A method and system for automatic water and medium feeding in a fermentation process

Abstract

The application provides a kind of control method and system of automatic water adding and medium adding in fermentation process, multiple-dimensional state parameters reflecting the metabolic state of bacteria and the fermentation environment in the fermentation process are collected, and then a digital twin model of the fermentation process is constructed to realize the virtual-real mapping and visualization of the fermentation state, based on the metabolic kinetics model and the real-time fermentation state, the growth trend of bacteria and the consumption rate of nutrients are predicted, a feed-forward-feedback composite control strategy is used to dynamically determine the feeding time, feeding rate and feeding ratio of water adding and medium adding, the fermentation response efficiency after each feeding is evaluated in real time, and the feeding control parameters are adaptively optimized. By using the scheme of the application, precise, adaptive automatic control of water adding and medium adding in the fermentation process can be realized, and the fermentation efficiency and product yield are improved.

Description

Technical Field

[0001] This invention relates to the field of bio-fermentation technology, and in particular to a method and system for automatically adding water and culture medium during the fermentation process. Background Technology

[0002] Fermentation is a key technological process that utilizes the metabolic activities of microorganisms to produce target products, and it is widely used in medicine, food, chemical industry, and agriculture. During fermentation, nutrients in the culture medium are continuously consumed as the microorganisms grow, reproduce, and metabolize. The volume of the fermentation broth also changes due to factors such as evaporation and sampling. Therefore, it is necessary to replenish water and fresh culture medium in a timely manner to maintain a suitable growth environment and nutrient supply for the microorganisms, thereby ensuring the efficient progress of the fermentation process and the high yield of the target product.

[0003] Currently, the control of water and culture medium addition during fermentation mainly employs two methods. The first is experience-based manual, timed, and quantitative feeding. This method is simple to operate, but heavily relies on the operator's experience level. The accuracy of feeding timing and quantity is difficult to guarantee, easily leading to problems such as insufficient nutrient supply resulting in limited cell growth, or excessive nutrient supply leading to the accumulation of metabolic byproducts and abnormal cell growth. Moreover, manual operation has a slow response speed and cannot adapt to the dynamic changes in the fermentation process. The second method is feedback control based on a single parameter, such as triggering feeding based on changes in dissolved oxygen concentration or pH value. Although this method can achieve a certain degree of automatic control, it relies on a single parameter for judgment and lacks a comprehensive perception of the overall metabolic state of the cells. Furthermore, it is a lagging feedback regulation that cannot predict nutrient requirements in advance, easily causing fluctuations in the fermentation process and affecting product yield and fermentation stability. Therefore, how to achieve precise and adaptive automatic control of water and culture medium addition during fermentation has become a challenge for the industry. Summary of the Invention

[0004] Based on this, the present invention provides a control method and system for automatically adding water and culture medium during the fermentation process, which can achieve precise and adaptive automatic control of adding water and culture medium during the fermentation process.

[0005] In a first aspect, the present invention provides a method for controlling the automatic addition of water and culture medium during a fermentation process, comprising the following steps: Collect multi-dimensional state parameters reflecting the metabolic state of microorganisms and the fermentation environment during the fermentation process; A digital twin model of the fermentation process is constructed, and the collected multi-dimensional state parameters are mapped to the digital twin model in real time to realize the virtual-real mapping and visualization of the fermentation state in virtual space. Based on the metabolic kinetics model and real-time fermentation status, predict cell growth trends and nutrient consumption rates. A feedforward-feedback composite control strategy is adopted to dynamically determine the timing, feeding rate, and water-to-material ratio of water and culture medium addition based on the predicted nutrient requirements and real-time state deviation. Real-time evaluation of fermentation response performance after each feeding, and adaptive optimization of feeding control parameters based on performance evaluation results.

[0006] In some embodiments, collecting multi-dimensional state parameters reflecting the metabolic state of microorganisms and the fermentation environment during fermentation specifically includes: Deploy multiple types of sensors at key detection points in the fermenter; By periodically collecting detection data from various sensors through a data acquisition system, multi-dimensional state parameters reflecting the metabolic state of the microorganisms and the fermentation environment are obtained.

[0007] In some embodiments, a digital twin model of the fermentation process is constructed, and the collected multi-dimensional state parameters are mapped to the digital twin model in real time to realize the virtual-real mapping and visualization of the fermentation state in virtual space. Specifically, this includes: A digital twin model of the fermentation process is constructed based on the fermenter equipment parameters, process principles and microbial metabolic characteristics. The collected multi-dimensional state parameters are mapped to the corresponding state variables of the digital twin model in real time; The simulation and visualization engines that drive the digital twin model enable the virtual-real mapping and 3D visualization of the fermentation state.

[0008] In some embodiments, predicting cell growth trends and nutrient consumption rates based on metabolic kinetics models and real-time fermentation status specifically includes: Construct a fermentation metabolic kinetic model that includes cell growth, substrate consumption, and product synthesis; The model parameters and the current state are updated based on the real-time state parameters using a state estimation algorithm. Based on the updated metabolic kinetics model, the cell growth trend and nutrient consumption rate in future periods are predicted.

[0009] In some embodiments, a feedforward-feedback composite control strategy is adopted to dynamically determine the timing, feeding rate, and water-to-material ratio of water and culture medium addition based on the predicted nutrient requirements and real-time state deviation. Specifically, this includes: Based on the predicted nutrient consumption rate, the feedforward amount required to meet the cell growth needs is calculated. Based on the deviation between real-time status parameters and set target values, calculate the feedback correction feed amount; By combining the feedforward feed rate and the feedback correction feed rate, the timing, feed rate, and water-to-material ratio for adding water and culture medium are dynamically determined.

[0010] In some embodiments, the fermentation response performance after each feeding is evaluated in real time, and the feeding control parameters are adaptively optimized based on the performance evaluation results, specifically including: Monitor changes in key status parameters within the set response period after each material replenishment; Calculate the feeding efficiency evaluation index based on the state changes before and after feeding; Based on the feeding efficiency evaluation index, the metabolic kinetic model parameters and feedforward-feedback control gain are adaptively adjusted.

[0011] In some embodiments, the multidimensional state parameters include dissolved oxygen concentration, pH value, cell concentration, carbon dioxide release rate, redox potential, fermentation broth temperature, and fermentation broth volume.

[0012] In some embodiments, the feeding timing includes the feeding start time in continuous feeding mode and the feeding trigger time in intermittent feeding mode; the feeding rate includes the amount of water added and the amount of culture medium added per unit time; the water-to-culture medium ratio is the volume ratio or mass ratio of the added water to the culture medium.

[0013] In some embodiments, the feeding efficiency evaluation index comprehensively considers the response of three dimensions: cell growth rate, substrate consumption rate, and product synthesis rate.

[0014] In a second aspect, the present invention provides a control system for automatically adding water and culture medium during a fermentation process, comprising: The data acquisition module is used to collect multi-dimensional state parameters that reflect the metabolic state of the microorganisms and the fermentation environment during the fermentation process; The modeling module is used to construct a digital twin model of the fermentation process, and to map the collected multi-dimensional state parameters to the digital twin model in real time, so as to realize the virtual-real mapping and visualization of the fermentation state in the virtual space. The prediction module is used to predict cell growth trends and nutrient consumption rates based on metabolic kinetics models and real-time fermentation status. The control module is used to dynamically determine the timing, feeding rate, and water-to-material ratio of water and culture medium addition based on the predicted nutrient requirements and real-time state deviation using a feedforward-feedback composite control strategy. The optimization module is used to evaluate the fermentation response efficiency after each feeding in real time and adaptively optimize the feeding control parameters based on the efficiency evaluation results.

[0015] Compared with the prior art, the present invention has the following advantages: The automatic water and culture medium addition control method and system for the fermentation process provided by this invention firstly collects multi-dimensional state parameters reflecting the metabolic state of the cells and the fermentation environment during the fermentation process. This step enables multi-dimensional and comprehensive state perception of the fermentation process, providing a comprehensive and accurate data foundation for subsequent modeling, prediction, and control, ensuring that the system can monitor the metabolic state of the cells and changes in the fermentation environment in real time. Secondly, a digital twin model of the fermentation process is constructed, and the collected multi-dimensional state parameters are mapped to the digital twin model in real time to achieve virtual-real mapping and visualization of the fermentation state in virtual space. This step can establish an accurate virtual mirror of the physical fermentation process through digital twin technology, realizing intuitive display and collaborative monitoring of the fermentation state, enabling operators to grasp the overall fermentation situation from the system level, and providing model support for subsequent simulation prediction and virtual debugging. Then, based on the metabolic kinetic model and real-time fermentation state, the cell growth trend and nutrient consumption rate are predicted. This step can predict the cell growth trend and nutrient demand in advance by combining the metabolic kinetic model with real-time state estimation, transforming traditional lag feedback control into forward-looking predictive control. This invention provides a scientific basis for feeding decisions, overcoming the blindness and lag of traditional experience-based control and single-parameter feedback control. Next, a feedforward-feedback composite control strategy is adopted. Based on predicted nutrient requirements and real-time state deviations, the timing, rate, and water-to-material ratio of water and culture medium addition are dynamically determined. This step allows feedforward control to meet predicted nutrient requirements for cell growth in advance, while feedback control corrects deviations from the target in real time. The combination of these two approaches achieves precise and dynamic control of feeding, ensuring both timely nutrient supply and stable fermentation, effectively improving fermentation efficiency and product yield. Finally, the fermentation response performance after each feeding is evaluated in real time, and the feeding control parameters are adaptively optimized based on the evaluation results. This step, through closed-loop evaluation of the feeding effect and adaptive parameter optimization, enables the system to continuously learn and adapt to the characteristics of different fermentation batches and different cell growth stages, continuously improving control performance and achieving self-learning and adaptive capabilities of the feeding control system. In summary, the solution of this invention can achieve precise and adaptive automatic control of water and culture medium addition during the fermentation process, improving fermentation efficiency and product yield. Attached Figure Description

[0016] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is an exemplary flowchart of a control method for automatically adding water and culture medium during a fermentation process, as shown in some embodiments of the present invention. Figure 2 This is a schematic diagram illustrating an application scenario of an automatic water and culture medium addition control system for a fermentation process, as shown in some embodiments of the present invention. Figure 3 This is a flowchart illustrating the process of determining the evaluation index of feeding efficiency according to some embodiments of the present invention; Figure 4 This is a schematic diagram of the structure of a control system for automatically adding water and culture medium during the fermentation process, as shown in some embodiments of the present invention. Figure 5 This is a schematic diagram of the structure of a computer device for controlling the automatic addition of water and culture medium during the fermentation process, as shown in some embodiments of the present invention. Detailed Implementation

[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0018] The present invention will be further described in detail below with reference to the accompanying drawings.

[0019] refer to Figure 1 The figure is an exemplary flowchart of a control method for automatically adding water and culture medium during a fermentation process according to some embodiments of the present invention. This control method for automatically adding water and culture medium during a fermentation process mainly includes the following steps: In step 101, multi-dimensional state parameters reflecting the metabolic state of the cells and the fermentation environment are collected during the fermentation process.

[0020] In practice, collecting multi-dimensional state parameters reflecting the metabolic state of the microorganisms and the fermentation environment during the fermentation process can be achieved in the following ways: First, deploy corresponding multi-type sensors at key detection points such as the top, middle, and bottom of the fermenter. Specifically, for dissolved oxygen concentration detection, a fluorescence dissolved oxygen sensor is installed in the fermentation broth area in the middle of the fermenter to detect the dissolved oxygen concentration in the fermentation broth in real time. This parameter indirectly reflects the respiration intensity and metabolic activity of the microorganisms. For pH detection, a composite pH electrode is installed in the middle of the fermenter to detect the acidity or alkalinity of the fermentation broth in real time. This parameter reflects the acid or alkali production during the microorganisms' metabolism. For microorganism concentration detection, an online turbidity sensor or near-infrared spectroscopy sensor installed on the side wall of the fermenter can be used for real-time detection. This parameter directly reflects the growth density and biomass of the microorganisms. For carbon dioxide release rate detection, an infrared carbon dioxide analyzer is installed in the exhaust pipe of the fermenter to detect the carbon dioxide concentration in the exhaust gas in real time. The carbon dioxide release rate is calculated by combining the ventilation rate. This parameter reflects the overall metabolic intensity of the microorganisms. For redox potential detection, a sensor is installed in the middle of the fermenter... A redox potential electrode is installed to detect the redox state of the fermentation broth in real time. This parameter reflects the metabolic environment and energy metabolism level of the cells. For the detection of fermentation broth temperature, platinum resistance temperature sensors are installed at the top, middle, and bottom of the fermenter to detect the temperature of the fermentation broth at different locations in real time. For the detection of fermentation broth volume, a hydrostatic level sensor or radar level gauge installed on the fermenter can be used for real-time detection, and the fermentation broth volume can be calculated by combining the geometric parameters of the fermenter. Then, all sensors are connected to the field data acquisition unit, and the detection data of each sensor is collected periodically according to the set sampling period, such as one minute, via Modbus or PROFIBUS industrial fieldbus. Finally, the collected data are classified and organized according to parameter type to form a complete dataset including dissolved oxygen concentration, pH value, cell concentration, carbon dioxide release rate, redox potential, fermentation broth temperature, and fermentation broth volume. This dataset is stored in the field data acquisition server for subsequent transmission and processing, thereby completing the acquisition of multi-dimensional state parameters of the fermentation process. Other methods can also be used in other embodiments, which are not limited here.

[0021] It should be noted that the multidimensional state parameters in this invention refer to a collection of various physical and chemical quantities collected during the fermentation process to comprehensively characterize the metabolic state of the cells and the fermentation environment. These parameters reflect the operating state of the fermentation process from different perspectives, providing comprehensive data support for subsequent state assessment, prediction and control.

[0022] In some embodiments, reference Figure 2As shown in the figure, this diagram illustrates an application scenario of the automatic water and culture medium addition control system for the fermentation process, as illustrated in some embodiments of the present invention. The figure includes three main components: a field acquisition and execution device, an edge control server, and a cloud data storage device. The field acquisition and execution device is responsible for collecting multi-dimensional state parameters of the fermentation process and executing the water and culture medium addition operations according to control commands. The field acquisition and execution device sends the collected data to the edge control server through an industrial communication network. The automatic water and culture medium addition control system for the fermentation process runs on the edge control server. The edge control server stores the processing results and historical data in the cloud data storage device and performs remote visual monitoring and management.

[0023] In step 102, a digital twin model of the fermentation process is constructed, and the collected multi-dimensional state parameters are mapped to the digital twin model in real time to realize the virtual-real mapping and visualization of the fermentation state in virtual space.

[0024] In some embodiments, constructing a digital twin model of the fermentation process and mapping the collected multi-dimensional state parameters to the digital twin model in real time to achieve virtual-real mapping and visualization of the fermentation state in virtual space can be achieved through the following steps: A digital twin model of the fermentation process is constructed based on the fermenter equipment parameters, process principles and microbial metabolic characteristics. The collected multi-dimensional state parameters are mapped to the corresponding state variables of the digital twin model in real time; The simulation and visualization engines that drive the digital twin model enable the virtual-real mapping and 3D visualization of the fermentation state.

[0025] In specific implementation, based on the fermenter equipment parameters, process principles, and microbial metabolic characteristics, a digital twin model of the fermentation process can be constructed in the following manner: First, collect the equipment specifications of the fermenter, including tank geometry, agitator type and power, aeration system parameters, heat exchange system parameters, and feeding system parameters; then, based on the principles of fluid mechanics, heat transfer, and mass transfer, construct simulation models of the flow field, temperature field, and concentration field within the fermenter. This model can simulate the mixing state, temperature distribution, and mass transfer of the fermentation broth; next, combining the metabolic kinetics of the microorganisms, construct a biochemical reaction kinetic model that includes microbial growth, substrate consumption, product synthesis, and the generation of metabolic byproducts; subsequently, establish a three-dimensional geometric model of the fermenter and a three-dimensional model of the equipment, and construct a visualized three-dimensional scene; finally, integrate the equipment simulation model, the biochemical reaction kinetic model, and the three-dimensional visualized scene to form a digital twin model of the fermentation process. This model can perform both numerical simulation of the fermentation process and three-dimensional visualization. Other methods can also be used in other embodiments, and are not limited here.

[0026] It should be noted that the digital twin model of the fermentation process in this invention refers to a high-fidelity simulation model constructed in virtual space that corresponds one-to-one with the physical fermentation process. It can simulate fluid flow, mass transfer and biochemical reactions in the fermentation process, reflect the fermentation state and provide three-dimensional visualization, providing virtual support for the monitoring, prediction and optimization of the fermentation process.

[0027] In specific implementation, the real-time mapping of collected multi-dimensional state parameters to the corresponding state variables of the digital twin model can be achieved in the following way: First, the multi-dimensional state parameters collected on-site are received through the data access layer, and the data is preprocessed such as format conversion, outlier filtering, and missing value interpolation; then, according to the parameter type identifier of the data source, the real-time data is mapped to the corresponding state variables in the digital twin model, for example, the dissolved oxygen concentration is mapped to the dissolved oxygen concentration state variable in the model, the cell concentration is mapped to the cell biomass state variable in the model, and the temperature is mapped to the temperature field state variable in the model; next, a data assimilation algorithm, such as extended Kalman filtering or ensemble Kalman filtering, is used to correct the state variables of the model using real-time observation data, so that the state of the virtual model is consistent with the actual state of the physical fermentation process; finally, the real-time mapping of multi-dimensional state parameters to the digital twin model is completed, ensuring that the digital twin model can accurately reflect the current actual fermentation state. Other methods can also be used in other embodiments, which are not limited here.

[0028] In specific implementation, the simulation engine and visualization engine driving the digital twin model to achieve virtual-real mapping and 3D visualization of the fermentation state can be implemented in the following way: First, the simulation engine driving the digital twin model performs the next simulation calculation based on the updated state variables and model parameters to predict the changing trend of the fermentation state; simultaneously, the 3D visualization engine is driven to update the visual representation of the 3D scene according to the real-time state of the digital twin model. Specific visualization updates include: dynamically refreshing and displaying the real-time values ​​and change curves of each state parameter at the corresponding positions of the fermenter 3D model; changing the color depth and transparency of the fermentation broth according to the cell concentration; changing the color of different areas of the fermentation broth according to the distribution of dissolved oxygen concentration; displaying a temperature field cloud map according to the temperature distribution; and displaying the fluid flow effect of the feeding pipeline according to the operating status of the feeding system. Finally, in the visualization interface of the digital twin model, the real-time operating status, key parameter indicators, and overall process flow of the fermentation process are simultaneously displayed in the form of a 3D scene combined with a data dashboard, thereby achieving virtual-real mapping and 3D visualization of the fermentation state. Other methods can also be used in other embodiments, and are not limited here.

[0029] It should be noted that the above steps can establish an accurate virtual mirror of the physical fermentation process through digital twin technology, enabling intuitive display and collaborative monitoring of the fermentation status. This allows operators to grasp the overall fermentation situation from a system perspective, while also providing model support for subsequent simulation prediction and virtual debugging.

[0030] In step 103, based on the metabolic kinetics model and real-time fermentation status, the cell growth trend and nutrient consumption rate are predicted.

[0031] In some embodiments, predicting cell growth trends and nutrient consumption rates based on metabolic kinetics models and real-time fermentation status specifically includes: Construct a fermentation metabolic kinetic model that includes cell growth, substrate consumption, and product synthesis; The model parameters and the current state are updated based on the real-time state parameters using a state estimation algorithm. Based on the updated metabolic kinetics model, the cell growth trend and nutrient consumption rate in future periods are predicted.

[0032] It should be noted that the metabolic kinetic model in this invention is a mathematical model that describes the quantitative relationship between cell growth, substrate consumption, and product synthesis during fermentation. Its core is to characterize the changes of each state variable over time through a series of differential equations, which is the basis for predicting and optimizing the fermentation process. Cell growth trend refers to the direction and rate of change of cell concentration or biomass over a future period of time, reflecting the growth status and development direction of the cells. Nutrient consumption rate refers to the amount of nutrients consumed by the cells per unit time, reflecting the rate at which the cells utilize the nutrients in the culture medium.

[0033] In practical implementation, constructing a fermentation metabolic kinetic model that includes cell growth, substrate consumption, and product synthesis can be achieved in the following way: First, based on the basic principles of microbial growth kinetics, construct a kinetic equation for cell growth, using the Monod equation or its improved form to describe the relationship between specific growth rate and limiting substrate concentration, while considering the effects of substrate inhibition, product inhibition, and environmental factors such as temperature and pH on the growth rate; then, construct a kinetic equation for substrate consumption, dividing substrate consumption into maintenance consumption for cell growth and metabolic consumption for product synthesis, and establishing corresponding consumption rate equations for each; next, construct... A kinetic equation for product synthesis is established. Based on the relationship between product synthesis and cell growth, a growth-coupled, partially growth-coupled, or non-growth-coupled kinetic model is used to describe the product synthesis rate. Subsequently, considering the volume changes during fermentation, a volume change equation incorporating feeding and sampling is established. Finally, the above equations are integrated to form a complete fermentation metabolic kinetic model that includes cell growth, substrate consumption, product synthesis, and volume changes. This model is expressed in the form of a set of ordinary differential equations, which can describe the dynamic changes of key state variables over time during fermentation. Other methods can also be used in other embodiments, which are not limited here.

[0034] In specific implementation, updating model parameters and the current state based on real-time state parameters using a state estimation algorithm can be achieved in the following way: First, select directly observable state variables from the multi-dimensional state parameters collected on-site as observation vectors, including dissolved oxygen concentration, pH value, cell concentration, carbon dioxide release rate, and fermentation broth volume; then, combine the parameters to be estimated in the metabolic kinetic model with the state variables to form an extended state vector, where the parameters to be estimated include maximum specific growth rate, substrate affinity constant, yield coefficient, and maintenance coefficient; next, use an extended Kalman filter algorithm or an unscented Kalman filter algorithm to recursively estimate the extended state vector using real-time observation data, continuously correcting the model's state variables and parameter values ​​so that the model output can optimally fit the actual observation data; finally, obtain the updated model parameters and the current state estimate, ensuring that the metabolic kinetic model can accurately reflect the actual characteristics of the current fermentation batch. Other methods can also be used in other embodiments, which are not limited here.

[0035] In specific implementation, based on the updated metabolic kinetic model, the prediction of bacterial growth trend and nutrient consumption rate in future periods can be achieved in the following way: First, the current state estimate is used as the initial condition, and the updated model parameters are used as the model parameters; then, a numerical integration method, such as the fourth-order Runge-Kutta method, is used to solve the differential equations of the metabolic kinetic model, and the state changes in the future are predicted by forward integration; next, the bacterial concentration change curve over time is extracted from the predicted state variable time series, which is the bacterial growth trend prediction result; at the same time, the predicted nutrient consumption rate per unit time is calculated based on the predicted substrate concentration change and bacterial biomass change; finally, the bacterial growth trend curve and nutrient consumption rate prediction value in the future set prediction period are output, providing a forward-looking basis for subsequent feeding decisions. Other methods can also be used in other embodiments, which are not limited here.

[0036] It should be noted that the above steps can be combined with metabolic kinetic models and real-time state estimation to predict the growth trend and nutritional needs of the bacteria in advance, transforming the traditional lagging feedback control into a forward-looking predictive control, thereby providing a scientific basis for feeding decisions and overcoming the blindness and lag of traditional experience control and single-parameter feedback control.

[0037] In step 104, a feedforward-feedback composite control strategy is adopted to dynamically determine the timing, feeding rate, and water-to-material ratio of water and culture medium addition based on the predicted nutrient requirements and real-time state deviation.

[0038] In some embodiments, a feedforward-feedback composite control strategy is adopted to dynamically determine the timing, feeding rate, and water-to-material ratio of water and culture medium addition based on the predicted nutrient requirements and real-time state deviation. Specifically, this includes: Based on the predicted nutrient consumption rate, the feedforward amount required to meet the cell growth needs is calculated. Based on the deviation between real-time status parameters and set target values, calculate the feedback correction feed amount; By combining the feedforward feed rate and the feedback correction feed rate, the timing, feed rate, and water-to-material ratio for adding water and culture medium are dynamically determined.

[0039] It should be noted that the feedforward-feedback composite control strategy in this invention refers to a control method that combines feedforward control and feedback control. Feedforward control adjusts in advance based on the prediction of disturbances to eliminate the influence of predictable disturbances, while feedback control adjusts based on the deviation between the actual output and the target to eliminate the influence of unpredictable disturbances and model errors. The combination of the two can achieve fast and accurate control effects. The feeding timing refers to the time point when the feeding operation is started or the condition that triggers feeding, including the feeding start time in continuous feeding mode and the feeding trigger time in intermittent feeding mode. The feeding rate refers to the volume or mass of water or culture medium added per unit time. The water-to-material ratio refers to the volume ratio or mass ratio of water to culture medium added, used to adjust the concentration of added nutrients.

[0040] In specific implementation, the feedforward amount to meet the growth requirements of the microorganisms, based on the predicted nutrient consumption rate, can be calculated in the following way: First, obtain the predicted nutrient consumption rate for the future control cycle from the prediction module, including the consumption rate of carbon source, nitrogen source, and key trace elements; then, based on the set target concentration range of each nutrient in the fermentation broth and the predicted consumption rate, calculate the feedforward amount required to maintain the target concentration; specifically, the calculation of the feedforward amount needs to consider the predicted microbial growth, the corresponding substrate consumption, and the volume loss caused by evaporation, sampling, etc., and obtain the total amount of culture medium and water to be added through material balance; next, according to the feedforward control cycle, the total feedforward amount is evenly distributed to each control cycle to obtain the feedforward feedforward rate; finally, output the feedforward control feedforward rate setpoint, including the feedforward rate of water addition and the feedforward rate of culture medium addition. This feedforward amount can meet the predicted nutrient requirements for microbial growth in advance. Other methods can also be used in other embodiments, which are not limited here.

[0041] In specific implementation, the feedback correction feed amount can be calculated based on the deviation between real-time state parameters and set target values ​​in the following manner: First, key state parameters are selected as controlled variables for feedback control, including dissolved oxygen concentration, cell growth rate, substrate concentration, and fermentation broth volume; then, the actual measured values ​​of each controlled variable are acquired in real time and compared with the corresponding set target values ​​to calculate the deviation values ​​of each variable; next, a proportional-integral-derivative (PID) control algorithm or a fuzzy adaptive PID control algorithm is used to calculate the feedback correction amount based on the deviation values ​​and their rates of change of each variable; for the deviation of dissolved oxygen concentration, the feeding rate is adjusted to indirectly affect the respiration intensity of the cells, thereby regulating the dissolved oxygen level; for the deviation of substrate concentration, the feeding rate of the culture medium is directly adjusted; for the deviation of volume, the water supply is adjusted; finally, the correction amounts of each feedback channel are weighted and combined to obtain the total feedback correction feed amount, including the water addition feedback correction amount and the culture medium addition feedback correction amount. This feedback correction amount can correct the deviation between the actual state and the target in real time. Other methods can also be used in other embodiments, which are not limited here.

[0042] In practice, the timing, rate, and water-to-material ratio of water and culture medium additions can be dynamically determined by combining the feedforward and feedback-corrected feed rates: First, the feedforward and feedback-corrected feed rates are superimposed to obtain the total feed rate, including the total water addition rate and the total culture medium addition rate. Then, the feeding timing is determined based on the fermentation stage and control mode. For continuous feeding mode, continuous feeding is initiated after entering the logarithmic growth phase, and the feed rate is dynamically adjusted according to the calculated value. For intermittent feeding mode, when the substrate concentration is below the set lower limit or the cell concentration is below the set lower limit, continuous feeding is initiated. When the concentration reaches a set threshold, feeding is triggered, and feeding is completed within a set time according to the calculated total amount of feed. Then, the water-to-material ratio, i.e., the volume ratio or mass ratio of the added water to the culture medium, is calculated based on the total water addition rate and the total culture medium addition rate. This ratio is dynamically adjusted according to the deviation between the target concentration and the actual concentration of the fermentation broth. Finally, a complete feeding control command containing the feeding timing, feeding rate, and water-to-material ratio is output and sent to the feeding actuator for execution, thereby realizing dynamic and precise control of water and culture medium addition. Other methods can also be used in other embodiments, which are not limited here.

[0043] It should be noted that the above steps can meet the predicted nutritional needs of cell growth in advance through feedforward control, and correct the deviation between the actual state and the target in real time through feedback control. The combination of the two achieves precise and dynamic control of feeding, which not only ensures the timeliness of nutrient supply, but also ensures the stability of fermentation state, effectively improving fermentation efficiency and product yield.

[0044] In step 105, the fermentation response efficiency after each feeding is evaluated in real time, and the feeding control parameters are adaptively optimized based on the efficiency evaluation results.

[0045] In some embodiments, the fermentation response performance after each feeding is evaluated in real time, and the feeding control parameters are adaptively optimized based on the performance evaluation results, specifically including: Monitor changes in key status parameters within the set response period after each material replenishment; Calculate the feeding efficiency evaluation index based on the state changes before and after feeding; Based on the feeding efficiency evaluation index, the metabolic kinetic model parameters and feedforward-feedback control gain are adaptively adjusted.

[0046] In some embodiments, reference Figure 3 As shown in the figure, this is a flowchart illustrating the process of determining the evaluation index of feeding efficiency in some embodiments of the present invention. In this embodiment, the real-time evaluation of the fermentation response efficiency after each feeding can be achieved by the following steps: In step 1031, the changes in key status parameters are monitored within a set response period after each material replenishment. In step 1032, the feeding efficiency evaluation index is calculated based on the state changes before and after feeding; In step 1033, the metabolic kinetic model parameters and feedforward-feedback control gain are adaptively adjusted based on the feeding efficiency evaluation index.

[0047] It should be noted that the feeding efficiency evaluation index in this invention is a comprehensive index that quantifies the impact of each feeding operation on the fermentation process. It can reflect the effectiveness and rationality of the feeding operation and is an important basis for achieving adaptive optimization of feeding control parameters. The set response period refers to the time required for the cells to produce a significant metabolic response after the feeding operation. It can be set according to the cell type and fermentation characteristics, and is usually several hours to more than ten hours.

[0048] In specific implementation, monitoring changes in key state parameters within a set response period after each feeding operation can be achieved in the following way: First, record feeding operation information such as the start time, end time, total feeding amount, and feeding rate of each feeding operation; then, starting from the end time of the feeding operation, initiate a monitoring window with a duration of the set response period; within the monitoring window, continuously collect data on key state parameters at a set sampling frequency, including cell concentration, substrate concentration, dissolved oxygen concentration, carbon dioxide release rate, and product concentration; simultaneously, record the corresponding parameter data within the same time period before the feeding operation as baseline data before feeding; finally, obtain the time series data of each key state parameter before and after feeding for subsequent performance evaluation. Other methods can also be used in other embodiments, which are not limited here.

[0049] In practice, the calculation of feeding efficiency evaluation indicators based on state changes before and after feeding can be achieved in the following way: First, calculate the rate of change of each key state parameter before and after feeding; for cell concentration, calculate the cell specific growth rate before and after feeding, i.e., the amount of growth per unit biomass per unit time, which reflects the growth activity of the cells; for substrate concentration, calculate the substrate specific consumption rate before and after feeding, i.e., the amount of substrate consumed per unit biomass per unit time, which reflects the efficiency of nutrient utilization by the cells; for product concentration, calculate the product specific synthesis rate before and after feeding, i.e., the amount of product consumed per unit biomass per unit time. The product synthesis amount within a time period is an indicator that reflects the product generation efficiency. Then, weighting coefficients are assigned to the above three dimensions of indicators. The weighting coefficients are adjusted according to the fermentation target. For example, when the product yield is the target, the weight of the product synthesis rate is increased, and when the cell growth is the target, the weight of the cell growth rate is increased. Next, a weighted summation method is used to sum the three dimensions of indicators to obtain a comprehensive feeding efficiency evaluation index. Finally, the value of the feeding efficiency evaluation index is output. The higher the index value, the better the feeding effect. Other methods can also be used in other embodiments, which are not limited here.

[0050] It should be noted that the feeding efficiency evaluation index comprehensively considers the response of three dimensions: cell growth rate, substrate consumption rate, and product synthesis rate. These three dimensions reflect the effect of feeding operation from the aspects of growth activity, nutrient utilization efficiency, and product formation efficiency, respectively, thereby achieving a comprehensive and objective evaluation of feeding efficiency.

[0051] In specific implementation, the adaptive adjustment of metabolic kinetic model parameters and feedforward-feedback control gain based on the feeding efficiency evaluation index can be achieved in the following way: First, the calculated feeding efficiency evaluation index is compared with the set expected efficiency value to obtain the efficiency deviation; then, based on the magnitude and direction of the efficiency deviation, the key parameters of the metabolic kinetic model are adjusted using gradient descent or particle swarm optimization algorithms, including the maximum specific growth rate, substrate yield coefficient, and product yield coefficient, so as to continuously improve the prediction accuracy of the model; next, also based on the efficiency deviation, the control gain of the feedforward-feedback composite controller is adaptively adjusted, including the feedforward gain coefficient, the proportional coefficient, integral coefficient, and derivative coefficient of the PID controller, etc. When the efficiency index is lower than the expected value, the feedback gain is appropriately increased to improve the control response speed; when the efficiency index is higher than the expected value and the system is stable, the gain is appropriately decreased to avoid overshoot; finally, through the above adaptive adjustment, the prediction accuracy of the metabolic kinetic model and the control performance of the feedforward-feedback controller can be continuously optimized as the fermentation process progresses, thereby continuously improving the effect of feeding control. Other methods can also be used in other embodiments, which are not limited here.

[0052] It should be noted that the above steps, through closed-loop evaluation of the feeding effect and adaptive parameter optimization, enable the system to continuously learn and adapt to the characteristics of different fermentation batches and different cell growth stages, thereby continuously improving control performance and realizing the self-learning and adaptive capabilities of the feeding control system.

[0053] Furthermore, in another aspect of the present invention, in some embodiments, the present invention provides a control system for automatically adding water and culture medium during the fermentation process, see reference. Figure 4 The figure is a schematic diagram of the structure of a control system for automatically adding water and culture medium during fermentation, according to some embodiments of the present invention. This control system includes: a data acquisition module 401, a modeling module 402, a prediction module 403, a control module 404, and an optimization module 405, which are described below: The acquisition module 401 in this invention is mainly used to acquire multi-dimensional state parameters reflecting the metabolic state of the cells and the fermentation environment during the fermentation process; Modeling module 402: In this invention, modeling module 402 is mainly used to construct a digital twin model of the fermentation process, and to map the collected multi-dimensional state parameters to the digital twin model in real time, so as to realize the virtual-real mapping and visualization of the fermentation state in virtual space. Prediction module 403, in this invention, is mainly used to predict cell growth trend and nutrient consumption rate based on metabolic kinetic model and real-time fermentation status. Control module 404, in this invention, is mainly used to adopt a feedforward-feedback composite control strategy to dynamically determine the timing, feeding rate and water-to-material ratio of adding water and culture medium based on the predicted nutrient requirements and real-time state deviation. The optimization module 405 in this invention is mainly used to evaluate the fermentation response efficiency after each feeding in real time, and adaptively optimize the feeding control parameters based on the efficiency evaluation results.

[0054] The various modules in the control system for automatically adding water and culture medium during the fermentation process described above can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0055] In another embodiment, the present invention provides a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The database stores historical operating data, model parameters, and control parameters for the automatic water and culture medium addition control of the fermentation process. The network interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a method for automatically adding water and culture medium to a fermentation process.

[0056] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the computer device to which the present invention is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0057] In one embodiment, a computer device is also provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described control method embodiment for automatically adding water and culture medium during the fermentation process.

[0058] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described control method embodiment for automatically adding water and culture medium during the fermentation process.

[0059] In one embodiment, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps in the above-described embodiment of the control method for automatically adding water and culture medium during a fermentation process.

[0060] In one embodiment, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps described in the embodiment of the modular livestock farm wastewater treatment plant operation and management method.

[0061] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0062] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0063] The technical solutions provided by the embodiments disclosed in this invention have the following beneficial effects: The automatic water and culture medium addition control method and system for the fermentation process provided by this invention firstly collects multi-dimensional state parameters reflecting the metabolic state of the cells and the fermentation environment during the fermentation process. This step enables multi-dimensional and comprehensive state perception of the fermentation process, providing a comprehensive and accurate data foundation for subsequent modeling, prediction, and control, ensuring that the system can monitor the metabolic state of the cells and changes in the fermentation environment in real time. Secondly, a digital twin model of the fermentation process is constructed, and the collected multi-dimensional state parameters are mapped to the digital twin model in real time to achieve virtual-real mapping and visualization of the fermentation state in virtual space. This step can establish an accurate virtual mirror of the physical fermentation process through digital twin technology, realizing intuitive display and collaborative monitoring of the fermentation state, enabling operators to grasp the overall fermentation situation from the system level, and providing model support for subsequent simulation prediction and virtual debugging. Then, based on the metabolic kinetic model and real-time fermentation state, the cell growth trend and nutrient consumption rate are predicted. This step can predict the cell growth trend and nutrient demand in advance by combining the metabolic kinetic model with real-time state estimation, transforming traditional lag feedback control into forward-looking predictive control. This invention provides a scientific basis for feeding decisions, overcoming the blindness and lag of traditional experience-based control and single-parameter feedback control. Next, a feedforward-feedback composite control strategy is adopted. Based on predicted nutrient requirements and real-time state deviations, the timing, rate, and water-to-material ratio of water and culture medium addition are dynamically determined. This step allows feedforward control to meet predicted nutrient requirements for cell growth in advance, while feedback control corrects deviations from the target in real time. The combination of these two approaches achieves precise and dynamic control of feeding, ensuring both timely nutrient supply and stable fermentation, effectively improving fermentation efficiency and product yield. Finally, the fermentation response performance after each feeding is evaluated in real time, and the feeding control parameters are adaptively optimized based on the evaluation results. This step, through closed-loop evaluation of the feeding effect and adaptive parameter optimization, enables the system to continuously learn and adapt to the characteristics of different fermentation batches and different cell growth stages, continuously improving control performance and achieving self-learning and adaptive capabilities of the feeding control system. In summary, the solution of this invention can achieve precise and adaptive automatic control of water and culture medium addition during the fermentation process, improving fermentation efficiency and product yield.

[0064] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A method for automatically adding water and culture medium during a fermentation process, characterized in that, Includes the following steps: Collect multi-dimensional state parameters reflecting the metabolic state of microorganisms and the fermentation environment during the fermentation process; A digital twin model of the fermentation process is constructed, and the collected multi-dimensional state parameters are mapped to the digital twin model in real time to realize the virtual-real mapping and visualization of the fermentation state in virtual space. Based on the metabolic kinetics model and real-time fermentation status, predict cell growth trends and nutrient consumption rates. A feedforward-feedback composite control strategy is adopted to dynamically determine the timing, feeding rate, and water-to-material ratio of water and culture medium addition based on the predicted nutrient requirements and real-time state deviation. Real-time evaluation of fermentation response performance after each feeding, and adaptive optimization of feeding control parameters based on performance evaluation results.

2. The automatic water and culture medium addition control method for the fermentation process as described in claim 1, characterized in that, The collection of multi-dimensional state parameters reflecting the metabolic state of microorganisms and the fermentation environment during the fermentation process specifically includes: Deploy multiple types of sensors at key detection points in the fermenter; By periodically collecting detection data from various sensors through a data acquisition system, multi-dimensional state parameters reflecting the metabolic state of the microorganisms and the fermentation environment are obtained.

3. The automatic water and culture medium addition control method for the fermentation process as described in claim 1, characterized in that, Constructing a digital twin model of the fermentation process and mapping the collected multi-dimensional state parameters to the digital twin model in real time to achieve virtual-real mapping and visualization of the fermentation state in virtual space specifically includes: A digital twin model of the fermentation process is constructed based on the fermenter equipment parameters, process principles and microbial metabolic characteristics. The collected multi-dimensional state parameters are mapped to the corresponding state variables of the digital twin model in real time; The simulation and visualization engines that drive the digital twin model enable the virtual-real mapping and 3D visualization of the fermentation state.

4. The automatic water and culture medium addition control method for the fermentation process as described in claim 1, characterized in that, Based on metabolic kinetics models and real-time fermentation status, the prediction of cell growth trends and nutrient consumption rates specifically includes: Construct a fermentation metabolic kinetic model that includes cell growth, substrate consumption, and product synthesis; The model parameters and the current state are updated based on the real-time state parameters using a state estimation algorithm. Based on the updated metabolic kinetics model, the cell growth trend and nutrient consumption rate in future periods are predicted.

5. The automatic water and culture medium addition control method for the fermentation process as described in claim 1, characterized in that, A feedforward-feedback composite control strategy is adopted to dynamically determine the timing, feeding rate, and water-to-material ratio of water and culture medium addition based on predicted nutrient requirements and real-time state deviations. Specifically, this includes: Based on the predicted nutrient consumption rate, the feedforward amount required to meet the cell growth needs is calculated. Based on the deviation between real-time status parameters and set target values, calculate the feedback correction feed amount; By combining the feedforward feed rate and the feedback correction feed rate, the timing, feed rate, and water-to-material ratio for adding water and culture medium are dynamically determined.

6. The automatic water and culture medium addition control method for the fermentation process as described in claim 1, characterized in that, Real-time evaluation of fermentation response performance after each feeding, and adaptive optimization of feeding control parameters based on performance evaluation results, specifically including: Monitor changes in key status parameters within the set response period after each material replenishment; Calculate the feeding efficiency evaluation index based on the state changes before and after feeding; Based on the feeding efficiency evaluation index, the metabolic kinetic model parameters and feedforward-feedback control gain are adaptively adjusted.

7. The automatic water and culture medium addition control method for the fermentation process as described in claim 2, characterized in that, The multi-dimensional state parameters include dissolved oxygen concentration, pH value, cell concentration, carbon dioxide release rate, redox potential, fermentation broth temperature, and fermentation broth volume.

8. The automatic water and culture medium addition control method for the fermentation process as described in claim 1, characterized in that, The feeding timing includes the feeding start time in continuous feeding mode and the feeding trigger time in intermittent feeding mode; the feeding rate includes the amount of water and culture medium added per unit time; the water-to-culture medium ratio is the volume ratio or mass ratio of the added water to the culture medium.

9. The automatic water and culture medium addition control method for the fermentation process as described in claim 6, characterized in that, The feeding efficiency evaluation index comprehensively considers the response of three dimensions: cell growth rate, substrate consumption rate, and product synthesis rate.

10. A control system for automatically adding water and culture medium during a fermentation process, characterized in that, include: The data acquisition module is used to collect multi-dimensional state parameters that reflect the metabolic state of the microorganisms and the fermentation environment during the fermentation process; The modeling module is used to construct a digital twin model of the fermentation process, and to map the collected multi-dimensional state parameters to the digital twin model in real time, so as to realize the virtual-real mapping and visualization of the fermentation state in the virtual space. The prediction module is used to predict cell growth trends and nutrient consumption rates based on metabolic kinetics models and real-time fermentation status. The control module is used to dynamically determine the timing, feeding rate, and water-to-material ratio of water and culture medium addition based on the predicted nutrient requirements and real-time state deviation using a feedforward-feedback composite control strategy. The optimization module is used to evaluate the fermentation response efficiency after each feeding in real time and adaptively optimize the feeding control parameters based on the efficiency evaluation results.

Images