Dynamic feedback control method and system in ganoderma lucidum pill fermentation process

Abstract

The invention provides a dynamic feedback control method and system in a ganoderma lucidum pill fermentation process, and the method comprises the steps: in the ganoderma lucidum pill fermentation process, obtaining concentration dynamic change information of ganoderan, ganoderic acid and dissolved oxygen in a fermentation liquid; based on the concentration dynamic change information, generating a space-time atlas reflecting a metabolite distribution state through a multi-modal data interaction fusion mode; inputting the space-time atlas into a pre-trained concentration correlation change prediction model, and outputting concentration correlation change data between ganoderan and ganoderic acid in a future time window; and based on the concentration correlation change data, iteratively correcting the adjustment amount of fermentation control parameters in a time sequence dimension and a space dimension, and generating a feedback control signal to dynamically adjust the fermentation process of the ganoderma lucidum pills, the fermentation control parameters including a material supplementing rate and a ventilation amount. According to the method, the output stability and proportion controllability of the effective components in the fermentation process of the ganoderma lucidum pills are improved.

Description

Technical Field

[0001] The present application relates to the field of biopharmaceutical technology, and in particular to a dynamic feedback control method and system during the fermentation process of Ganoderma lucidum pills. Background Art

[0002] Lingzhi pills are traditional Chinese medicines, and the content and ratio of their core active ingredients, ganoderma polysaccharides and ganoderic acid, directly impact product quality. In industrial fermentation production, the concentrations of these two components must be monitored in real time, and the feed rate and ventilation volume precisely adjusted based on their dynamic relationship to ensure efficient synthesis and stable accumulation of the target metabolites.

[0003] Currently, there is a control method that uses offline detection combined with static model prediction. Specifically, it is to regularly sample and detect the concentration of polysaccharides and acids in the fermentation broth, generate fixed-duration control instructions through a preset concentration-parameter mapping table, and adjust feeding and ventilation operations.

[0004] The above method relies on offline detection, which leads to control lag. At the same time, when the dissolved oxygen concentration changes suddenly, the static model is prone to mismatch between feeding and ventilation regulation, resulting in loss of control of the target component ratio. Summary of the Invention

[0005] The present application provides a dynamic feedback control method and system for the fermentation process of Ganoderma lucidum pills, which is used to solve the problems of poor output stability and weak controllability of the ratio of active ingredients in the fermentation process of Ganoderma lucidum pills in the prior art.

[0006] In a first aspect, the present application provides a method for dynamic feedback control during the fermentation process of Ganoderma lucidum pills, comprising:

[0007] During the fermentation process of Lingzhi pills, the dynamic changes in the concentrations of Ganoderma lucidum polysaccharides, ganoderic acid and dissolved oxygen in the fermentation liquid were obtained;

[0008] Based on the dynamic concentration change information, a spatiotemporal map reflecting the distribution state of metabolites is generated through interactive fusion of multimodal data;

[0009] Inputting the spatiotemporal map into a pre-trained concentration correlation change prediction model to output concentration correlation change data between the Ganoderma lucidum polysaccharide and the Ganoderma lucidic acid in a future time window;

[0010] Based on the concentration-related change data, the adjustment amounts of the fermentation control parameters in the temporal and spatial dimensions are iteratively corrected to generate feedback control signals to dynamically regulate the fermentation process of the Ganoderma lucidum pills. The fermentation control parameters include the feeding rate and the ventilation volume.

[0011] Optionally, the iterative correction of the adjustment amount of the fermentation control parameter in the time dimension and the space dimension based on the concentration correlation change data to generate a feedback control signal includes:

[0012] Based on the concentration correlation change data and the dissolved oxygen concentration dynamic change information, combined with preset fuzzy compensation rules, a control instruction sequence is generated;

[0013] constructing an adaptive matching relationship between a predicted deviation and a compensation threshold parameter based on the control instruction sequence and the concentration-related change data;

[0014] Based on the adaptive matching relationship, a feedback control signal is generated.

[0015] Optionally, generating a feedback control signal based on the adaptive matching relationship includes:

[0016] Determining an adjustment direction of the fermentation control parameter according to a degree of correspondence between the predicted deviation amount and the compensation threshold parameter in the adaptive matching relationship, wherein the adjustment direction includes an incremental change trend in a temporal dimension and a parameter distribution offset pattern in a spatial dimension;

[0017] A feedback control signal is generated based on the incremental change trend in the temporal dimension and the parameter distribution offset pattern in the spatial dimension.

[0018] Optionally, generating a feedback control signal based on the incremental change trend in the temporal dimension and the parameter distribution offset pattern in the spatial dimension includes:

[0019] Calculating a single-step adjustment factor of the control parameter based on the incremental change trend in the time series dimension;

[0020] Applying an upper limit constraint on the adjustment amplitude to the single-step adjustment factor to obtain an amplitude-limited adjustment factor;

[0021] Performing a smoothness threshold check on the amplitude-limited adjustment factor to generate a single-step adjustment factor sequence;

[0022] generating a spatial coefficient matrix according to the parameter distribution offset pattern in the spatial dimension;

[0023] performing amplitude capping on matrix elements of the spatial coefficient matrix;

[0024] Combining the single-step adjustment factor sequence with the pruned spatial coefficient matrix according to a preset spatiotemporal coupling rule;

[0025] The physical feasibility of the fermentation environment is verified for the combination result, and a feedback control signal corresponding to the combination result is output if the verification passes.

[0026] Optionally, the generating of a control instruction sequence based on the concentration correlation change data and the dissolved oxygen concentration dynamic change information in combination with a preset fuzzy compensation rule includes:

[0027] generating a compensation intensity value based on the concentration ratio of ganoderma polysaccharide to ganoderma acid in the concentration correlation change data and the dynamic change information of dissolved oxygen concentration;

[0028] Generate a feeding rate based on the product relationship between the compensation intensity value and the control parameter base value in the preset fuzzy compensation rule;

[0029] Expanding the feed rate into discrete command points in a time series;

[0030] According to the compensation intensity change rate between adjacent discrete instruction points, a transition instruction point is inserted in the time sequence to generate a feeding rate instruction sequence;

[0031] According to the distribution difference of the dynamic change information of the dissolved oxygen concentration in the fermentation container, the spatial priority of the fermentation area is divided, and an independent ventilation correction coefficient is assigned to each fermentation area;

[0032] Adding the ventilation correction coefficient to the ventilation base value of the corresponding fermentation area according to the spatial priority to generate ventilation instruction sequences for different fermentation areas;

[0033] The feed rate instruction sequence is combined with the ventilation volume instruction sequence to generate a control instruction sequence.

[0034] Optionally, the step of inputting the spatiotemporal map into a pre-trained concentration correlation change prediction model to output concentration correlation change data between the Ganoderma lucidum polysaccharide and the Ganoderma lucidic acid in a future time window comprises:

[0035] The spatiotemporal map is input into a pre-trained concentration correlation change prediction model, and the predicted concentration ratio and spatial distribution offset between Ganoderma lucidum polysaccharide and Ganoderma lucidic acid at each future time point are output in sequence through a multi-step recursive prediction algorithm;

[0036] The predicted values ​​of the concentration ratio between the Ganoderma lucidum polysaccharide and ganoderic acid and the spatial distribution offset at all future time points are integrated according to the length of the time window to generate concentration correlation change data between the Ganoderma lucidum polysaccharide and ganoderic acid in the future time window.

[0037] Optionally, generating a spatiotemporal map reflecting the distribution state of metabolites by interactively fusing multimodal data based on the concentration dynamic change information includes:

[0038] Dividing the dynamic change information of the concentrations of the ganoderma polysaccharide, ganoderma acid and dissolved oxygen into time series segments according to the time dimension;

[0039] Each time series segment is labeled with metabolic association features;

[0040] The fermentation container used for Ganoderma lucidum pill fermentation was divided into multiple independent fermentation areas, and the spatial distribution difference ratios of Ganoderma lucidum polysaccharides, ganoderic acid, and dissolved oxygen in each fermentation area were extracted to generate spatial distribution characteristic maps of each fermentation area.

[0041] The metabolic association feature labeling results are cross-bound with the spatial distribution feature map at the corresponding time to generate a spatiotemporal map.

[0042] In a second aspect, the present application provides a dynamic feedback control system for a Ganoderma lucidum pill fermentation process, comprising:

[0043] The acquisition module is used to obtain the dynamic change information of the concentration of ganoderma polysaccharides, ganoderma acid and dissolved oxygen in the fermentation liquid during the fermentation process of ganoderma pills;

[0044] A generation module, configured to generate a spatiotemporal map reflecting the distribution state of metabolites by interactively fusing multimodal data based on the concentration dynamic change information;

[0045] An input module, configured to input the spatiotemporal map into a pre-trained concentration correlation change prediction model, and output concentration correlation change data between the Ganoderma lucidum polysaccharide and the Ganoderma lucidic acid in a future time window;

[0046] The correction module is used to iteratively correct the adjustment amount of the fermentation control parameters in the time dimension and the space dimension based on the concentration correlation change data, and generate a feedback control signal to dynamically adjust the fermentation process of the Ganoderma lucidum pills. The fermentation control parameters include the feeding rate and the ventilation volume.

[0047] In a third aspect, the present application provides a computing device comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute a dynamic feedback control method in a Ganoderma lucidum pill fermentation process as described in any one of the first aspects.

[0048] In a fourth aspect, the present application provides a computer storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the dynamic feedback control method in the fermentation process of Ganoderma lucidum pills as described in any one of the first aspects.

[0049] In the present application, a dynamic feedback control method for the fermentation process of Ganoderma lucidum pills is provided, which comprises: obtaining dynamic concentration change information of Ganoderma lucidum polysaccharides, Ganoderma lucidum acid and dissolved oxygen in the fermentation liquid during the fermentation process of Ganoderma lucidum pills; generating a spatiotemporal map reflecting the distribution state of metabolites through interactive fusion of multimodal data based on the dynamic concentration change information; inputting the spatiotemporal map into a pre-trained concentration correlation change prediction model, and outputting concentration correlation change data between the Ganoderma lucidum polysaccharides and the Ganoderma lucidum acid in a future time window; based on the concentration correlation change data, iteratively correcting the adjustment amount of the fermentation control parameters in the time dimension and the space dimension, and generating a feedback control signal to dynamically adjust the fermentation process of Ganoderma lucidum pills, wherein the fermentation control parameters include the feed rate and the ventilation volume.

[0050] The technical solution provided by this application has the following beneficial effects:

[0051] This application realizes real-time monitoring of key parameters (polysaccharides, acids, dissolved oxygen) of the fermentation process, provides a data basis for dynamic regulation, and improves the output stability of the effective ingredients in the fermentation process of Ganoderma lucidum pills. Through multimodal data fusion, the distribution state of metabolites in time and space dimensions is intuitively presented, breaking through the limitations of single-dimensional analysis. Based on the model, the synergistic change trend of polysaccharides and acids in the future period is predicted to realize feedforward control of the fermentation process. Through iterative correction of parameters in the dual dimensions of time and space, precise coordinated regulation of feed rate and ventilation volume is achieved.

[0052] Furthermore, this application also generates preliminary control instructions by fusing concentration correlation change data with dissolved oxygen dynamic information and combining them with fuzzy compensation rules. This then establishes an adaptive matching mechanism between predicted deviations and compensation thresholds, ultimately outputting dynamically optimized feedback control signals. This process achieves an upgrade from static rule-based control parameters to dynamic adaptation.

[0053] In addition, this solution effectively solves the regulation mismatch problem caused by response lag in traditional control methods. Through the adaptive compensation mechanism, it improves the control accuracy and stability of the proportion of effective ingredients in the Ganoderma lucidum pill fermentation process, and is particularly suitable for complex working conditions such as dissolved oxygen fluctuations.

[0054] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0056] Figure 1 A flow chart of a dynamic feedback control method for a Ganoderma lucidum pill fermentation process provided in an embodiment of the present application;

[0057] Figure 2 A schematic diagram of the structure of a dynamic feedback control system in the fermentation process of Ganoderma lucidum pills provided in an embodiment of the present application;

[0058] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0060] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0061] Researchers have found that the existing Ganoderma lucidum pill fermentation process control method is difficult to capture the dynamic correlation characteristics of polysaccharides and ganoderic acid in real time, and cannot effectively predict the spatiotemporal impact of dissolved oxygen changes on metabolic processes, resulting in hysteresis and spatial heterogeneity in feeding and ventilation regulation. Based on this, the embodiment of the present application provides a dynamic feedback control method for the fermentation process of Ganoderma lucidum pills. The method constructs a spatiotemporal map through multimodal data fusion, combines the prediction model to prospectively output the concentration correlation change trend, and iteratively optimizes the control parameters based on the dual dimensions of time and space to achieve precise coordinated regulation of feeding rate and ventilation volume. The technical solution of the present application can be applied to medicinal fungus fermentation scenarios that require precise control of the proportions and spatial distribution of multiple metabolites.

[0062] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0063] Figure 1 This is a flow chart of a dynamic feedback control method for the fermentation process of Ganoderma lucidum pills provided in an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0064] Step 101: During the fermentation process of the Ganoderma lucidum pills, the concentration dynamic change information of Ganoderma lucidum polysaccharides, Ganoderma lucidum acid and dissolved oxygen in the fermentation liquid is obtained.

[0065] In this step, the dynamic changes in the concentration of Ganoderma lucidum polysaccharides indicate the polysaccharide content in the fermentation broth (unit: g / L), reflecting the metabolic activity of the bacteria. The dynamic changes in the concentration of ganoderic acid indicate the triterpenoid content in the fermentation broth (unit: mg / L), indicating the degree of product accumulation. The dynamic changes in the concentration of dissolved oxygen indicate the oxygen content in the fermentation broth (unit: % saturation), which affects the respiratory metabolism of the bacteria.

[0066] In an embodiment of the present application, fermentation broth parameters are collected in real time by an online sensor array (including a near-infrared spectroscopy probe, a dissolved oxygen electrode, etc.) installed in the fermentation tank, where the polysaccharide and acid concentrations are collected every 5 minutes using spectral analysis, and the dissolved oxygen is continuously monitored by electrodes; the collected data is processed by a signal conversion module and transmitted to a central control system to form a three-parameter time series data set containing a timestamp.

[0067] For example, when cultivating Ganoderma mycelium in a 500-liter stainless steel fermenter, three sets of online monitoring devices were installed: a near-infrared sensor for real-time monitoring of polysaccharide concentration (the detection wavelength was selected at the characteristic absorption peak of Ganoderma polysaccharide at 2100 nm, and the current value was measured to be 15.3 g / L); an HPLC probe for monitoring ganoderic acid concentration (the current value was measured to be 1.2 mg / mL by comparing it with the retention time of a standard); and a dissolved oxygen electrode (using a polarized probe, which measured the dissolved oxygen content in the middle of the tank at 45% saturation). The system automatically collects data every 10 minutes and stores the monitored values ​​of these three parameters in a database along with the acquisition timestamp. For example, at the 36th hour of cultivation, the system recorded a typical set of data: polysaccharide 15.3 g / L, acid 1.2 mg / mL, and dissolved oxygen 45%.

[0068] Step 102: Based on the concentration dynamic change information, a spatiotemporal map reflecting the distribution state of metabolites is generated through multimodal data interactive fusion.

[0069] In this step, multimodal data includes monitoring data from different sources (spectral, electrode, etc.) and types (concentration values, rate of change, etc.). Metabolite distribution refers to the spatiotemporal distribution characteristics of three key substances in the fermentation system: Ganoderma lucidum polysaccharides, ganoderic acid, and dissolved oxygen. The spatiotemporal map represents a metabolite distribution visualization matrix with time and spatial coordinates as the dimensions.

[0070] In the embodiment of the present application, the time series concentration data is bound to the three-dimensional coordinates of the fermentation tank, and a continuous distribution field is generated by the spatiotemporal interpolation algorithm; the tensor decomposition technology is used to fuse the multi-parameter data to construct a time layer (one slice per hour), a spatial layer (1 cm 3 Finally, the dimensions are unified through normalization and a standardized spatiotemporal distribution matrix is ​​output.

[0071] For example, based on data collected over 36 hours, the system first calculates the rate of change of polysaccharide and acid concentrations (by comparing the data from the first six time points, it is found that polysaccharides increase by 0.5 grams per hour and acids increase by 0.08 milligrams per hour). The fermentation tank is then divided into three layers, upper, middle, and lower, with five monitoring points deployed on each layer. A three-dimensional concentration distribution map is generated using a spatial interpolation algorithm. Finally, the time series data and spatial distribution data are integrated to generate a dynamic spatiotemporal map, which shows that the polysaccharide concentration in the lower layer of the tank is high (up to 16.1 grams per liter) but the dissolved oxygen is low (only 38%), while the acid synthesis in the upper layer is active (1.25 mg per milliliter) and the dissolved oxygen is sufficient (52%).

[0072] Step 103: inputting the spatiotemporal map into a pre-trained concentration correlation change prediction model, and outputting the concentration correlation change data between the Ganoderma lucidum polysaccharide and the Ganoderma lucidic acid in a future time window.

[0073] In this step, the prediction model represents a deep learning model based on a hybrid architecture of long short-term memory and transformer. The concentration correlation change represents the trend of the ratio of polysaccharide to acid concentration in the future period.

[0074] In an embodiment of the present application, the spatiotemporal map is split into time series segments and spatial feature maps, which are input into the time encoder and spatial encoder of the model respectively; the cross-regional metabolic associations are captured through the attention mechanism, and the spatiotemporal features are fused in the fully connected layer; the predicted value and confidence interval of the polysaccharide / acid concentration ratio are output every 15 minutes in the next 3 hours to form a change curve.

[0075] For example, after inputting the generated spatiotemporal map into the prediction model, the model analysis revealed that, based on current trends, the accumulation of polysaccharides in the lower layer will slow over the next three hours (predicted to drop from 0.5 g / hour to 0.3 g / hour), while the synthesis of acids in the upper layer will accelerate (from 0.08 mg / hour to 0.12 mg / hour). The model specifically identifies the risk point in the lower right quadrant of the tank (coordinates x3-y4-z1), where insufficient dissolved oxygen (predicted to drop below 35%) will lead to metabolic imbalance, and issues a warning signal.

[0076] Step 104: Based on the concentration-related change data, iteratively correct the adjustment amounts of the fermentation control parameters in the temporal dimension and the spatial dimension to generate a feedback control signal to dynamically adjust the fermentation process of the Ganoderma lucidum pills. The fermentation control parameters include the feeding rate and the ventilation volume.

[0077] In this step, the adjustment amount of the time dimension represents the change gradient of the feeding rate over time (mL / min 2 The adjustment amount of the spatial dimension represents the distribution coefficient of ventilation in different regions (0-1 normalized value). The feedback control signal refers to the set of execution instructions automatically generated by the system, including timing control instructions, spatial control instructions and safety constraints.

[0078] In an embodiment of the present application, the basic feed amount is calculated based on the predicted concentration ratio change using a proportional-integral-differential and fuzzy logic composite control algorithm; the ventilation weight of each area is determined through fluid mechanics simulation in combination with the spatial distribution of dissolved oxygen; the parameters are iterated every 30 minutes, and the adjustment instructions are sent to the distributed actuator (feed pump partition valve group), and the control parameters are synchronously updated through the OPC protocol.

[0079] For example, based on the predicted results, the control system developed an adjustment plan: increasing the feed rate from the current 50 ml / min to 65 ml / min (the increase is calculated by multiplying the predicted deviation by a compensation factor of 0.8). At the same time, the ventilation distribution was adjusted, increasing the lower layer ventilation share from 30% to 40% (calculated based on the regional volume share and the dissolved oxygen gap). After two hours of execution, monitoring data showed that the dissolved oxygen in the lower layer had rebounded to 42%, the polysaccharide accumulation rate had returned to 0.45 g / h, and the acid synthesis rate had stabilized at 0.09 mg / h, indicating that the metabolism of each region had returned to equilibrium.

[0080] By establishing a complete technology chain from real-time monitoring to intelligent control, this solution improves the consistency of target metabolites (polysaccharides and acids) in different areas of the fermentation tank during the Lingzhi pill fermentation process, making the overall distribution more uniform. It effectively maintains the stability of metabolite ratios and reduces quality fluctuations between production batches. At the same time, this solution reduces inefficient consumption of energy and raw materials through precise control.

[0081] In order to solve the problem of insufficient precision in dynamic control of metabolite concentration during the fermentation process of Ganoderma lucidum pills, in some embodiments, step 104: iteratively correcting the adjustment amount of the fermentation control parameter in the temporal and spatial dimensions based on the concentration correlation change data to generate a feedback control signal, includes:

[0082] Step 201: Based on the concentration correlation change data and the dissolved oxygen concentration dynamic change information, combined with preset fuzzy compensation rules, a control instruction sequence is generated.

[0083] In step 201, the fuzzy compensation rule refers to an empirical comparison table of "deviation degree - compensation strength" established based on historical fermentation data. The deviation degree is the percentage by which the polysaccharide-to-acid concentration ratio deviates from the target range, and the compensation strength corresponds to the adjustment level of the feed rate and ventilation volume (categorized as strong, medium, and weak). The control instruction sequence is a sequential instruction set consisting of control parameter adjustment schemes at multiple time nodes. Each node contains the feed rate adjustment value and the ventilation volume allocation ratio for each zone.

[0084] In the embodiment of the present application, the system first analyzes the current concentration correlation change data of Ganoderma lucidum polysaccharides and Ganoderma lucidum acid, combines the spatial distribution characteristics of dissolved oxygen, and calls the preset fuzzy compensation rule library for matching analysis. By comparing the deviation state of the real-time concentration ratio and the target interval, the basic compensation level is determined, and the regional correction coefficient is calculated based on the distribution difference of dissolved oxygen in each area. The basic compensation value and the regional correction coefficient are superimposed and calculated to generate a preliminary control instruction sequence containing multiple control time nodes. This sequence clearly stipulates the change trend of the feeding rate in different time periods and the distribution plan of the ventilation volume in each area.

[0085] Step 202: Based on the control instruction sequence and the concentration-related change data, an adaptive matching relationship between the predicted deviation and the compensation threshold parameter is constructed.

[0086] In step 202, the predicted deviation refers to the degree of difference between the actual concentration change and the model's predicted value. The predicted deviation is calculated by comparing the predicted parameters in the control instruction sequence with the actual detection values ​​of the online sensor array, which are updated in real time. The compensation threshold parameter is the upper limit of the allowable deviation set according to different fermentation stages. Based on the fluctuation characteristics of the concentration-related change data, the compensation threshold parameter is dynamically generated to adapt to the current fermentation stage. The adaptive matching relationship is a feedback mechanism that dynamically adjusts the compensation parameter to ensure that the predicted deviation remains within the threshold range.

[0087] In the examples of this application, during the execution of control instructions, the system continuously monitors the degree of deviation between actual metabolite concentration changes and predicted values. When the deviation exceeds a preset threshold for consecutive cycles, an adaptive adjustment mechanism is activated. By dynamically evaluating the correspondence between the deviation amount and the compensation effect, the matching coefficient of the compensation parameter is recalculated, and the correction amplitude of the feed rate and ventilation volume is adjusted. This process establishes a real-time deviation-compensation mapping relationship, ensuring that the control parameters are always optimally matched to the current fermentation state.

[0088] Step 203: Generate a feedback control signal based on the adaptive matching relationship.

[0089] In the present embodiment, based on the optimized matching relationship between compensation parameters, the system integrates timing control requirements and spatial distribution characteristics to generate a final feedback control signal. This signal converts the feed rate adjustment plan into a specific flow control curve, defining how the flow rate changes over time. It also converts the ventilation volume distribution plan into precise opening instructions for the intake valves in each zone, forming a closed-loop control strategy that coordinates time and space. The control signal is transmitted to the actuator in real time via an industrial bus, completing the entire process from analytical decision-making to physical control.

[0090] Here's a specific example:

[0091] During the cultivation of Ganoderma mycelium in a 500-liter stainless steel fermenter, when the system detected that the polysaccharide concentration reached 18.5 g / L at the 48th hour of fermentation (data was collected every 10 minutes by a near-infrared sensor, and the average of the last six measurements was taken), and the ganoderic acid concentration was 1.4 mg / mL (HPLC detection value), the calculated polysaccharide / acid concentration ratio was 13.2 (18.5 ÷ 1.4), with a deviation of +10% from the target ratio of 12.0. The system combines the dissolved oxygen data of each area of ​​the tank (40% at the bottom, 45% in the middle, and 50% at the top) and queries a preset fuzzy compensation rule base (this rule base is established based on historical data and stipulates that a deviation of 10-15% corresponds to an intermediate compensation). It determines that the basic feeding rate needs to be increased by 15% (the current feeding rate is 60 ml / min, so it is increased by 9 ml to 69 ml / min). At the same time, based on the dissolved oxygen gradient distribution, it is calculated that an additional 5% increase in ventilation volume is required in the bottom area (the compensation coefficient is calculated as: basic compensation value 15% × regional dissolved oxygen deviation coefficient 0.33, where 0.33 is determined by the difference between the bottom dissolved oxygen of 40% and the average value of 45%). During the execution process, the system monitored the actual concentration changes in real time and found that the acid synthesis rate was 0.05 mg / h lower than the predicted value (predicted value 0.08, actual value 0.03). The compensation parameters were recalculated through the deviation feedback algorithm (new feeding rate = original compensation value × (1 + deviation correction coefficient 0.2) = 69 × 1.2 ≈ 83 ml / min), and finally generated a phased control instruction: the feeding rate increased linearly from 60 to 70 ml / min in the first 30 minutes, and increased to 83 ml / min in the next 30 minutes; the ventilation volume distribution was adjusted to 45% at the bottom, 40% in the middle, and 15% at the top, so that the ratio returned to 12.1 after 50 hours of cultivation, and the dissolved oxygen in each area was stabilized at 42±3%.

[0092] In the embodiments of the present application, by establishing a complete regulatory chain from deviation detection to parameter adaptation, the optimal ratio of polysaccharide and acid concentrations is ensured, the dissolved oxygen balance in each area of ​​the fermentation tank is maintained, and the impact of environmental fluctuations on the fermentation process is eliminated through a dynamic compensation mechanism.

[0093] In order to solve the problem of unclear adjustment direction of control parameters during the fermentation process of Ganoderma lucidum pills, in some embodiments, step 203: generating a feedback control signal based on the adaptive matching relationship includes:

[0094] Step 301: Determine the adjustment direction of the fermentation control parameter according to the correspondence between the predicted deviation and the compensation threshold parameter in the adaptive matching relationship, wherein the adjustment direction includes the incremental change trend in the time dimension and the parameter distribution offset pattern in the space dimension.

[0095] In step 301, the incremental change trend in the temporal dimension refers to the direction and magnitude of the feed rate increase or decrease over time, and is categorized into five levels: rapid increase, slow increase, steady state, slow decrease, and rapid decrease. The parameter distribution shift pattern in the spatial dimension refers to the distribution and adjustment scheme of ventilation volume in different areas of the fermenter, including three basic modes: center-aggregation (enhanced central ventilation), edge-compensation (enhanced edge ventilation), and uniform diffusion (proportional adjustment).

[0096] In the present embodiment, the system determines the adjustment direction of the fermentation control parameters by analyzing the relative relationship between the predicted deviation and the compensation threshold parameter. In the temporal dimension, the incremental change trend is divided into different levels according to the degree of deviation, including characteristics such as the rate of change and duration. In the spatial dimension, the parameter distribution deviation pattern is identified based on the abnormal distribution areas of metabolites, and the key locations in each area requiring strengthened or weakened regulation are determined. This step transforms the abstract adaptive matching relationship into a specific regulatory strategy guide by establishing a mapping rule between the degree of deviation and the adjustment direction.

[0097] Step 302: Generate a feedback control signal based on the incremental change trend in the temporal dimension and the parameter distribution offset pattern in the spatial dimension.

[0098] In this embodiment, based on the determined adjustment direction, the system converts the incremental change trend in the temporal dimension into a continuous control curve for the feed rate, setting the slope and duration of the curve according to the trend level. Simultaneously, the distribution offset pattern in the spatial dimension is converted into a regional distribution scheme for ventilation volume, adjusting the ventilation ratio according to the identified key areas. Finally, the temporal control curve and the spatial distribution scheme are fused and calibrated to generate a feedback control signal that combines both time series control instructions and spatial distribution parameters, enabling precise regulation of the fermentation process.

[0099] Here's a specific example:

[0100] During the cultivation of Ganoderma mycelium in a 500-liter stainless steel fermenter, the system detected a polysaccharide / acid concentration ratio deviation of +12% (current ratio 13.4, target ratio 12.0) at the 50th hour of fermentation. Based on the ratio of the deviation to the compensation threshold parameter (1.2, 12% ÷ 10%), the system determined the need for a "rapid rise" temporal adjustment trend and a "centrally concentrated" spatial offset mode. Using a preset adjustment intensity calculation formula (base adjustment amount × deviation coefficient 1.2), the system determined the feed rate adjustment strategy: a 0.6 ml / min increase (base slope 0.5 ml × 1.2) in the first phase (0-20 minutes), followed by a 0.4 ml / min increase in the second phase (20-40 minutes). At the same time, based on the dissolved oxygen distribution (42% in the central area, 45% in the transition area, and 48% in the edge area), and in accordance with the calculation rules of the central aggregation model (central area compensation amount = basic compensation amount × (1 + central area dissolved oxygen deviation coefficient 0.8)), the ventilation volume distribution was adjusted as follows: the ventilation proportion in the central area was increased to 46% (original 40% + 6%), the transition area maintained at 40%, and the edge area was reduced to 14%. After 2 hours of implementing this control plan, the dissolved oxygen in the central area increased to 45%, and the polysaccharide / acid ratio deviation was reduced to +6%.

[0101] In the embodiments of the present application, by establishing clear adjustment direction determination rules and control signal generation mechanisms, the accuracy and consistency of control parameter adjustment are ensured, the response speed to sudden fermentation anomalies is improved, and the uniformity of material distribution in the fermentation tank is effectively improved through spatiotemporal coordinated regulation.

[0102] To further improve the accuracy and reliability of the Ganoderma lucidum pill fermentation control signal, in some embodiments, step 302: generating a feedback control signal based on the incremental change trend in the temporal dimension and the parameter distribution offset pattern in the spatial dimension, includes:

[0103] Step 401: Calculate a single-step adjustment factor of the control parameter based on the incremental change trend in the time series dimension.

[0104] In step 401, the single-step adjustment factor refers to the change in the control parameter within a single time unit and is used to quantify the adjustment amplitude of each control step in the time series dimension. Its numerical value is determined by the type and strength of the incremental change trend. Exemplary calculation process: Extract the concentration change data from the current moment to the previous N time steps, calculate the difference in the rate of change of the concentration of Ganoderma lucidum polysaccharide and Ganoderma lucidic acid (for example, the growth rate of Ganoderma lucidic acid is 0.2% / min faster than that of polysaccharide); according to the preset trend-adjustment mapping table, convert the rate of change difference into the initial adjustment factor base value (for example, a difference of 0.2% / min corresponds to a base value + 0.5). The initial base value is weighted based on the direction of change in dissolved oxygen concentration (increase / decrease): if the dissolved oxygen decreases, the weight of the Ganoderma lucidic acid growth rate increases, and the adjustment factor base value is multiplied by 1.2; otherwise, it is multiplied by 0.8; and output the unconstrained single-step adjustment factor (for example, final value = 0.5 × 1.2 = 0.6).

[0105] In this embodiment, the system uses a trend-to-factor conversion model to calculate a basic single-step adjustment factor based on the identified "rapidly rising" trend type. This model takes into account factors such as the current fermentation stage and historical control effects, and outputs a theoretical adjustment to the feed rate within each control period (e.g., 15 minutes).

[0106] Step 402: applying an upper limit constraint on the adjustment amplitude to the single-step adjustment factor to obtain an amplitude-limited adjustment factor.

[0107] In step 402, the upper limit constraint of the adjustment amplitude refers to the maximum allowable single-step adjustment amount set to ensure fermentation safety, including two types of constraints: the absolute value upper limit and the relative change rate upper limit. Exemplarily, the implementation process is as follows: read the upper limit of the adjustment amplitude of the current fermentation stage from the predefined process parameter library (such as the maximum allowable single-step adjustment of the feed rate of ±10L / h); convert the upper limit value into an equivalent constraint range of the adjustment factor (for example, ±0.8). If the single-step adjustment factor is greater than the upper limit value (such as 0.6>0.8), the original value is retained; if the single-step adjustment factor is less than the negative upper limit value (such as -1.0<-0.8), it is forced to be set to -0.8; the output amplitude-limited adjustment factor (for example, 0.6 is 0.6, -1.0 is -0.8). The amplitude-limited adjustment factor refers to the single-step adjustment factor after the adjustment amplitude upper limit constraint is processed, which is used to ensure that the change in the control parameter (such as the feed rate adjustment amount) within each time unit does not exceed the maximum safety range allowed by the equipment or process.

[0108] In this embodiment, the system retrieves the current stage's safety parameter range from the fermentation process database and performs a double check on the calculated single-step adjustment factor: first, the absolute value is compared to see if it exceeds the equipment's allowable limit, and then the increase relative to the current value is calculated to see if it exceeds the limit. The factor that passes the check becomes the amplitude-limited adjustment factor.

[0109] Step 403: Perform a smoothness threshold check on the amplitude-limited adjustment factor to generate a single-step adjustment factor sequence.

[0110] In step 403, the smoothness threshold check is a filtering condition used to ensure that the adjustment amount of adjacent time units changes smoothly, avoiding drastic fluctuations in the control parameters. The single-step adjustment factor sequence is a continuous control instruction set composed of the amplitude-limited adjustment factors of multiple time units in chronological order.

[0111] In this embodiment, the system establishes a sliding time window (e.g., three control cycles) and calculates the gradient of the adjustment factor within the window. When a new factor causes the gradient to exceed a threshold, a smooth correction is performed based on the average rate of change over the previous two cycles, ultimately outputting a single-step adjustment factor sequence that meets the continuity requirement.

[0112] Step 404: Generate a spatial coefficient matrix according to the parameter distribution offset pattern in the spatial dimension.

[0113] In step 404, the spatial coefficient matrix is ​​a two-dimensional array that quantifies the control intensity of each area of ​​the fermentation tank. The rows and columns of the matrix correspond to spatial positions, and the element values ​​represent the parameter adjustment weights of the area.

[0114] In an embodiment of the present application, according to the "center-aggregated" distribution offset mode, the system generates an initial coefficient matrix: the central area is assigned the highest coefficient, which decreases layer by layer toward the periphery. The coefficient value is determined by the regional importance weight and the current degree of dissolved oxygen deviation. The specific process is: the fermentation container is divided into K independent areas (such as three layers above and below × three areas on the left and right), and the concentration gradient difference between ganoderma lucidum acid and polysaccharide in each area is calculated (such as the gradient difference in area A = 0.3% / cm); the offset direction is determined according to the sign of the gradient difference (positive / negative) (such as a positive value means aggregation towards the center of the area). Main diagonal elements: represent the self-adjustment intensity of each area, and the value is the absolute value of the gradient difference in the area (such as the element corresponding to area A = 0.3); non-diagonal elements: represent the synergistic influence between regions. If the offset directions of the two areas are the same, the value is the average of the gradient differences of the two areas (such as areas A and B are in the same direction, element = 0.3 + 0.2 / 2 = 0.25); output the uncropped spatial coefficient matrix (K × K dimensions). Exemplary:

[0115] Example matrix (3-area simplified version)

[0116]

[0117] Step 405: Perform amplitude upper limit clipping on the matrix elements of the spatial coefficient matrix.

[0118] In step 405, amplitude upper limit clipping is a normalization process for the elements of the spatial coefficient matrix to prevent excessive adjustment in local areas.

[0119] In this embodiment, the system sets a maximum adjustment coefficient limit for each region, scaling the elements in the matrix that exceed the limit proportionally, while maintaining the relative proportional relationship between the regions. For example, amplitude clipping means that the absolute value upper limit is applied to all elements (such as a single element maximum of 0.5), and the upper limit value of the exceeding elements is retained by sign (such as 0.6 corresponds to 0.5, and -0.7 corresponds to -0.5).

[0120] Step 406: Combine the single-step adjustment factor sequence and the pruned spatial coefficient matrix according to a preset spatiotemporal coupling rule.

[0121] In step 406, the spatiotemporal coupling rule is an operation method that integrates the timing adjustment sequence with the spatial coefficient matrix to ensure that the temporal and spatial control are coordinated and consistent.

[0122] In this embodiment of the present application, the system performs a tensor product operation on the single-step adjustment factor sequence and the spatial coefficient matrix to generate a spatiotemporal joint control tensor. The time dimension of this tensor records the adjustment amount for each time period, and the spatial dimension records the allocation ratio for each region.

[0123] Step 407: Performing physical feasibility verification of the fermentation environment on the combination result, and outputting a feedback control signal corresponding to the combination result if the verification passes.

[0124] In step 407, physical feasibility verification includes three levels: equipment execution capability verification, process compliance check, and safety assessment.

[0125] In this embodiment, the system integrates a tensor input verification module to sequentially check whether the actuator can achieve the required rate of change, whether the adjustment plan meets process specifications, and whether key parameters exceed safety thresholds. The verified tensor is then converted into the final control signal.

[0126] Here's a specific example:

[0127] During the cultivation of Ganoderma mycelium in a 500-liter stainless steel fermenter, the system detected a +15% deviation in the polysaccharide / acid concentration ratio (current ratio 13.8, target ratio 12.0) at the 52nd hour of fermentation. Based on the ratio of the deviation to the compensation threshold parameter (1.5, 15% ÷ 10%), the system determined that a "rapidly rising" temporal adjustment trend and a "centrally concentrated" spatial offset mode were required. The system first calculated the single-step adjustment factor: a base adjustment of 0.5 ml / min (a benchmark determined by historical data) was multiplied by the deviation factor of 1.5 to obtain an initial single-step adjustment factor of 0.75 ml / min. After applying the upper limit (the equipment allows a maximum single-step adjustment of 0.7 ml / min), the adjustment factor was corrected to 0.7 ml / min. After smoothness verification (adjustment volume variation between adjacent time periods does not exceed 0.2 ml), a single-step adjustment factor sequence was generated: +0.6 ml / min for 0-15 minutes, +0.7 ml / min for 15-30 minutes, and +0.7 ml / min for 30-45 minutes. Simultaneously, based on the dissolved oxygen distribution (41% in the central region, 44% in the transition region, and 47% in the peripheral region), a spatial coefficient matrix was generated according to the central aggregation model: a coefficient of 0.5 for the central region (calculated as: base coefficient 0.4 × (1 + dissolved oxygen deviation coefficient 0.25)), 0.3 for the transition region, and 0.2 for the peripheral region. After amplitude clipping (single-region coefficient upper limit 0.45), the matrix was corrected to 0.45 for the central region, 0.3 for the transition region, and 0.2 for the peripheral region. After spatiotemporal coupling, the actual increase in feed volume in the central region during the first 15 minutes was 0.6 × 0.45 = 0.27 ml / min. It was verified that the scheme met all safety constraints, and 2 hours after the final control signal was executed, the dissolved oxygen in the central area increased from 41% to 44%, and the deviation of the polysaccharide / acid ratio decreased to +8%.

[0128] In the embodiment of the present application, by establishing a complete spatiotemporal parameter processing and verification mechanism, the stability and safety of the control signal are ensured, the accuracy of multi-region coordinated regulation is improved, and the risk of misoperation is effectively prevented through physical feasibility verification.

[0129] In order to further improve the accuracy and adaptability of control instruction generation during the fermentation process of Ganoderma lucidum pills, in some embodiments, step 201: generating a control instruction sequence based on the concentration correlation change data and the dissolved oxygen concentration dynamic change information in combination with a preset fuzzy compensation rule includes:

[0130] Step 501: generating a compensation intensity value based on the concentration ratio of ganoderma polysaccharide to ganoderma acid in the concentration correlation change data and the dynamic change information of dissolved oxygen concentration.

[0131] In step 501, the compensation intensity value is a quantitative indicator of the control intensity calculated based on the current fermentation state. Its value ranges from 0 to 1, with 0 indicating no compensation and 1 indicating maximum compensation intensity. This value is determined by the degree of deviation of the polysaccharide / acid concentration ratio and the trend of dissolved oxygen variation.

[0132] In this embodiment, the system first calculates the percentage deviation between the current polysaccharide / acid ratio and the target value. Then, combined with the instantaneous value of dissolved oxygen and its slope of change, it uses a three-dimensional fuzzy rule table (deviation degree × dissolved oxygen value × dissolved oxygen change trend) to determine the compensation intensity value. For example, if the deviation is +15% and the dissolved oxygen is 45% and declining, the output compensation intensity value is 0.8.

[0133] Step 502: Generate a feeding rate based on the product relationship between the compensation intensity value and the control parameter base value in the preset fuzzy compensation rule.

[0134] In step 502, the control parameter base value refers to the standard control amount preset for each fermentation stage, including the basic feeding rate and the basic ventilation volume, which are pre-set according to the process requirements. The feeding rate refers to the flow rate of adding fresh culture medium or nutrients to the fermentation tank per unit time (such as per minute), and its numerical value directly affects the growth rate of the bacteria and the accumulation of metabolites. For example, according to the standard feeding amount preset for the current fermentation stage (such as 50 ml / min); the coefficient reflecting the current required control intensity (such as 0.8); the actual feeding rate = base value × (1 ± compensation intensity value), the feeding amount is increased during positive compensation (such as 50×1.8=90 ml / min), and the feeding amount is reduced during negative compensation (such as 50×0.2=10 ml / min).

[0135] In this embodiment of the present application, the system retrieves the base feed rate for the current stage (e.g., 50 ml / min) from the process database and multiplies it by the compensation intensity value to obtain the actual feed rate (50×0.8=40 ml / min). This calculation ensures that the control amount accurately matches the fermentation status.

[0136] Step 503: Expand the feeding rate into discrete instruction points in time series.

[0137] In step 503, discrete instruction points refer to key control nodes of control parameters on the time axis, and each node includes a specific timestamp and parameter value.

[0138] In this embodiment, the system divides the total control time (e.g., 2 hours) into several time periods (e.g., 30-minute periods) and sets a feeding rate value at each time period node, forming a preliminary sequence of discrete control points. The feeding rate value at each point is proportionally allocated based on the time period length and the total adjustment amount.

[0139] Step 504: inserting transition instruction points in time sequence according to the compensation intensity change rate between adjacent discrete instruction points to generate a feed rate instruction sequence.

[0140] In step 504 , the transition instruction point is an intermediate control point inserted between discrete instruction points to ensure control continuity and is used to smooth the parameter change curve.

[0141] In this embodiment, the system calculates the feed rate gradient between adjacent discrete points. When the gradient exceeds a set threshold (e.g., 0.5 ml / min), one or two transition points are inserted between the two points to achieve a linear transition. The resulting feed rate instruction sequence includes all discrete points and necessary transition points.

[0142] Step 505: According to the distribution difference of the dissolved oxygen concentration dynamic change information in the fermentation container, the spatial priority of the fermentation area is divided, and an independent ventilation correction coefficient is allocated to each fermentation area.

[0143] In step 505, the spatial priority refers to the importance ranking of the fermentation areas according to the degree of abnormality of the dissolved oxygen distribution. The higher the priority, the more necessary the area is to be adjusted.

[0144] In this embodiment, the system divides the fermentation tank vertically into three layers: upper, middle, and lower. The system calculates the deviation of dissolved oxygen from the target value for each layer and prioritizes them in descending order (lower layer > middle layer > upper layer). Aeration correction coefficients are assigned based on priority, with higher priority levels receiving larger coefficients.

[0145] Step 506: superimposing the ventilation correction coefficient onto the ventilation base value of the corresponding fermentation area according to the spatial priority to generate ventilation instruction sequences for different fermentation areas.

[0146] In step 506, the ventilation instruction sequence refers to the ventilation adjustment plan for each region at different time points, including time dimension and space dimension information.

[0147] In this embodiment of the present application, the system multiplies the base ventilation volume (e.g., 30 cubic meters / hour) by the correction coefficient for each zone to obtain the actual ventilation volume for the zone (e.g., a coefficient of 1.2 corresponds to 36 cubic meters / hour). The system generates ventilation volume change curves for each zone at different time periods according to the timing requirements, forming a complete zone instruction sequence.

[0148] Step 507: Combine the feeding rate instruction sequence with the ventilation volume instruction sequence to generate a control instruction sequence.

[0149] In an embodiment of the present application, the system aligns and merges the feeding rate instruction sequence and the ventilation volume instruction sequence of each area along the time axis. Each time point contains the feeding rate value and the ventilation volume value of each area, forming a control instruction table that can be directly issued to the actuator.

[0150] Here's a specific example:

[0151] During the cultivation of Ganoderma mycelium in a 500-liter stainless steel fermenter, the system detected a polysaccharide / acid concentration ratio deviation of +15% (current ratio 13.8, target ratio 12.0) at the 52nd hour of fermentation. Based on the deviation and the dynamic changes in dissolved oxygen (42% in the center region, 44% in the transition region, and 47% in the edge region), a three-dimensional fuzzy compensation rule table was used to determine a compensation intensity value of 0.9 (calculation process: a 15% deviation corresponds to an intensity base value of 0.8, with a correction factor of +0.1 added to account for the downward trend in dissolved oxygen). Based on a base feed rate of 60 ml / min, the actual feed rate was calculated to be 60 × (1 + 0.9) = 114 ml / min. This feed rate was then expanded into discrete command points in a time series: 60 ml at 0 minutes and 114 ml at 60 minutes. Because the rate of change in adjacent point compensation intensity (0.9 / 60 = 0.015) exceeded the threshold of 0.01, the system inserted a transition command point of 87 ml at 30 minutes (calculated as: 60 + (114 - 60) × 0.5), resulting in a feed rate command sequence of [0:60, 30:87, 60:114]. Furthermore, based on the differences in dissolved oxygen distribution (the lowest in the central zone was 42%), spatial priorities were prioritized as central zone > transition zone > edge zone. Ventilation correction coefficients of 1.3, 1.1, and 0.8 were assigned based on priority (calculated as: base coefficient 1.0 + priority weight × 0.3). Based on a base ventilation rate of 30 cubic meters per hour, the resulting regional ventilation command sequence was: 39 (30 × 1.3) in the central zone, 33 (30 × 1.1) in the transition zone, and 24 (30 × 0.8) in the edge zone.

[0152] In the embodiments of the present application, by establishing a full-process method from parameter analysis to instruction generation, the dynamic matching of control instructions and fermentation status is ensured, the accuracy of multi-parameter coordinated regulation is improved, and the stability of the control process is guaranteed through the transition instruction point mechanism.

[0153] In order to further improve the prediction accuracy of metabolite concentration changes during the fermentation process of Ganoderma lucidum pills, in some embodiments, step 103: inputting the spatiotemporal map into a pre-trained concentration correlation change prediction model to output the concentration correlation change data between the Ganoderma lucidum polysaccharide and the Ganoderma lucidic acid in the future time window includes:

[0154] Step 601: Input the spatiotemporal map into a pre-trained concentration correlation change prediction model, and output the predicted concentration ratio and spatial distribution offset between Ganoderma lucidum polysaccharide and Ganoderma lucidic acid at each future time point in sequence through a multi-step recursive prediction algorithm.

[0155] In step 601, the multi-step recursive prediction algorithm is a method that iteratively predicts data at multiple time points, with the output of each prediction step serving as the input for the next step. The predicted concentration ratio represents the concentration ratio of Ganoderma lucidum polysaccharides to Ganoderma lucidic acid at a specific point in the future. The spatial distribution offset quantifies the degree of change in the distribution of metabolites across different fermentation zones.

[0156] In the present application embodiment, the prediction model first decomposes the spatiotemporal map into time series data and spatial feature graphs, extracts the concentration change trend characteristics of polysaccharides and acids through a time encoder, and analyzes the dissolved oxygen distribution pattern in each region through a space encoder. In the prediction stage, the model takes the spatiotemporal characteristics of the current moment as a starting point, and gradually predicts the concentration ratio and regional offset of subsequent time points. Each prediction step is 15 minutes, and 12 steps (3 hours) are continuously predicted. Each prediction step is dynamically corrected with reference to the result of the previous step.

[0157] Step 602: Integrate the predicted values ​​of the concentration ratio between the Ganoderma lucidum polysaccharide and ganoderic acid and the spatial distribution offset at all future time points according to the length of the time window to generate concentration correlation change data between the Ganoderma lucidum polysaccharide and ganoderic acid in the future time window.

[0158] In step 602, the time window length refers to the total time length covered by the prediction, and the concentration correlation change data is a structured data set that integrates the concentration ratio and spatial offset information of all time points within the prediction period.

[0159] In this example, the system arranges the multi-step prediction results in chronological order, uses cubic spline interpolation to generate a continuous curve for the concentration ratio data, and performs cluster analysis on the spatial offset by region, marking the offset hotspots. The final output is a data combination consisting of a time-concentration ratio curve and a spatial offset heat map.

[0160] Here's a specific example:

[0161] During the 72nd hour of cultivating Ganoderma mycelium in a 500-liter stainless steel fermenter, the system was fed into the prediction model using a temporal map. The map showed a polysaccharide concentration of 19.2 g / L (average of six measurements), ganoderic acid of 1.5 mg / mL, and a spatial distribution of dissolved oxygen (DO) with the concentrations at the bottom (42%), the middle (45%), and the top (48%). The model first used a time encoder to analyze the concentrations, revealing an increase in polysaccharide of 0.35 g per hour (linear regression slope), an increase in acid of 0.07 mg per hour (cubic spline fit), and a decrease in dissolved oxygen of 0.8% per hour (bottom region). In a multi-step recursive prediction, the model uses a 15-minute step length, dividing one hour into four steps. The first step predicts an increase in polysaccharide by 0.0875 g (0.35 ÷ 4), an increase in acid by 0.0175 mg (0.07 ÷ 4), and a decrease in bottom dissolved oxygen by 0.2% (0.8% ÷ 4). The calculated concentration ratio changes from 12.8 (19.2 ÷ 1.5) to 12.9 ((19.2 + 0.0875) ÷ (1.5 + 0.0175)), with a bottom zone offset of +0.5% (current 42% minus predicted 41.8%). After 12 consecutive prediction steps (three hours), the integrated data shows that after three hours, the concentration ratio will reach 13.4 (calculated by: 12.8 + (0.1 × 6)), the bottom zone dissolved oxygen will drop to 40.4%, and the central zone offset will accumulate by +1.2%.

[0162] In the embodiments of the present application, through multi-step recursive prediction and spatiotemporal data integration, the effective warning time window is extended, the accuracy of regionalized offset prediction is improved, and sufficient data support is provided for active regulation.

[0163] In order to further improve the spatiotemporal resolution of metabolite monitoring during the fermentation process of Lingzhi pills, in some embodiments, step 102: generating a spatiotemporal map reflecting the distribution state of metabolites by interactively fusing multimodal data based on the concentration dynamic change information, includes:

[0164] Step 701: Divide the dynamic change information of the concentrations of Ganoderma lucidum polysaccharides, Ganoderma lucidic acid and dissolved oxygen into time series segments according to the time dimension.

[0165] In step 701, the time series segments refer to time period units in which the continuous monitoring data are divided according to a fixed time length, and each segment contains the concentration change curves and interaction relationship data of Ganoderma lucidum polysaccharides, Ganoderma lucidic acid and dissolved oxygen in the time period.

[0166] In the embodiment of the present application, the system cuts the original monitoring data into 30-minute basic units and calculates three key indicators for each segment: the slope of the polysaccharide concentration change (linear regression coefficient), the acceleration of the acid concentration change (second-order derivative), and the response delay time of dissolved oxygen to the changes of the two (cross-correlation analysis).

[0167] Step 702: Perform metabolic association feature tagging on each time series segment.

[0168] In step 702, metabolic association feature labels refer to feature sets that quantify the interaction relationships between different metabolites, including association patterns such as promotion, inhibition, and synergy.

[0169] In this example, the system uses Granger causality tests to analyze the lead-lag relationship between polysaccharide and acid changes, and uses correlation coefficients to quantify the strength of the effect of dissolved oxygen on both. The results are encoded as three-color markers: red indicates strong inhibition (correlation coefficient <-0.7), green indicates strong promotion (>0.7), and yellow indicates weak correlation (-0.3 to 0.3).

[0170] Step 703: Divide the fermentation container used for the fermentation of the Ganoderma lucidum pills into multiple independent fermentation areas, extract the spatial distribution difference ratios of Ganoderma lucidum polysaccharides, Ganoderma lucidum acid and dissolved oxygen in each fermentation area, and generate a spatial distribution feature map of each fermentation area.

[0171] In step 703, the spatial distribution characteristic map is a quantitative map reflecting the uneven distribution of metabolites in the fermentation container, including regional concentration gradients and boundary diffusion trends.

[0172] In the present embodiment, the fermentation tank was vertically divided into three layers: upper, middle, and lower. Each layer was annularly divided into three zones: center, middle, and edge, for a total of 9 zones. For each zone, the following were calculated: the regional proportion of polysaccharide concentration (concentration in the zone / average concentration), the coefficient of variation of acid concentration (standard deviation / mean), and the dissolved oxygen gradient (difference in concentration from adjacent zones). A continuous distribution surface was generated by Kriging interpolation.

[0173] Step 704: Cross-binding the metabolic association feature labeling result with the spatial distribution feature map at the corresponding time to generate a spatiotemporal map.

[0174] In step 704, cross-binding refers to a fusion process of establishing a mapping relationship between temporal features and spatial features to form a data entity with dual temporal and spatial attributes.

[0175] In an embodiment of the present application, the system matches the spatial feature map of the corresponding moment for each time segment, and fuses the two through a spatiotemporal encoder: the temporal features are converted into position codes, and the spatial features are converted into channel attention weights, and the final output dimension is a three-dimensional spatiotemporal map with the dimension of [time step × spatial partition × feature channel].

[0176] Here's a specific example:

[0177] During the 36th hour of Ganoderma mycelium cultivation in a 500-liter stainless steel fermenter, the continuous monitoring data was first segmented into 30-minute time series segments. For example, the segment from 36 to 36.5 hours showed an increase in polysaccharide concentration from 16.2 g / L to 16.5 g / L (slope calculated as (16.5 - 16.2) × 2 = +0.6 g / L / hour), while the acid concentration measured by HPLC increased from 1.1 mg / mL to 1.12 mg / mL (acceleration calculated as [(1.12 - 1.1) × 2 - previous period acceleration 0.015] × 2 = +0.01 mg / mL / hour²). Dissolved oxygen decreased from 46% to 44% (the delayed response time was determined to be 15 minutes through cross-correlation analysis). Based on the correlation coefficient of -0.65 between the decrease in dissolved oxygen and the increase in acidity in this segment, the data was labeled "Moderate inhibition of acid synthesis by dissolved oxygen." The fermenter was divided into three independent zones: the upper zone (z1), the middle zone (z2), and the lower zone (z3). The results showed that the polysaccharide region in the lower zone accounted for 118% (16.5 ÷ average concentration 14.0), the acid coefficient of variation was 0.18 (standard deviation 0.2 ÷ mean 1.12), and the dissolved oxygen gradient was -4% / cm (lower zone 44% - middle zone 48%). Finally, the temporal segment features were linked to the spatial distribution data to generate a spatiotemporal map showing that at 36.5 hours, the lower zone exhibited a metabolic state characterized by "high polysaccharide (118%), low oxygen (44%), and acid inhibition," providing precise spatiotemporal positioning for subsequent regulation.

[0178] In the examples of the present application, through data fusion and feature labeling in both time and space dimensions, local metabolic abnormality areas that are difficult to detect with traditional methods are revealed, a visual expression of the interaction relationship between metabolites is established, and a spatial positioning basis is provided for precise regulation.

[0179] Figure 2 This is a schematic diagram of the structure of a dynamic feedback control system in the fermentation process of Ganoderma lucidum pills provided in an embodiment of the present application, such as Figure 2 As shown, the system includes:

[0180] The acquisition module 21 is used to obtain the dynamic change information of the concentrations of ganoderma polysaccharides, ganoderma acid and dissolved oxygen in the fermentation liquid during the fermentation process of the ganoderma pills.

[0181] The generation module 22 is used to generate a spatiotemporal map reflecting the distribution state of metabolites through multimodal data interactive fusion based on the concentration dynamic change information.

[0182] The input module 23 is used to input the spatiotemporal map into a pre-trained concentration correlation change prediction model, and output the concentration correlation change data between the Ganoderma lucidum polysaccharide and the Ganoderma lucidic acid in a future time window.

[0183] The correction module 24 is used to iteratively correct the adjustment amount of the fermentation control parameters in the time dimension and the space dimension based on the concentration correlation change data, and generate a feedback control signal to dynamically adjust the fermentation process of the Ganoderma lucidum pills. The fermentation control parameters include the feeding rate and the ventilation volume.

[0184] Figure 2 The dynamic feedback control system in the fermentation process of the Ganoderma lucidum pills can be implemented Figure 1 The implementation principle and technical effects of the dynamic feedback control method for the fermentation process of Ganoderma lucidum pills described in the illustrated embodiment will not be elaborated on here. The specific manner in which each module and unit performs operations in the dynamic feedback control system for the fermentation process of Ganoderma lucidum pills in the above embodiment has been described in detail in the relevant embodiments of the method and will not be elaborated on here.

[0185] In one possible design, Figure 2 The dynamic feedback control system of the Ganoderma lucidum pill fermentation process of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0186] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0187] The processing component 32 is as follows Figure 1 The embodiment provides a dynamic feedback control method for the fermentation process of Ganoderma lucidum pills.

[0188] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0189] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random-access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0190] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0191] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0192] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0193] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0194] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The embodiment shown is a dynamic feedback control method for the fermentation process of Ganoderma lucidum pills.

[0195] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0196] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0197] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0198] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A dynamic feedback control method for the fermentation process of Ganoderma lucidum pills, characterized in that: include: During the fermentation process of Lingzhi pills, the dynamic changes in the concentrations of Ganoderma lucidum polysaccharides, ganoderic acid and dissolved oxygen in the fermentation liquid were obtained; Based on the dynamic concentration change information, a spatiotemporal map reflecting the distribution state of metabolites is generated through interactive fusion of multimodal data; Inputting the spatiotemporal map into a pre-trained concentration correlation change prediction model to output concentration correlation change data between the Ganoderma lucidum polysaccharide and the Ganoderma lucidic acid in a future time window; Based on the concentration-related change data, the adjustment amounts of the fermentation control parameters in the temporal and spatial dimensions are iteratively corrected to generate feedback control signals to dynamically regulate the fermentation process of the Ganoderma lucidum pills. The fermentation control parameters include the feeding rate and the ventilation volume.

2. The method according to claim 1, characterized in that The iterative correction of the adjustment amount of the fermentation control parameter in the time dimension and the space dimension based on the concentration correlation change data to generate a feedback control signal includes: Based on the concentration correlation change data and the dissolved oxygen concentration dynamic change information, combined with preset fuzzy compensation rules, a control instruction sequence is generated; constructing an adaptive matching relationship between a predicted deviation and a compensation threshold parameter based on the control instruction sequence and the concentration-related change data; Based on the adaptive matching relationship, a feedback control signal is generated.

3. The method according to claim 2, characterized in that Generating a feedback control signal based on the adaptive matching relationship includes: Determining an adjustment direction of the fermentation control parameter according to a degree of correspondence between the predicted deviation amount and the compensation threshold parameter in the adaptive matching relationship, wherein the adjustment direction includes an incremental change trend in a temporal dimension and a parameter distribution offset pattern in a spatial dimension; A feedback control signal is generated based on the incremental change trend in the temporal dimension and the parameter distribution offset pattern in the spatial dimension.

4. The method according to claim 3, characterized in that The generating of the feedback control signal based on the incremental change trend in the time dimension and the parameter distribution offset pattern in the space dimension includes: Calculating a single-step adjustment factor of the control parameter based on the incremental change trend in the time series dimension; Applying an upper limit constraint on the adjustment amplitude to the single-step adjustment factor to obtain an amplitude-limited adjustment factor; Performing a smoothness threshold check on the amplitude-limited adjustment factor to generate a single-step adjustment factor sequence; generating a spatial coefficient matrix according to the parameter distribution offset pattern in the spatial dimension; performing amplitude capping on matrix elements of the spatial coefficient matrix; Combining the single-step adjustment factor sequence with the pruned spatial coefficient matrix according to a preset spatiotemporal coupling rule; The physical feasibility of the fermentation environment is verified for the combination result, and a feedback control signal corresponding to the combination result is output if the verification passes.

5. The method according to claim 2, characterized in that The generating of a control instruction sequence based on the concentration correlation change data and the dissolved oxygen concentration dynamic change information in combination with a preset fuzzy compensation rule includes: generating a compensation intensity value based on the concentration ratio of ganoderma polysaccharide to ganoderma acid in the concentration correlation change data and the dynamic change information of dissolved oxygen concentration; Generate a feeding rate based on the product relationship between the compensation intensity value and the control parameter base value in the preset fuzzy compensation rule; Expanding the feed rate into discrete command points in a time series; According to the compensation intensity change rate between adjacent discrete instruction points, a transition instruction point is inserted in the time sequence to generate a feeding rate instruction sequence; According to the distribution difference of the dynamic change information of the dissolved oxygen concentration in the fermentation container, the spatial priority of the fermentation area is divided, and an independent ventilation correction coefficient is assigned to each fermentation area; Adding the ventilation correction coefficient to the ventilation base value of the corresponding fermentation area according to the spatial priority to generate ventilation instruction sequences for different fermentation areas; The feed rate instruction sequence is combined with the ventilation volume instruction sequence to generate a control instruction sequence.

6. The method according to claim 1, wherein The step of inputting the spatiotemporal map into a pre-trained concentration correlation change prediction model and outputting concentration correlation change data between the Ganoderma lucidum polysaccharide and the Ganoderma lucidic acid in a future time window comprises: The spatiotemporal map is input into a pre-trained concentration correlation change prediction model, and the predicted concentration ratio and spatial distribution offset between Ganoderma lucidum polysaccharide and Ganoderma lucidic acid at each future time point are output in sequence through a multi-step recursive prediction algorithm; The predicted values ​​of the concentration ratio between the Ganoderma lucidum polysaccharide and ganoderic acid and the spatial distribution offset at all future time points are integrated according to the length of the time window to generate concentration correlation change data between the Ganoderma lucidum polysaccharide and ganoderic acid in the future time window.

7. The method according to claim 1, characterized in that The method of generating a spatiotemporal map reflecting the distribution state of metabolites based on the concentration dynamic change information by interactively fusing multimodal data includes: Dividing the dynamic change information of the concentrations of the ganoderma polysaccharide, ganoderma acid and dissolved oxygen into time series segments according to the time dimension; Each time series segment is labeled with metabolic association features; The fermentation container used for Ganoderma lucidum pill fermentation was divided into multiple independent fermentation areas, and the spatial distribution difference ratios of Ganoderma lucidum polysaccharides, ganoderic acid, and dissolved oxygen in each fermentation area were extracted to generate spatial distribution characteristic maps of each fermentation area. The metabolic association feature labeling results are cross-bound with the spatial distribution feature map at the corresponding time to generate a spatiotemporal map.

8. A dynamic feedback control system for the fermentation process of Ganoderma lucidum pills, characterized in that: include: The acquisition module is used to obtain the dynamic change information of the concentration of ganoderma polysaccharides, ganoderma acid and dissolved oxygen in the fermentation liquid during the fermentation process of ganoderma pills; A generation module, configured to generate a spatiotemporal map reflecting the distribution state of metabolites by interactively fusing multimodal data based on the concentration dynamic change information; An input module, configured to input the spatiotemporal map into a pre-trained concentration correlation change prediction model, and output concentration correlation change data between the Ganoderma lucidum polysaccharide and the Ganoderma lucidic acid in a future time window; The correction module is used to iteratively correct the adjustment amount of the fermentation control parameters in the time dimension and the space dimension based on the concentration correlation change data, and generate a feedback control signal to dynamically adjust the fermentation process of the Ganoderma lucidum pills. The fermentation control parameters include the feeding rate and the ventilation volume.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a dynamic feedback control method in a Ganoderma lucidum pill fermentation process as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the dynamic feedback control method in the fermentation process of the ganoderma lucidum pills according to any one of claims 1 to 7 is implemented.

Images