment method, resulting in poor subsequent biological response.

[0042] Furthermore, the simulation module uses the digital twin to build a real-time virtual model of the physical system, that is, a cell culture model, to achieve all-round real-time monitoring, dynamic simulation and accurate prediction of the system status. Through real-time data collection by sensors, etc., it can reflect the operating status of the physical system and provide an accurate data basis for subsequent analysis. The dynamic adjustment module is implemented through an artificial neural network, which uses the real-time data of the digital twin as input to model, analyze and optimize decisions of complex systems. Through deep learning technology, it adaptively mines the operating laws of the system, predicts future trends, and proposes optimized control plans to achieve intelligent management of the system. The effects achieved after the combination are mainly that the system can predict possible problems or anomalies in advance by combining real-time data with learning capabilities; and improve the operating efficiency and stability of the system through intelligent optimization.

[0043] In one embodiment of the present application, the dynamic adjustment module includes a material adjustment submodule and an environment adjustment submodule. The material adjustment submodule adjusts the biological culture process by inputting materials into the bioreactor, and the materials include pH regulators, gases, biological nutrients, and catalytic enzymes; the environment adjustment submodule adjusts the biological culture process by adjusting the reaction environment of the bioreactor, and the environment includes temperature, pressure, and speed data in the bioreactor; In the data analysis module, the different adjustment methods include inputting different materials and / or adjusting environmental data, and the real-time reaction process data includes biological types and growth stages, and different growth stages of different biological types are obtained, and the impact results on the biological reaction process under different adjustment methods.

[0044] It should be noted that the adjustment method includes adding materials in the bioreactor process through the material adjustment submodule, and changing the bioreactor environment, such as pH value and oxygen content, through the added materials; and also includes changing the bioreactor environment by adjusting the design parameters, such as adjusting the temperature and pressure, and controlling the rotation speed to change the flow field dynamics in the bioreactor, thereby changing the material mixing time, mixing method, and gas content data; In one embodiment of the present application, a design adjustment module is also included for adjusting the design parameters of the bioreactor before the biological reaction, including the tank size, impeller type, impeller combination form, impeller size, ventilation method and baffle size. Different design parameters correspond to different fluid dynamic simulation results through the simulation model, and the design parameters are associated with cell culture performance indicators. The design parameters corresponding to the highest cell culture performance indicators are used to adjust the bioreactor.

[0045] It should be noted that, considering that the adjustment of design parameters of the bioreactor during the reaction process may cause contamination to the biological reaction, before the reaction, a simulation model is established, including an integrated model based on coupled fluid dynamics and cell growth dynamics, and a data-driven method is used to analyze and compare the possible correlations between various variables across scales; specifically, back propagation neural networks, convolutional neural networks, and long short-term memory neural networks are used to convert fluid dynamics simulation results, such as shear force, velocity field, kla, mixing time, and cell growth results, into a coding form that can be recognized by the software, and deep training and learning are carried out to obtain the optimal weights, and then the optimal design parameters and key process parameters are obtained through genetic algorithms; In one embodiment of the present application, a parameter adjustment module is also included. After the dynamic adjustment module completes the adjustment of the current biological culture process, the real-time reaction process data is saved as the historical reaction process data, and the adjustment parameter indicators in the reaction process are learned and updated by inputting multiple batches of the historical reaction process data into the neural network model.

[0046] In this embodiment, it is considered to adjust the biological reaction process by adding materials, and it is also considered to adjust the biological reaction process by adjusting the design parameters. With the support of a large amount of data, different growth stages of different biological types are obtained. Under different adjustment methods, the impact results on the biological reaction process, different adjustment methods include adjustment of materials and / or design parameters, which may be the adjustment of a single material, or the adjustment of a single design parameter, or the impact of both on the reaction process; further, the amount of data is expanded through the growth and metabolism of multiple batches of cells. After each adjustment of the dynamic adjustment module, the actual adjustment result is compared with the adjustment result estimated by the simulation module. According to the difference between the two, the cell culture model learns and updates the adjustment parameter indicators in the reaction process. At the same time, only a part of the data of each batch is used in the adjustment of the parameters to avoid occasional parameter errors, and gradually update the parameters of the current bioreactor, so as to predict a more accurate adjustment method for the subsequent simulation module.

[0047] In one embodiment of the present application, the dynamic adjustment module also includes an early warning adjustment submodule, which obtains the type of reaction data collected by the sensor and sets corresponding thresholds for various reaction data of different biological types at different growth stages. The sensor obtains real-time reaction process data of the bioreactor and obtains a change curve of the reaction data. When the change curve shows an abnormal trend, an early warning operation is performed; when the reaction process data reaches the corresponding threshold, a corresponding adjustment operation is performed according to the threshold reached by the reaction process data.

[0048] In this embodiment, in the historical reaction process data, the data before the abnormality occurs are selected for model training. When the same data as the data before the abnormality appears in the real-time reaction process data, it is regarded as a trend of abnormality. At this time, an early warning operation is performed, and the staff makes adjustments in advance to avoid damage to the biological activity when the abnormal state is reached; further, if the data has reached the abnormal threshold, it means that the staff has failed to perform effective adjustment operations. The bioreactor performs corresponding adjustment operations according to the threshold reached by the reaction process data through automatic settings. For example, when the oxygen content exceeds the minimum amount required for biological survival, oxygen is correspondingly introduced into the bioreactor; through the hierarchical setting of early warning followed by automatic operation, on the one hand, the number of abnormalities in the biological reaction process is effectively reduced to ensure biological activity, and on the other hand, the staff is given reaction time to perform adjustment operations. At the same time, the staff's experience and actual conditions are used to judge whether an abnormal situation will occur, thereby avoiding unnecessary adjustment operations and saving resources.

[0049] In one embodiment of the present application, the biological types include single biological types and composite biological types. The single biological type refers to the presence of one organism in the biological reaction process, and the composite biological type refers to the presence of multiple organisms in the biological reaction process; the growth stages include the growth stages of non-reproductive organisms and reproductive organisms. The non-reproductive organisms refer to the organisms in the biological reaction process that do not reproduce each other, and the corresponding growth stages include an adjustment period, an exponential growth period, a stable period, and a decay period; the reproductive organisms refer to the organisms in the biological reaction process that reproduce each other, and the corresponding growth stages include the initial stage of the group, the group reproduction period, and the group cycle period.

[0050] It should be noted that the biological type in the bioreactor is usually the same organism, and the corresponding cell culture model is established according to the living habits of different organisms and the response to different adjustment methods; however, there are also composite biological types in the bioreactor that react with multiple organisms; in addition, the growth stage includes the growth stage of non-reproductive organisms and reproductive organisms, the non-reproductive organisms do not reproduce during the reaction process, and all organisms are basically in the same growth stage synchronously, and the adjustment methods in different stages are different, for example, the oxygen content, nutrients and biological enzyme requirements are different, and the adjustment amount is also different; reproductive organisms carry out reproductive activities during the reaction process, so there may be multiple growth stages in the bioreactor, and the above-mentioned different stages corresponding to different adjustment methods are no longer applicable. At this time, the entire group is divided into stages, and the initial stage of the group is the biological development stage, all organisms are in a growth stage without reproduction, and sufficient growth nutrients are given; in the group reproduction period, the organisms begin to reproduce, and nutrients that promote reproduction are given; the group cycle period is the stable emergence of new organisms and dead organisms, and the dead organisms need to be cleaned up regularly; In one embodiment of the present application, when the biological type is a composite organism, the historical reaction process data is analyzed by the data analysis module to obtain the optimal proportion of each organism in the bioreactor, and the proportion of each organism is maintained by the dynamic adjustment module during the biological reaction process.

[0051] In this embodiment, when the biological type is a composite organism, the historical reaction process data is analyzed by the data analysis module, and the simulation module is used to simulate and predict the results of biological reactions under different proportions, thereby obtaining the optimal proportion of each organism. In the subsequent reaction process, if the proportion of a certain organism decreases, the suitability of the growth environment of the organism is improved to promote the growth of the organism; correspondingly, if the proportion of a certain organism increases, the suitability of the growth environment of the organism is reduced to inhibit the growth of the organism, thereby achieving a balance in the proportion of the organisms.

[0052] In one embodiment of the present application, the data acquisition module also includes an external data sub-module, which collects and identifies the operations of the staff through the sensor, screens and classifies standard operations and erroneous operations, and trains the recognition models respectively. The trained recognition models identify the operations of the staff, and when the operations of the staff are inconsistent with the standard operations, or are consistent with the erroneous operations, corresponding prompts and processing are performed.

[0053] It should be noted that the process optimization of biological reactions includes adjusting the amount and method of adding materials in the bioreactor, as well as the adjustment of design parameters, so as to change the biological reaction environment by obtaining a better adjustment method, thereby achieving process optimization; it also includes the optimization of operations of staff outside the bioreactor, through more standardized and more accurate execution of the corresponding adjustment method, to optimize possible biological contamination caused by human factors, and abnormal biological reactions caused by improper operation; In one embodiment of the present application, the external data submodule obtains the adjustment method issued by the dynamic adjustment module. When the staff adjusts the biological reaction process according to the adjustment method, the recognition model matches the corresponding standard operation according to the adjustment method, and judges whether the staff's adjustment operation is correct according to the standard operation.

[0054] In this embodiment, the identification model is used to identify the staff's operations. When the staff's operations are inconsistent with the standard operations, or are consistent with the erroneous operations, corresponding prompts and processing are performed. For example, when depressurizing the bioreactor, the standard operation is to first check whether the reaction is completed, then stop stirring and gas supply, and finally slowly open the pressure relief valve. When the staff releases the pressure, they follow the standard operating procedures for comparison; if an erroneous operation occurs, a prompt is given, and the correct operation is informed, further including the position of each operating component, such as identifying the position of the pressure relief valve, determining whether the pressure relief valve is slowly opened by the staff, and effectively regulating the staff's operations to avoid the negative impact of erroneous operations on the biological reaction; further, when the dynamic adjustment module gives the adjustment method, the standard operation corresponding to the adjustment method is called to compare whether the actual operation performed by the staff is correct. The comparison process is more targeted and the operation accuracy is higher.

[0055] In one embodiment of the present application, the recognition model also includes identifying the identity of the staff, establishing an operation database for all staff, marking operations that are inconsistent with the standard operations or consistent with the erroneous operations, and prompting staff members who have been marked more than a threshold number of times before performing subsequent operations.

[0056] In this embodiment, an operation database of all staff members is established to record the operations of each staff member. When a staff member makes mistakes in performing the same operation multiple times, it means that the staff member is prone to habitual errors in the operation. In this case, the staff member is reminded in advance when performing the operation subsequently. Furthermore, the staff member's operations are analyzed to find out the operations that each staff member is prone to making mistakes in, and corresponding adjustments are made to reduce erroneous operations. It is also analyzed whether the same operation occurs before the erroneous operation, thereby analyzing the pattern of the staff member's erroneous operations, and prompting when the same operation occurs subsequently, to prevent the staff from making erroneous operations in the future based on habit.

[0057] In the second aspect of the present application, reference is made to Figure 4 As shown, a bioreactor process optimization method is provided, comprising: Collect historical reaction process data of the bioreactor in the database, obtain real-time reaction process data of the bioreactor through sensors, and save them; Analyze the historical reaction process data to obtain different effects of different adjustment methods on the biological reaction process; Establishing a cell culture model according to the corresponding relationship obtained from the different effects of different adjustment methods on the biological reaction process, and adding the biological type and growth stage as independent variables to the cell culture model; The real-time reaction process data is obtained, and the biological culture process is affected by adjusting the bioreactor. The specific adjustment method is obtained through the real-time reaction process data and the required adjustment result.

[0058] In one embodiment of the present application, the biological culture process is adjusted by feeding materials into the bioreactor, wherein the materials include a pH regulator, a gas, biological nutrients, and a catalytic enzyme; the biological culture process is adjusted by adjusting the reaction environment of the bioreactor, wherein the environment includes temperature, pressure, and rotation speed data in the bioreactor; The different adjustment methods include inputting different materials and / or adjusting environmental data. The real-time reaction process data includes biological types and growth stages, and different growth stages of different biological types are obtained, and the impact results on the biological reaction process under different adjustment methods.

[0059] In one embodiment of the present application, the design parameters of the bioreactor, including the tank size, impeller type, impeller combination, impeller size, ventilation method and baffle size, are adjusted before the biological reaction. Different design parameters correspond to different fluid dynamic simulation results through the simulation model, and the design parameters are associated with cell culture performance indicators. The design parameters corresponding to the highest cell culture performance indicators are used to adjust the bioreactor.

[0060] In one embodiment of the present application, after the current biological culture process is adjusted, the real-time reaction process data is saved as the historical reaction process data, and the adjustment parameter indicators in the reaction process are learned and updated by inputting multiple batches of the historical reaction process data into the neural network model.

[0061] In one embodiment of the present application, the type of reaction data collected by the sensor is obtained, and thresholds of various reaction data of different biological types at different growth stages are set accordingly. The sensor obtains real-time reaction process data of the bioreactor and obtains a change curve of the reaction data. When the change curve shows an abnormal trend, an early warning operation is performed; when the reaction process data reaches the corresponding threshold, a corresponding adjustment mode operation is performed according to the threshold reached by the reaction process data.

[0062] In one embodiment of the present application, the biological types include single biological types and composite biological types. The single biological type refers to the presence of one organism in the biological reaction process, and the composite biological type refers to the presence of multiple organisms in the biological reaction process; the growth stages include the growth stages of non-reproductive organisms and reproductive organisms. The non-reproductive organisms refer to the organisms in the biological reaction process that do not reproduce each other, and the corresponding growth stages include early, middle and late stages; the reproductive organisms refer to the organisms in the biological reaction process that reproduce each other, and the corresponding growth stages include the early stage of the group, the group reproduction period and the group circulation period.

[0063] In one embodiment of the present application, when the biological type is a composite organism, the historical reaction process data is analyzed to obtain the optimal proportion of each organism in the bioreactor, and the proportion of each organism is maintained through continuous adjustment during the biological reaction process.

[0064] In one embodiment of the present application, the sensor is used to collect and identify the staff's operations, and standard operations and erroneous operations are screened and classified. The recognition models are trained respectively, and the trained recognition models are used to identify the staff's operations. When the staff's operations are inconsistent with the standard operations, or are consistent with the erroneous operations, corresponding prompts and processing are performed.

[0065] In one embodiment of the present application, when a staff member performs an adjustment operation on a biological reaction process according to the adjustment method, the recognition model matches the corresponding standard operation according to the adjustment method, and determines whether the adjustment operation of the staff member is correct according to the standard operation.

[0066] In one embodiment of the present application, the recognition model also includes identifying the identity of the staff, establishing an operation database for all staff, marking operations that are inconsistent with the standard operations or consistent with the erroneous operations, and prompting staff members who have been marked more than a threshold number of times before performing subsequent operations.

[0067] It should be noted that the specific implementation of a bioreactor process optimization method in an embodiment of the present application refers to the specific implementation of a bioreactor process optimization system proposed in the first aspect of the aforementioned embodiment of the present application, and will not be repeated here.

[0068] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that an article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such article or device. In the absence of more restrictions, the elements defined by the sentence "includes..." do not exclude the existence of other identical elements in the article or device including the elements.

[0069] The above is a detailed introduction to a bioreactor process optimization system provided. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the bioreactor process optimization system of the present application and its core idea; at the same time, for general technicians in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A bioreactor process optimization system, characterized in that: include: The data acquisition module collects the historical reaction process data of the bioreactor in the database, obtains the real-time reaction process data of the bioreactor through the sensor, and saves it; A data analysis module analyzes the historical reaction process data to obtain different effects of different adjustment methods on the biological reaction process; A simulation module, which establishes a cell culture model according to the corresponding relationship obtained by the data analysis module, and adds the biological type and growth stage as independent variables to the cell culture model; The dynamic adjustment module obtains the real-time reaction process data and affects the biological culture process by adjusting the bioreactor. The specific adjustment method is obtained by inputting the real-time reaction process data and the required adjustment results into the simulation module.

2. A bioreactor process optimization system according to claim 1, characterized in that: The dynamic adjustment module includes a material adjustment submodule and an environment adjustment submodule. The material adjustment submodule adjusts the biological culture process by inputting materials into the bioreactor, and the materials include pH regulators, gases, biological nutrients and catalytic enzymes; the environment adjustment submodule adjusts the biological culture process by adjusting the reaction environment of the bioreactor, and the environment includes temperature, pressure and speed data in the bioreactor; In the data analysis module, the different adjustment methods include inputting different materials and / or adjusting environmental data, and the real-time reaction process data includes biological types and growth stages, and different growth stages of different biological types are obtained, and the impact results on the biological reaction process under different adjustment methods.

3. A bioreactor process optimization system according to claim 2, characterized in that: It also includes a design adjustment module for adjusting the design parameters of the bioreactor before the biological reaction, including tank size, impeller type, impeller combination form, impeller size, ventilation method and baffle size. Different design parameters correspond to different fluid dynamic simulation results through the simulation model, and the design parameters are associated with cell culture performance indicators. The design parameters corresponding to the highest cell culture performance indicator are used to adjust the bioreactor.

4. A bioreactor process optimization system according to claim 2, characterized in that: It also includes a parameter adjustment module. After the dynamic adjustment module completes the adjustment of the current biological culture process, the real-time reaction process data is saved as the historical reaction process data. By inputting multiple batches of the historical reaction process data into the neural network model, learning and updating the adjustment parameter indicators in the reaction process.

5. A bioreactor process optimization system according to claim 1, characterized in that: The dynamic adjustment module also includes an early warning adjustment submodule, which obtains the type of reaction data collected by the sensor and sets thresholds for various reaction data of different biological types at different growth stages. The sensor obtains real-time reaction process data of the bioreactor and obtains a change curve of the reaction data. When the change curve shows an abnormal trend, an early warning operation is performed; when the reaction process data reaches a corresponding threshold, a corresponding adjustment operation is performed according to the threshold reached by the reaction process data.

6. A bioreactor process optimization system according to any one of claims 1 to 5, characterized in that: The data acquisition module also includes an external data submodule, which collects and identifies the operations of the staff during the biological reaction process through the sensor, screens and classifies standard operations and erroneous operations, and trains the recognition models respectively. The trained recognition models identify the operations of the staff, and when the operations of the staff are inconsistent with the standard operations, or consistent with the erroneous operations, corresponding prompts and processing are performed.

7. A bioreactor process optimization system according to claim 6, characterized in that: The external data submodule obtains the adjustment method issued by the dynamic adjustment module. When the staff adjusts the biological reaction process according to the adjustment method, the recognition model matches the corresponding standard operation according to the adjustment method, and determines whether the staff's adjustment operation is correct according to the standard operation.

8. A bioreactor process optimization system according to claim 7, characterized in that: The recognition model also includes identifying the identity of the staff, establishing an operation database for all staff, marking operations that are inconsistent with the standard operations or consistent with the erroneous operations, and prompting staff who have been marked more than a threshold before performing subsequent operations.

9. A bioreactor process optimization method, characterized in that: include: Collect historical reaction process data of the bioreactor in the database, obtain real-time reaction process data of the bioreactor through sensors, and save them; Analyze the historical reaction process data to obtain different effects of different adjustment methods on the biological reaction process; Establishing a cell culture model according to the corresponding relationship obtained from the different effects of different adjustment methods on the biological reaction process, and adding the biological type and growth stage as independent variables to the cell culture model; The real-time reaction process data is obtained, and the biological culture process is affected by adjusting the bioreactor. The specific adjustment method is obtained through the real-time reaction process data and the required adjustment result.

Images