A digital twin operation virtual-real fusion method for an Internet of Things cell drug factory

Abstract

The present application provides a method for integrating virtual and real operations of a digital twin of an Internet of Things cell drug factory, including: constructing geometric models of a culture medium storage tank and a bioreactor through 3D modeling software, mapping and associating the historical process parameters of the culture medium supply system and the real-time monitored temperature, pH value, and dissolved oxygen concentration with the geometric models to construct a digital twin model of the reactor; analyzing historical production data through machine learning algorithms, extracting the quantitative relationships between key parameters such as cell metabolic rate, glucose concentration and pH value in the culture medium, and reactor temperature, establishing a cell growth prediction model to predict the growth trend and metabolic requirements of cells within a target time period; integrating the virtual-real mapping mechanism of the culture medium supply system with the intelligent manufacturing execution system to enable collaborative optimization control of the culture medium supply process and the entire drug production process, and automatically scheduling the process steps of the culture medium production process according to the production plan and cell growth requirements.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a virtual-reality fusion method for digital twin operation of a cell drug factory combined with the Internet of Things. Background Art

[0002] In the culture medium supply system of the cell drug factory, the real-time fluctuation of the redox potential of the culture medium storage tank will affect cell metabolism. The dissolved oxygen fluctuation signal collected by the IoT sensor needs to be analyzed to predict and dynamically adjust the culture medium replacement cycle. However, in the actual production process, due to the complex composition of the culture medium, the interaction between the components is difficult to accurately grasp, resulting in unclear correlation between the dissolved oxygen fluctuation signal and cell metabolism. At the same time, there are differences in the growth state and metabolic level of cells in different batches, making it impossible for a fixed culture medium replacement cycle to meet the needs of dynamic adjustment. In addition, fluctuations in environmental factors during the culture process, such as temperature and pH value, will also affect the dissolved oxygen level, further increasing the difficulty of prediction. How to establish an accurate mapping relationship between dissolved oxygen fluctuation signals and cell metabolism in a complex production environment, and based on this, to achieve dynamic optimization of the culture medium replacement cycle, is a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present invention provides a virtual-real fusion method for digital twin operation of a cell drug factory combined with the Internet of Things, which mainly includes:

[0004] A dissolved oxygen sensor is used to monitor the fluctuation of dissolved oxygen concentration in the culture medium in the bioreactor in real time, and the time series data of dissolved oxygen concentration is obtained. The time series data is input into the long short-term memory neural network model for training and learning, and the change trend of dissolved oxygen concentration in a preset time period is predicted. Combined with the preset dissolved oxygen concentration threshold, it is determined whether the culture medium replacement cycle needs to be adjusted;

[0005] If adjustment is needed, the cell density and cell viability during the cell growth process are collected in real time to generate a cell growth curve. The support vector machine algorithm is used to establish a correlation model between the cell growth curve and the culture medium replacement cycle, and the changes in the cell growth curve under different replacement cycle conditions are obtained to determine the target replacement cycle;

[0006] The geometric model of the culture medium storage tank and bioreactor is constructed through 3D modeling software. The historical process parameters of the culture medium supply system and the real-time monitored temperature, pH value and dissolved oxygen concentration are mapped and associated with the geometric model to build a digital twin model of the reactor.

[0007] Obtain the internal environmental parameters of the culture medium storage tank and the bioreactor in real time, synchronize the internal environmental parameters with the virtual parameters in the digital twin model in real time, construct a virtual-real fusion closed-loop control model for the culture medium supply system, and dynamically adjust the flow rate and concentration of the supply system according to the cell growth requirements and metabolic status;

[0008] Combined with the digital twin model, perform multi-scale and multi-physical field simulations of the culture medium storage tank and the bioreactor during the cell growth process, simulate and analyze the effects of fluctuations in the oxidation-reduction potential and dissolved oxygen concentration of the culture medium on cell growth, and change the ratio and supply of key nutrients in the culture medium;

[0009] Analyze historical production data through machine learning algorithms, extract the quantitative relationships between key parameters such as cell metabolic rate, glucose concentration, pH value in the culture medium, and reactor temperature, establish a cell growth prediction model, and predict the cell growth trend and metabolic requirements during the target time period;

[0010] Integrate the virtual-real mapping mechanism of the culture medium supply system with the intelligent manufacturing execution system to enable the collaborative optimization control of the culture medium supply process and the entire process of drug production, and automatically schedule the processes of the culture medium production process according to the production plan and cell growth requirements.

[0011] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0012] The present invention discloses a virtual-real fusion method for digital twin operation of an Internet of Things-based cell drug factory. The dissolved oxygen concentration of the culture medium in the bioreactor is monitored in real time through a dissolved oxygen sensor, and its change trend is predicted using a long short-term memory neural network. Combining cell density and viability data, an association model between the cell growth curve and the culture medium replacement cycle is established. The present invention constructs digital twin models of the culture medium storage tank and the bioreactor, realizing closed-loop control of virtual-real fusion. Through multi-scale and multi-physical field simulation analysis of the effects of culture medium parameters on cell growth, and using machine learning algorithms to predict cell growth trends and metabolic requirements, intelligent control of the culture medium supply process is achieved. The present invention integrates the culture medium supply system with the intelligent manufacturing execution system, realizing collaborative optimization control of the entire process and improving the efficiency and quality of biopharmaceutical production. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a flowchart of a virtual-real fusion method for digital twin operation of an Internet of Things-based cell drug factory according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this specification without creative efforts shall fall within the scope of protection of this specification.

[0015] As Figure 1 , a method for integrating virtual and real operations of a digital twin of an Internet of Things cell drug factory in this embodiment may specifically include:

[0016] Step S101, use a dissolved oxygen sensor to continuously monitor the fluctuation of the dissolved oxygen concentration in the culture medium in the bioreactor, obtain the time series data of the dissolved oxygen concentration, input the time series data into a long short-term memory neural network model for training and learning, predict the change trend of the dissolved oxygen concentration within a preset time period, and combine the preset dissolved oxygen concentration threshold to determine whether it is necessary to adjust the culture medium replacement cycle.

[0017] Obtain the dissolved oxygen concentration value collected by the dissolved oxygen sensor of the bioreactor, and obtain the dissolved oxygen concentration time series data set according to the sampling time mark; use a normalization method to standardize the dissolved oxygen concentration time series data set, and obtain a standardized dissolved oxygen concentration time series data set by removing outliers and supplementing missing values; extract the mean, variance, and peak statistical features of the dissolved oxygen concentration from the standardized dissolved oxygen concentration time series data set, and input them into a long short-term memory neural network for training to obtain a dissolved oxygen concentration prediction model; predict the dissolved oxygen concentration within a future fixed time period through the dissolved oxygen concentration prediction model. If the predicted dissolved oxygen concentration is lower than the preset concentration threshold and the decline rate exceeds the preset rate threshold, it is marked as the culture medium replacement warning time point.

[0018] Specifically, the dissolved oxygen data of the culture medium in the bioreactor is collected by a dissolved oxygen sensor, and the dissolved oxygen concentration values are obtained at fixed sampling time intervals. According to the sampling time marks, a time series dataset of dissolved oxygen concentration is formed. The time series dataset of dissolved oxygen concentration is subjected to data cleaning to remove outliers and fill in missing values. The normalization method is used to standardize the dissolved oxygen concentration values, and a standardized time series dataset of dissolved oxygen concentration is obtained. Statistical features such as the mean, variance, and peak value of the dissolved oxygen concentration are extracted from the standardized time series dataset of dissolved oxygen concentration, a training sample matrix is constructed, and it is input into a long short-term memory neural network for training to obtain a dissolved oxygen concentration prediction model. The dissolved oxygen concentration prediction model is used to predict the dissolved oxygen concentration within a fixed future time period, and a predicted time series dataset of dissolved oxygen concentration is generated. According to the predicted time series dataset of dissolved oxygen concentration, the decline rate of the dissolved oxygen concentration is calculated. When the decline rate exceeds the preset rate threshold and the predicted dissolved oxygen concentration is lower than the preset concentration threshold, it is marked as the warning time point for culture medium replacement. The historical culture medium replacement time records are extracted from the dissolved oxygen monitoring database, and the average replacement cycle is calculated in combination with the warning time point to generate the next culture medium replacement time. During the cell culture process in the bioreactor, the dissolved oxygen concentration is a key parameter affecting the cell growth state, and the change of the dissolved oxygen concentration reflects the intensity of cell metabolic activities. When the dissolved oxygen concentration in the culture medium continuously decreases, it indicates that the cells are metabolically active, the oxygen consumption rate increases, and at the same time, it also indicates that the nutrients in the culture medium are gradually consumed. In practical applications, the dissolved oxygen sensor collects the dissolved oxygen concentration values every 10 minutes and continuously monitors for 48 hours to obtain the dissolved oxygen concentration data at 288 time points. The normal range of the dissolved oxygen concentration fluctuates between 35% and 70%. During the data cleaning process, the outliers below 20% or above 85% are removed, and the missing data is filled in by the linear interpolation method to ensure the continuity of the data. During the standardization process, 50% of the mean value of the dissolved oxygen concentration in the historical data is selected as the reference value, and the variance is 15%. The collected dissolved oxygen concentration data is transformed into a standard normal distribution. The extracted statistical features include indicators such as the arithmetic mean, geometric mean, standard deviation, kurtosis, and skewness of the dissolved oxygen concentration within every 4 hours. These features can comprehensively reflect the change law of the dissolved oxygen concentration. The long short-term memory neural network model adopts a three-layer network structure. The input layer contains 12 neurons corresponding to the dissolved oxygen concentration data of the previous 12 time points, the hidden layer contains 8 neurons, and the output layer predicts the dissolved oxygen concentration values of the next 4 time points. The model is trained with 120 sets of historical data, of which 80% is used for training and 20% is used for verification. The prediction results show that when the cell growth enters the logarithmic phase, the dissolved oxygen concentration shows an obvious downward trend, with an average decrease of 2% to 3% per hour. The warning threshold is set as the dissolved oxygen concentration of 35%, and the decline rate threshold is 3% per hour. When the predicted dissolved oxygen concentration is lower than 35% within the next 6 hours and the decline rate exceeds 3% per hour, a warning for culture medium replacement is triggered.By analyzing historical culture data, it is found that the average medium replacement cycle during cell culture is 72 hours, but the actual replacement time fluctuates with factors such as cell density and metabolic activity. During the period of vigorous cell growth, the replacement cycle is shortened to 54 hours, while during the growth plateau phase, the replacement cycle can be extended to 96 hours. Combining the warning time point with the historical replacement records, it is calculated that the next replacement time is generally between 12 hours and 24 hours after the warning time point.

[0019] Step S102, if adjustment is needed, then in real time collect the cell density and cell viability during cell growth, generate a cell growth curve, use the support vector machine algorithm to establish an association model between the cell growth curve and the medium replacement cycle, obtain the changes in the cell growth curve under different replacement cycle conditions, and determine the target replacement cycle.

[0020] Obtain the cell density value and fluorescence intensity signal in the culture solution collected by the cell density detector, and record according to a fixed time interval to obtain the original cell growth data set; perform outlier detection on the original cell growth data set, and use the cubic spline interpolation method to supplement the missing data to obtain the processed cell growth data set; calculate the cell density change rate and cell viability change rate for the processed cell growth data set, and construct a cell growth feature vector in combination with cell morphology parameters; extract the historical medium replacement time points from the medium replacement record database, establish the corresponding relationship between the cell growth feature vector and the replacement time point, input it into the support vector machine algorithm to obtain the medium replacement cycle model, and predict the cell density integral value and cell viability average value under different replacement time intervals according to the medium replacement cycle model; determine whether the cell density integral value and cell viability average value are within the preset threshold interval, and if so, select the time point with the largest cell density integral value and the highest cell viability average value as the target medium replacement cycle.

[0021] Specifically, the cell density values in the culture medium are collected in real time by a cell density detector, and the fluorescence intensity signals are synchronously collected to calculate the cell viability values. The original cell growth data set is recorded at fixed time intervals. Outlier detection is performed on the original cell growth data set, and the cell density values and cell viability values that exceed the physiological range are removed. The cubic spline interpolation method is used to supplement the missing data, and the processed cell growth data set is obtained. According to the processed cell growth data set, characteristic parameters such as the cell density change rate, cell viability change rate, and density-viability ratio are calculated, and combined with the cell morphology parameters at the corresponding time points, a cell growth feature vector is constructed. The historical medium replacement time points are extracted from the medium replacement record database, and the corresponding relationship between the cell growth feature vector and the replacement time points is established and input into the support vector machine algorithm to train the medium replacement cycle model. The medium replacement cycle model is used to predict the cell density curves at different replacement time intervals, and the cell density integral value and the average cell viability value corresponding to each time interval are calculated. According to the preset cell density threshold interval and cell viability threshold interval, the eligible medium replacement cycle values are screened out, and the time point with the largest cell density integral value and the highest average cell viability value is selected as the target medium replacement cycle. During the cell culture process, cell density and cell viability are important indicators reflecting the cell growth state. The cell number in the culture medium is monitored in real time by a cell density detector, and data is collected every 30 minutes. The measurement range of cell density is between 1×10^5 and 1×10^7 cells / ml. At the same time, live cells are labeled with a fluorescent dye, and the cell viability is calculated by detecting the fluorescence intensity. The fluorescence intensity is positively correlated with the cell metabolic activity. In practical applications, due to detection equipment failures or external interferences, outliers often appear in the original data. For example, the cell density suddenly drops to 0 or surges to 1×10^8 cells / ml. These data that clearly exceed the cell physiological range need to be removed. For the time points with missing data, the cubic spline interpolation method is used for supplementation. This method can maintain the continuity and smoothness of the data curve and avoid drastic fluctuations. The cell growth characteristic parameters include multiple dimensions. The cell density change rate reflects the proliferation speed of the cell population. Under normal circumstances, the density change rate during the logarithmic growth phase is between 0.15 and 0.25. The cell viability change rate represents the change trend of cell metabolic activity. A decrease in viability exceeding 20% usually indicates an adverse change in the culture environment. The density-viability ratio comprehensively reflects the metabolic level of a single cell. A value between 2 and 5 indicates a good cell growth state. Historical culture data shows that different medium replacement cycles will result in significant differences in the cell growth curve. If the replacement cycle is too short, the cells fail to fully utilize the nutrients in the medium, causing waste of resources; if the replacement cycle is too long, the cell growth is inhibited by the accumulation of metabolic wastes, resulting in a decrease in cell viability.By analyzing historical data, it is found that when the cell density reaches 5×10^6 cells / ml and the cell viability remains above 85%, changing the culture medium can achieve a higher cell proliferation efficiency. The support vector machine algorithm learns the patterns in the historical culture data to establish a mapping relationship between the cell growth characteristics and the optimal replacement time point. In the model validation stage, simulations are performed for different replacement cycles, and multiple candidate replacement time points are set within the range of 48 hours to 96 hours at 24-hour intervals. By calculating the integral value of the cell density and the average viability value corresponding to each time point, the total cell production and quality indicators are comprehensively evaluated. The selected target replacement cycle should ensure that the integral value of the cell density reaches the expected level while maintaining a high average cell viability.

[0022] Step S103: Construct a geometric model of the culture medium storage tank and the bioreactor using 3D modeling software, and map and associate the historical process parameters of the culture medium supply system and the real-time monitored temperature, pH value, and dissolved oxygen concentration with the geometric model to construct a digital twin model of the reactor.

[0023] Use 3D modeling software to construct a three-dimensional structure model of the storage tank and the bioreactor. The three-dimensional structure model includes geometric data of the feed inlet, discharge outlet, stirring device, and sensor installation positions. Establish parameter monitoring points based on the surface of the three-dimensional structure model, obtain culture medium supply system data from the process parameter database, and collect real-time monitored data at the sensor installation positions. Divide the fluid force field, temperature field, and concentration field for the parameter monitoring points, and calculate the distribution values of the fluid force field, temperature field, and concentration field in space through the finite element algorithm. Use the deep neural network algorithm to train the coupling relationship between the fluid force field, temperature field, and concentration field, and establish a digital mapping structure of the reactor according to the coupling relationship to obtain a digital twin model.

[0024] Specifically, according to the physical size parameters of the culture medium storage tank and the bioreactor, a three-dimensional modeling software is used to construct a three-dimensional structural model of the storage tank and the reactor, including the geometric data of the feed inlet, the discharge outlet, the stirring device, and the sensor installation positions, to generate the first structural model. The flow rate, pressure, and stirring rate values of the culture medium supply system are extracted from the process parameter database, and the values are normalized. The real-time monitoring data of the temperature sensor, pH sensor, and dissolved oxygen sensor in the bioreactor are collected to generate a process parameter dataset. According to the numerical distribution in the process parameter dataset, parameter monitoring points are established on the surface of the first structural model, and the interpolation algorithm is used to calculate the distribution law of the parameters in space to generate the second parameter structural model. In the second parameter structural model, the fluid force field, temperature field, and concentration field are divided, and a system of equations for calculating the field quantities is established. The finite element algorithm is used to calculate the distribution law of the field quantities in space to generate the third physical field model. The deep neural network algorithm is used to train the coupling relationship between the fluid force field, temperature field, and concentration field. The feature vector including the field quantity values, field quantity gradients, and inter-field forces is input, and the evolution law under the interaction of the field quantities is output. According to the coupling evolution law of the field quantities, a digital mapping structure of the reactor is established on the basis of the third physical field model to realize the dynamic association of process parameters, monitoring data, and physical field distribution, and to form a digital twin model. The digital twin modeling of the bioreactor involves the coupling analysis of multiple physical fields, and the reactor structure directly affects the flow characteristics and mass transfer efficiency of the internal culture medium. Taking the stirred bioreactor as an example, the reactor height is 300 mm, the diameter is 200 mm, and it is equipped with a four-blade propeller stirring device. The diameter of the stirring paddle is 1 / 3 of the inner diameter of the container, and the height from the bottom is 1 / 4 of the container height. The feed inlet is located at the top center, the discharge outlet is located at the bottom side wall, and the dissolved oxygen sensor, pH sensor, and temperature sensor are evenly arranged along the reactor height. The acquisition frequency of the process parameters is set according to the parameter change characteristics. The temperature and pH value are collected every 10 minutes, the dissolved oxygen concentration is collected every 5 minutes, and the stirring speed and flow rate are collected in real time. The original data is normalized, and the temperature range of 35 to 38 degrees Celsius is mapped to the interval of 0 to 1, the pH value of 6.8 to 7.2 is mapped to the interval of 0 to 1, and the dissolved oxygen concentration of 30% to 70% is mapped to the interval of 0 to 1. A monitoring point grid is arranged on the surface of the reactor structure, and the grid density changes with the parameter gradient. The grid is encrypted in the area where the parameters change violently. Taking the dissolved oxygen distribution as an example, the monitoring point spacing is set to 10 mm in the stirring area and near the gas-liquid interface, and the spacing in other areas is set to 20 mm. The radial basis function interpolation algorithm is used to calculate the parameter distribution between the grid points to realize the continuous distribution of the parameters in space. The division of the physical field is based on the fluid mechanics and heat and mass transfer theories. The fluid force field includes the pressure field and the velocity field. The pressure field reflects the pressure distribution of the culture medium under the action of gravity and stirring, and the velocity field describes the flow characteristics of the culture medium. The temperature field is affected by the heating device and the ambient temperature, and a temperature gradient is formed near the reactor wall surface.The dissolved oxygen concentration field is affected by both the bubble distribution and cell oxygen consumption. Deep neural networks are used to establish the coupling relationship between physical fields, and the input features include local field quantity values, spatial gradients of field quantities, and inter-field forces. Taking the coupling of the temperature field and the dissolved oxygen field as an example, an increase in temperature will reduce the dissolved oxygen content in the culture medium. For every 1-degree Celsius increase, the dissolved oxygen concentration decreases by approximately 2%. At the same time, the density difference caused by temperature changes will affect the natural convection of the culture medium, thereby changing the mass transfer efficiency of dissolved oxygen. The digital twin model reflects the internal state of the reactor by real-time updating the physical field distribution. When the stirring speed is increased from 200 revolutions per minute to 400 revolutions per minute, the model can predict the dynamic process of the dissolved oxygen concentration increasing from 45% to 65% within 15 minutes, while showing slight fluctuations in the temperature field and the stability of the pH value. This digital mapping of multi-physical field coupling can accurately reflect the dynamic change law of the culture environment.

[0025] Step S104: Real-time obtain the internal environment parameters of the culture medium storage tank and the bioreactor, perform real-time mapping synchronization between the internal environment parameters and the virtual parameters in the digital twin model, construct a virtual-real fusion closed-loop control model for the culture medium supply system, and dynamically adjust the flow rate and concentration of the supply system according to the cell growth requirements and metabolic status.

[0026] Collect the internal environment parameters of the culture medium storage tank and the bioreactor through temperature sensors, pH sensors, and dissolved oxygen sensors, and use the sliding window method to perform noise reduction processing on the environment parameters to obtain an environment parameter data set; collect the glucose concentration and lactic acid concentration during the cell culture process according to the environment parameter data set, and combine the cell density data to obtain a cell metabolic status data set; use the cell metabolic status data set to calibrate the sensor collection positions in the digital twin structure, and calculate the parameter spatial distribution through the Kriging interpolation algorithm to obtain a virtual parameter distribution data set; calculate the root mean square error value according to the virtual parameter distribution data set and the environment parameter data set, and use the deep reinforcement learning algorithm to train the culture medium supply controller to obtain the culture medium flow compensation value and concentration compensation value.

[0027] Specifically, the internal environmental parameters of the culture medium storage tank and the bioreactor are collected by temperature sensors, pH sensors, dissolved oxygen sensors, liquid level sensors, and pressure sensors. The sliding window method is used to denoise and process outliers in the collected data to generate an environmental parameter dataset. The glucose concentration, lactic acid concentration, and amino acid concentration during the cell culture process are collected, and the change trend of metabolite concentration is calculated. Combining the cell density data and cell viability data, a cell metabolic state dataset is generated. The sensor collection positions are calibrated in the digital twin structure, the corresponding relationship between the actual environmental parameters and the virtual environmental parameters is established, and the Kriging interpolation algorithm is used to calculate the spatial distribution of parameters to generate a virtual parameter distribution dataset. The root mean square error between the virtual parameter distribution dataset and the environmental parameter dataset is calculated, an error compensation function is established, and the virtual parameters are corrected to generate a corrected virtual parameter distribution dataset. The culture medium supply controller is trained using a deep reinforcement learning algorithm. The cell metabolic state dataset and the corrected virtual parameter distribution dataset are input, and the culture medium flow compensation value and concentration compensation value are output. According to the compensation values, the rotation speed of the culture medium supply pump is adjusted, and the mixing ratio of nutrients in the culture medium preparation unit is synchronously adjusted. The actual supply parameters are collected, and the virtual parameter distribution dataset is continuously updated. The virtual-real fusion control of the culture medium supply system involves the real-time monitoring and regulation of multiple environmental parameters. In practical applications, the measurement range of the temperature sensor is 30 to 40 degrees Celsius, and the accuracy is 0.1 degree Celsius; the measurement range of the pH sensor is 6.0 to 8.0, and the accuracy is 0.01; the measurement range of the dissolved oxygen sensor is 0 to 100%, and the accuracy is 0.1%. The sliding time window of 60 seconds is used to process the original data, and outliers beyond the normal range are removed. The monitoring of the cell metabolic state mainly focuses on the glucose consumption rate and the accumulation of metabolites. The glucose concentration gradually decreases from the initial 4.5 g / L and triggers a replenishment signal when it drops to 1.0 g / L. Lactic acid, as the main metabolite, has a concentration that gradually increases from 0 and inhibits cell growth when it exceeds 2.0 g / L. The amino acid monitoring includes the concentration changes of 20 amino acids such as glutamine and alanine. 48 sensor collection positions are set in the digital twin structure to form a spatial monitoring grid in the reactor. The Kriging interpolation algorithm predicts the parameter distribution of unknown points based on the measured values of known points, considering spatial correlation and anisotropy characteristics. Taking the dissolved oxygen distribution as an example, the spatial correlation distance in the stirring area is 20 mm, while the correlation distance in the static area is 50 mm. There is an error between the virtual parameters and the actual parameters. The root mean square error of the temperature field is usually within 0.5 degree Celsius, the error of the pH field is within 0.05 units, and the error of the dissolved oxygen field is within 2%. By establishing an error compensation function, the virtual parameters are dynamically corrected according to the statistical law of historical data. The culture medium supply controller is trained using a deep reinforcement learning algorithm, and the reward function is set to include multiple indicators such as the cell density growth rate and the metabolite concentration change rate.The controller outputs regulation instructions every 5 minutes. The adjustment range of the culture medium flow rate is 50 to 200 milliliters per hour, and the concentration adjustment range is 0.5 to 2 times the base concentration. The culture medium preparation unit includes multiple peristaltic pumps and mixers, and adjusts the ratio of each nutrient according to the cell metabolism requirements. The glucose stock solution concentration is 200 grams per liter, and the amino acid stock solution concentration is 50 times the base culture medium concentration. The on-demand supply of substances is achieved by precisely controlling the valve opening. The real-time feedback of the supply parameters is synchronized with the update frequency of the virtual parameters to ensure that the digital twin model can accurately reflect the dynamic changes of the actual culture environment. During the culture process, the cell density doubles every 24 hours, and the metabolite concentration shows periodic fluctuations. The culture medium supply controller maintains a steady-state environment for cell growth through predictive regulation.

[0028] Step S105: Combine with the digital twin model to conduct multi-scale and multi-physical field simulation of the fusion culture medium storage tank and bioreactor during the cell growth process, simulate and analyze the influence of the fluctuations of the oxidation-reduction potential and dissolved oxygen concentration of the culture medium on cell growth, and change the ratio and supply of the key nutrient components in the culture medium.

[0029] Obtain the geometric structure data of the storage tank and reactor in the digital twin structure, and establish the Navier-Stokes equation describing fluid motion and the convective diffusion equation describing the mass transfer process according to the geometric structure data; establish a multi-scale grid structure using the adaptive grid division method according to the Navier-Stokes equation and the convective diffusion equation, and obtain the multi-physical field distribution data through the multi-scale grid structure; use the discrete element algorithm to construct a cell-scale calculation module according to the multi-physical field distribution data, and obtain the oxidation-reduction potential distribution and dissolved oxygen concentration gradient data in the microenvironment through the cell-scale calculation module; establish a cell metabolism flux balance equation set according to the dissolved oxygen concentration gradient data, and calculate the glucose, amino acid, fatty acid consumption rate and metabolite generation rate data through the cell metabolism flux balance equation set.

[0030] Specifically, geometric structure data of storage tanks and reactors are extracted from the digital twin structure. According to the spatial distribution laws of the potential field, flow field, temperature field, and concentration field, the Navier-Stokes equations describing fluid motion are established at the macroscopic scale, and the convective diffusion equations describing mass transfer processes are established at the mesoscopic scale. An adaptive mesh generation method is adopted to refine the mesh in regions with intense fluid motion and sparse the mesh in static fluid regions, establishing a multi-scale mesh structure. The finite element method is used to solve the multi-physics field governing equations. The discrete element algorithm is employed to construct a computational module at the cellular scale, dividing it into three computational domains: cell membrane, cytoplasm, and nucleus, and calculating the distribution of redox potential and the dissolved oxygen concentration gradient in the microenvironment around the cell. Based on the computational results at the cellular scale, a system of equations for cellular metabolic flux balance is established to calculate the consumption rates of substances such as glucose, amino acids, and fatty acids and the production rates of metabolites. A deep neural network algorithm is used to establish a metabolic flux predictor, which inputs feature vectors such as cell density, metabolite concentration, and environmental parameters to predict the cell growth curve under different culture conditions. According to the predicted results of metabolic flux, the matching relationship between the consumption rate and replenishment rate of nutrient components in the culture medium is calculated to generate a new culture medium formula, and the formula parameters are continuously optimized through a feedback regulation mechanism. The multi-scale and multi-physics field simulation of the cell culture process involves cross-scale calculations from the molecular level to the reactor scale. At the macroscopic level, the fluid motion in the bioreactor is mainly driven by the agitator, forming a complex three-dimensional flow field structure. When the agitation speed is 100 revolutions per minute, the flow velocity in the central region of the reactor reaches 0.5 m / s, while the flow velocity in the edge region drops below 0.1 m / s. This velocity difference directly affects the mass transfer efficiency of dissolved oxygen. During mesh generation, a finer mesh is used for regions with a large flow velocity gradient, and the mesh size is dynamically adjusted between 0.5 mm and 5 mm. The mesh density is the highest near the agitator, with 1000 mesh elements per unit volume, while the mesh density in the static region is reduced to 100 elements per cubic centimeter. This adaptive mesh can accurately capture the detailed features of the flow field. The cellular-scale calculation focuses on the microenvironment around a single cell. The cell membrane is about 10 nm thick and constitutes the main barrier for mass transfer. The cytoplasm has a volume of about 2000 cubic micrometers and is the main site of metabolic reactions. In the microenvironment, the redox potential drops from 150 mV outside the cell to -50 mV inside the cell, forming a transmembrane potential difference, which affects the metabolic activity of the cell. Metabolic flux analysis shows that a single cell consumes 1 picomole of glucose per hour, producing 2 picomoles of lactic acid and 0.5 picomole of carbon dioxide. The consumption patterns of amino acids show significant differences. The consumption rate of glutamine is the highest, reaching 0.8 picomoles per hour, while the consumption rate of tryptophan is only 0.05 picomoles per hour. This difference reflects the complexity of the cellular metabolic network. The deep neural network has a three-layer structure. The input layer contains 48 nodes, corresponding to cell density, the concentrations of 12 key metabolites, and various environmental parameters.The hidden layer processes these features through 1024 neurons, and the output layer predicts the cell growth curve within the next 24 hours. The training dataset contains 1000 sets of historical culture data, and the validation set accounts for 20%. The optimization of the culture medium formula is based on the dynamic balance of metabolic requirements. When the cell density reaches 2×10^6 cells / ml, the glucose consumption rate increases to 3 times the initial value, while the glutamine consumption rate increases to 4 times. According to this trend, the glucose concentration in the culture medium is increased from 4.5 g / L to 6.0 g / L, glutamine is increased from 2.0 mmol / L to 3.5 mmol / L, and the content of other essential amino acids is appropriately increased. Through this dynamic regulation, the specific growth rate of the cells is maintained between 0.028 and 0.032 / hour, and the metabolic homeostasis is well maintained.

[0031] Step S106, analyze the historical production data through a machine learning algorithm, extract the quantitative relationship between the cell metabolic rate, the glucose concentration and pH value in the culture medium, and the key parameters of the reactor temperature, establish a cell growth prediction model, and predict the cell growth trend and metabolic requirements within the target time period.

[0032] Obtain the culture medium parameter data in the historical database, process the culture medium parameter data through the maximum-minimum normalization method to obtain a normalized culture parameter dataset; according to the normalized culture parameter dataset, use the wavelet transform denoising algorithm and the preset decomposition depth to perform signal reconstruction on the normalized culture parameter dataset to obtain a smoothed dataset; for the smoothed dataset, calculate the eigenvalues and eigenvectors through the principal component analysis method. If the cumulative contribution rate of the eigenvalues reaches the preset threshold, obtain a dimensionality-reduced feature dataset; train a recurrent neural network model using the dimensionality-reduced feature dataset. If the prediction error of the validation set is less than the preset threshold, obtain a cell growth prediction model, and calculate the metabolite generation rate through the cell growth prediction model.

[0033] Specifically, the values ​​of glucose concentration, pH value, temperature, dissolved oxygen concentration, and cell density in the culture medium were extracted from the historical database, and the missing values ​​and outliers were eliminated. The data were normalized using the maximum and minimum value standardization method to generate a standardized culture parameter data set. The wavelet transform denoising algorithm was applied to the standardized culture parameter data set, the db4 wavelet basis function was selected, and the four-layer decomposition depth was set. After removing the high-frequency noise, the signal was reconstructed to obtain the culture parameter data set after smoothing. According to the culture parameter data set after smoothing, the Pearson correlation coefficient between the parameters was calculated, the parameter association matrix was constructed, the principal component analysis method was used to extract the eigenvalues ​​and eigenvectors, and the principal components whose cumulative contribution rate reached the preset threshold were selected to generate a reduced dimension feature data set. The recursive neural network algorithm was used to construct a cell growth prediction model, the number of neurons in the input layer was set to be the same as the dimension of the reduced dimension feature data set, the hidden layer used a long short-term memory unit structure, and the training samples were divided in chronological order. According to the preset training parameters, the recursive neural network was iteratively trained, the root mean square error was used as the loss function, and the network weights were updated by the back propagation algorithm until the prediction error on the validation set was less than the preset threshold. Using the trained cell growth prediction model, the current culture parameters are input, and the cell density change trend in the future time period is predicted by the autoregressive method, and the metabolite generation rate and nutrient consumption rate at each time point are calculated. The parameter monitoring data during the cell culture process reflects the dynamic changes in cell growth status and metabolic activity. In the original monitoring data, the measurement range of glucose concentration is 0 to 10 g / L, the pH range is 6.5 to 7.8, the temperature range is 35 to 39 degrees Celsius, and the dissolved oxygen concentration range is 20% to 95%. These parameters are difficult to compare directly under different dimensions and numerical ranges. Through the maximum and minimum value normalization processing, all parameters are mapped to the range of 0 to 1 to make the data comparable. During the data preprocessing process, it was found that sensor failure and external interference can cause mutations and anomalies in the measurement data. For example, the temperature sensor drifts after the sterilization process, and the reading suddenly increases by 2 degrees Celsius; the pH electrode jumps during calibration, and the pH value deviates from 0.5 units instantly. Wavelet transform was used to smooth these anomalies. The reason for choosing db4 wavelet basis function is that it has good local feature extraction ability. After four-layer decomposition, it can effectively remove high-frequency noise while retaining the main features of the signal. There is a significant correlation between the parameters. By calculating the Pearson correlation coefficient, it was found that the correlation coefficient between glucose consumption rate and cell density reached 0.85, indicating that the two are strongly positively correlated. The change in pH value is also closely related to the accumulation of metabolites, with a correlation coefficient of -0.72, reflecting that the acidic substances produced during cell metabolism will reduce the pH value of the culture medium. Principal component analysis shows that the cumulative contribution rate of the first three principal components reached 87%, which means that these three principal components can be used to represent the main information of the original data. The recursive neural network adopts a long short-term memory unit structure, which has the advantage of processing long sequence data.The input layer contains 12 neurons, corresponding to the values of 4 culture parameters at 3 consecutive time points. The hidden layer is set with 64 LSTM units, which can capture the temporal characteristics of parameter changes. During the training process, a 24-hour sliding time window is adopted to predict the parameter changes in the next 12 hours. When the root mean square error on the validation set drops below 0.15, it is considered that the model training reaches convergence. The prediction results show that the model can accurately predict different stages of cell growth. During the exponential growth phase, the cell density doubles every 24 hours, and the glucose consumption rate reaches 0.5 g / L / hour; after entering the stationary phase, the growth rate decreases, and the glucose consumption is maintained at 0.2 g / L / hour. Based on these prediction results, the nutrient requirements of cells can be calculated in advance. For example, when the cell density reaches 5×10^6 cells / mL, 0.3 mmol of glutamine is consumed per hour, and 0.5 mmol of lactic acid is produced.

[0034] Step S107, integrate the virtual-real mapping mechanism of the culture medium supply system with the intelligent manufacturing execution system to enable the collaborative optimization control of the culture medium supply process and the entire process of drug production, and automatically schedule the processes of the culture medium production process according to the production plan and the cell growth requirements.

[0035] Collect the production plan scheduling data and the cell culture progress data, generate a production resource status data set according to the scheduling data and the progress data; obtain the production equipment maintenance records and failure records, and calculate the equipment status evaluation data set by using the equipment reliability evaluation index system; according to the production resource status data set and the equipment status evaluation data set, optimize and calculate the initial scheduling plan for the culture medium production by using the integer programming algorithm; for the initial scheduling plan for the culture medium production, construct a culture medium production scheduler by using the deep reinforcement learning algorithm, and process the production plan priority and the equipment status characteristics through the input layer of the scheduler to obtain a scheduling instruction sequence.

[0036] Specifically, collect production plan scheduling data and cell culture progress data, combine the liquid level information of the culture medium storage tank and the status information of the preparation equipment, and use a relational database to record the occupancy status of production equipment, raw materials, and human resources, generating a production resource status data set. Collect the maintenance records, fault records, and repair records of production equipment, establish an equipment reliability evaluation index system, calculate the availability score during the equipment idle period, and generate an equipment status evaluation data set. According to the culture medium consumption prediction results in the virtual-real mapping database, combine the production resource status data set and the equipment status evaluation data set, and use the integer programming algorithm to optimize the culture medium production batch arrangement, generating an initial scheduling plan. Use the deep reinforcement learning algorithm to construct a culture medium production scheduler. The input layer includes features such as production plan priority, equipment status, raw material inventory, and personnel scheduling, and the output layer generates a scheduling instruction sequence. Train the deep reinforcement learning algorithm based on historical scheduling data, set the reward function to include indicators such as production plan completion rate, resource utilization rate, and scheduling stability, and optimize the scheduling strategy through the policy gradient method. Allocate production resources according to the scheduling instruction sequence, generate production work orders, and monitor production process parameters. When equipment failures, raw material shortages, or quality anomalies are detected, trigger the online optimization of the scheduling plan. The culture medium production scheduling involves the collaborative management of multiple resource dimensions. In the monitoring of production resource status, the liquid level of the culture medium storage tank is sampled every 5 minutes through a liquid level sensor, and a replenishment signal is triggered when the liquid level is below 30%. The status of the preparation equipment includes indicators such as running duration, maintenance interval, and fault frequency. For example, the continuous running duration of the agitator exceeds 72 hours and preventive maintenance is required, and the cleaning equipment needs to be disinfected once every 8 hours of operation. The equipment reliability evaluation adopts a multi-layer index system, including equipment intact rate, mean time between failures, and repair recovery time. Taking the sterilizer as an example, the monthly intact rate requirement is above 95%, the average time between failures is not less than 720 hours, and the single-fault repair time does not exceed 4 hours. In the evaluation of the equipment idle period, factors such as equipment status, maintenance plan, and personnel configuration are comprehensively considered to calculate the equipment availability score for each period. The integer programming algorithm considers multiple constraints when optimizing the culture medium production batch. The time window constraint of the production equipment requires that the production time for each batch is not less than 4 hours, and the equipment cleaning and disinfection time is not less than 2 hours. The raw material inventory constraint ensures sufficient raw material supply during the production process. For example, the inventory of glucose mother liquor should at least meet the production demand for 3 days. The personnel scheduling constraint ensures that sufficient operators are equipped in each period. The deep reinforcement learning algorithm processes complex scheduling scenarios. The input features include 48 dimensions, covering production plans for the next 7 days, equipment status predictions, raw material inventory warnings, and other information. The action space of the scheduling instruction includes decision variables such as equipment allocation, batch arrangement, and personnel deployment. In the design of the reward function, the weight of the production plan completion rate is 0.4, the weight of the resource utilization rate is 0.3, and the weight of the scheduling stability is 0.3. In practical applications, production anomalies need to be responded to in a timely manner.When the temperature of the agitator bearing is detected to rise abnormally, the algorithm will predict that the equipment may malfunction and adjust the production plan in advance, transferring the original planned tasks to backup equipment. When the delay in raw material delivery by the supplier leads to a decrease in inventory levels, the scheduling algorithm will re-plan the production batch size, giving priority to ensuring the production of high-priority products. The quality monitoring of the culture medium production process runs through the entire process. Key parameters such as pH value, conductivity, and osmotic pressure are monitored online, and an alarm is triggered when the osmotic pressure deviates from the target value by ±10%. The production work order contains information such as raw material batch number, process parameters, and quality inspection results, enabling full traceability of the production process. Through multi-dimensional collaborative management, the culture medium production scheduling not only meets the requirements of the production plan but also ensures the stability of the production process and the consistency of product quality.

[0037] The above are only embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be included in the patent protection scope of the present invention by the same token.

Claims

1. A virtual-real fusion method for digital twin operation of a cell drug factory combined with the Internet of Things, characterized in that: The method comprises: A dissolved oxygen sensor is used to monitor the fluctuation of dissolved oxygen concentration in the culture medium in the bioreactor in real time, and the time series data of dissolved oxygen concentration is obtained. The time series data is input into the long short-term memory neural network model for training and learning, and the change trend of dissolved oxygen concentration in a preset time period is predicted. Combined with the preset dissolved oxygen concentration threshold, it is determined whether the culture medium replacement cycle needs to be adjusted; If adjustment is needed, the cell density and cell viability during the cell growth process are collected in real time to generate a cell growth curve. The support vector machine algorithm is used to establish a correlation model between the cell growth curve and the culture medium replacement cycle, and the changes in the cell growth curve under different replacement cycle conditions are obtained to determine the target replacement cycle; The geometric model of the culture medium storage tank and bioreactor is constructed through 3D modeling software. The historical process parameters of the culture medium supply system and the real-time monitored temperature, pH value and dissolved oxygen concentration are mapped and associated with the geometric model to build a digital twin model of the reactor. Acquire the internal environmental parameters of the culture medium storage tank and bioreactor in real time, map and synchronize the internal environmental parameters with the virtual parameters in the digital twin model in real time, build a virtual-real fusion closed-loop control model for the culture medium supply system, and dynamically adjust the flow and concentration of the supply system according to cell growth requirements and metabolic status; Combined with the digital twin model, multi-scale and multi-physics simulation of the fusion culture medium tank and bioreactor during cell growth was carried out to simulate and analyze the impact of fluctuations in the redox potential and dissolved oxygen concentration of the culture medium on cell growth, and to change the ratio and supply of key nutrients in the culture medium; Analyze historical production data through machine learning algorithms, extract the quantitative relationship between key parameters such as cell metabolic rate, glucose concentration and pH value in the culture medium, and reactor temperature, establish a cell growth prediction model, and predict the growth trend and metabolic needs of cells in the target time period; Integrate the virtual-reality mapping mechanism of the culture medium supply system with the intelligent manufacturing execution system to achieve coordinated optimization control of the culture medium supply process and the entire drug production process, and automatically schedule the culture medium production process according to the production plan and cell growth requirements.

2. The method according to claim 1, characterized in that The method uses a dissolved oxygen sensor to monitor the fluctuation of the dissolved oxygen concentration of the culture medium in the bioreactor in real time, obtains time series data of the dissolved oxygen concentration, inputs the time series data into a long short-term memory neural network model for training and learning, predicts the change trend of the dissolved oxygen concentration in a preset time period, and determines whether the culture medium replacement cycle needs to be adjusted in combination with a preset dissolved oxygen concentration threshold, including: Obtain the dissolved oxygen concentration value collected by the dissolved oxygen sensor of the bioreactor, and obtain the dissolved oxygen concentration time series data set according to the sampling time mark; The dissolved oxygen concentration time series data set is standardized by using a normalization method, and a standardized dissolved oxygen concentration time series data set is obtained by removing outliers and supplementing missing values; Extracting the mean, variance and peak statistical features of dissolved oxygen concentration according to the standardized dissolved oxygen concentration time series data set, and inputting them into the long short-term memory neural network training to obtain a dissolved oxygen concentration prediction model; The dissolved oxygen concentration prediction model is used to predict the dissolved oxygen concentration in a fixed time period in the future. If the predicted dissolved oxygen concentration is lower than a preset concentration threshold and the decreasing rate exceeds a preset rate threshold, it is marked as a warning time point for culture medium replacement.

3. The method according to claim 1, characterized in that If adjustment is required, the cell density and cell viability during the cell growth process are collected in real time to generate a cell growth curve, and a support vector machine algorithm is used to establish a correlation model between the cell growth curve and the culture medium replacement cycle, to obtain the change of the cell growth curve under different replacement cycle conditions, and to determine the target replacement cycle, including: A cell density detector is used to collect cell density values ​​and fluorescence intensity signals in the culture medium, and the original cell growth data set is obtained according to the fixed time interval recording; Performing outlier detection on the original cell growth data set, supplementing missing data using a cubic spline interpolation method, and obtaining a processed cell growth data set; Calculating the cell density change rate and the cell viability change rate for the processed cell growth data set, and constructing a cell growth feature vector in combination with cell morphology parameters; Extracting historical culture medium replacement time points from a culture medium replacement record database, establishing a correspondence between the cell growth feature vector and the replacement time point, inputting the result into a support vector machine algorithm to obtain a culture medium replacement cycle model, and predicting the cell density integral value and the average value of cell viability at different replacement time intervals according to the culture medium replacement cycle model; It is determined whether the cell density integral value and the cell viability average value are within a preset threshold range. If they are within the preset threshold range, the time point with the maximum cell density integral value and the highest cell viability average value is selected as the target culture medium replacement cycle.

4. The method according to claim 1, characterized in that: The geometric model of the culture medium storage tank and the bioreactor is constructed by using the three-dimensional modeling software, and the historical process parameters of the culture medium supply system and the real-time monitored temperature, pH value and dissolved oxygen concentration are mapped and associated with the geometric model to construct a digital twin model of the reactor, including: A three-dimensional structural model of the storage tank and the bioreactor is constructed using three-dimensional modeling software, wherein the three-dimensional structural model includes geometric data of the feed port, the discharge port, the stirring device, and the sensor installation position; Establish parameter monitoring points based on the surface of the three-dimensional structure model, obtain culture medium supply system data from the process parameter database, and collect real-time monitoring data of the sensor installation position; Dividing the fluid force field, temperature field and concentration field according to the parameter monitoring point, and calculating the distribution values ​​of the fluid force field, temperature field and concentration field in space by using a finite element algorithm; A deep neural network algorithm is used to train the coupling relationship between the fluid force field, temperature field and concentration field, and a reactor digital mapping structure is established according to the coupling relationship to obtain a digital twin model.

5. The method according to claim 1, characterized in that The real-time acquisition of the internal environmental parameters of the culture medium storage tank and the bioreactor, the real-time mapping and synchronization of the internal environmental parameters with the virtual parameters in the digital twin model, the construction of a virtual-real fusion closed-loop control model of the culture medium supply system, and the dynamic adjustment of the flow and concentration of the supply system according to the cell growth requirements and metabolic state, including: The internal environmental parameters of the culture medium storage tank and the bioreactor are collected by using a temperature sensor, a pH sensor, and a dissolved oxygen sensor, and the environmental parameters are subjected to noise reduction processing by using a sliding window method to obtain an environmental parameter data set; According to the environmental parameter data set, glucose concentration and lactate concentration in the cell culture process are collected, and combined with the cell density data to obtain a cell metabolic state data set; The cell metabolic state dataset is used to calibrate the sensor acquisition position in the digital twin structure, and the parameter space distribution is calculated by the Kriging interpolation algorithm to obtain a virtual parameter distribution dataset; The root mean square error value is calculated based on the virtual parameter distribution data set and the environmental parameter data set, and the culture medium supply controller is trained using a deep reinforcement learning algorithm to obtain the culture medium flow compensation value and the concentration compensation value.

6. The method according to claim 1, characterized in that The digital twin model is combined to perform multi-scale multi-physics simulation of the fusion culture medium storage tank and bioreactor during cell growth, simulate and analyze the impact of fluctuations in the redox potential and dissolved oxygen concentration of the culture medium on cell growth, and change the ratio and supply of key nutrients in the culture medium, including: Acquire geometric structure data of the storage tank and the reactor in the digital twin structure, and establish the Navier-Stokes equations describing fluid motion and the convection-diffusion equations describing the mass transfer process based on the geometric structure data; A multi-scale grid structure is established by using an adaptive grid division method according to the Navier-Stokes equations and the convection-diffusion equations, and multiple physical field distribution data are obtained through the multi-scale grid structure; A discrete element algorithm is used to construct a cell-scale calculation module according to the multi-physical field distribution data, and the redox potential distribution and dissolved oxygen concentration gradient data in the microenvironment are obtained through the cell-scale calculation module; A cellular metabolic flux balance equation group is established according to the dissolved oxygen concentration gradient data, and the glucose, amino acid and fatty acid consumption rate and metabolite generation rate data are calculated by the cellular metabolic flux balance equation group.

7. The method according to claim 1, characterized in that The method analyzes historical production data through a machine learning algorithm, extracts the quantitative relationship between key parameters such as cell metabolic rate, glucose concentration in the culture medium, pH value, and reactor temperature, establishes a cell growth prediction model, and predicts the growth trend and metabolic demand of cells within a target time period, including: Acquire culture medium parameter data from a historical database, and process the culture medium parameter data using a maximum and minimum value standardization method to obtain a standardized culture parameter data set; According to the standardized culture parameter data set, a wavelet transform denoising algorithm and a preset decomposition depth are used to perform signal reconstruction on the standardized culture parameter data set to obtain a smoothed data set; For the smoothed data set, eigenvalues ​​and eigenvectors are calculated by principal component analysis, and if the cumulative contribution rate of the eigenvalues ​​reaches a preset threshold, a reduced-dimensional feature data set is obtained; The recursive neural network model is trained using the dimension reduction feature data set. If the prediction error of the validation set is less than a preset threshold, a cell growth prediction model is obtained, and the metabolite generation rate is calculated using the cell growth prediction model.

8. The method according to claim 1, characterized in that The virtual-real mapping mechanism of the culture medium supply system is integrated with the intelligent manufacturing execution system to achieve coordinated optimization control of the culture medium supply process and the entire drug production process, and automatically schedule the culture medium production process according to the production plan and cell growth requirements, including: Collecting production plan scheduling data and cell culture progress data, and generating a production resource status data set according to the scheduling data and progress data; Obtain the maintenance records and fault records of production equipment, and use the equipment reliability evaluation index system to calculate the equipment status evaluation data set; According to the production resource status data set and the equipment status evaluation data set, an integer programming algorithm is used to optimize and calculate to obtain an initial scheduling plan for culture medium production; According to the initial scheduling plan for culture medium production, a deep reinforcement learning algorithm is used to construct a culture medium production scheduler. The production plan priority and equipment status characteristics are processed through the scheduler input layer to obtain a scheduling instruction sequence.

Images