Cordyceps sinensis culture control method and system

Abstract

The invention discloses a cordyceps sinensis culture control method and system, and core culture targets are set in generations by identifying the subculture generations G1, G2 and G3 of cordyceps sinensis strains: the G1 generation activates a cordycepin and adenosine synthesis pathway, the G2 generation maximizes the total nucleoside content and antioxidant activity, and the G3 generation improves the biomass and delays degradation. The system monitors environmental parameters such as temperature and humidity, illumination, gas concentration and the like, and thallus physiological indexes such as biomass, metabolism precursor concentration, active substance content and the like in real time, and adopts algorithms such as fuzzy PID, NSGA-II multi-objective optimization, model prediction control and the like to dynamically adjust culture conditions such as nutrition supply, dissolved oxygen, illumination period and the like. The method provided by the invention solves the problem that traditional static regulation neglects intergenerational difference of strains, realizes subgeneration refined targeted regulation, remarkably improves the yield and quality of target products, prolongs the service cycle of the strains, and is suitable for industrial bionic cultivation of cordyceps sinensis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cultivation, and particularly relates to a method and system for cultivating Ophiocordyceps sinensis. BACKGROUND

[0002] As a key biological resource, the industrial production of edible fungi has strategic significance for meeting the nutritional and health needs of mankind. Among them, rare medicinal fungi such as Ophiocordyceps sinensis have become the core raw material in the field of biological medicine and health care due to their unique physiological active substances and significant pharmacological value. Limited by the scarcity of natural resources, artificial bionic cultivation has become an inevitable direction for industrial development, and subculture is the basis for realizing large-scale production; stable strain sources are provided through strain multiplication to support subsequent industrial cultivation. The bionic cultivation cabin, as the core equipment, mainly simulates natural conditions by adjusting environmental parameters such as culture medium composition, temperature, humidity, light, ventilation, and pH value to ensure mycelium survival and biomass accumulation, laying a foundation for the initial development of Ophiocordyceps sinensis industry.

[0003] The existing artificial cultivation technology mainly relies on static or rough segmented macro-control of environmental parameters. Although the basic substances required for mycelial growth are provided by optimizing the nutrient ratio, the metabolism is promoted by adjusting the temperature and humidity to simulate the natural environment, and the normal respiration and physiological metabolism are maintained by moderate light and ventilation, which supports the subculture of edible fungi and provides stable strain sources for industrial production. However, it has obvious defects. The subculture process is not designed for the differences between generations of strains, and the subculture process is considered as a relatively uniform stage. The macro-control mode based on environmental parameters lacks deep understanding and targeted regulation of physiological characteristics of strains at different generations, and cannot focus on target products for fine and differentiated strategy deployment according to different subculture generations, resulting in insufficient potential of key metabolic product accumulation peak, and timely and accurate intervention to delay degradation of strain activity, affecting production efficiency and product quality, and limiting the preservation period of the strain.

[0004] Therefore, it is necessary to improve the control method and system of the existing Ophiocordyceps sinensis bionic cultivation cabin to solve the above problems. SUMMARY

[0005] The present application overcomes the shortcomings of the prior art and provides a method and system for cultivating Ophiocordyceps sinensis, which aims to solve the problem of ignoring the differences between generations of strains in the subculture process of Ophiocordyceps sinensis in the prior art.

[0006] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows: a method for cultivating Ophiocordyceps sinensis, comprising:

[0007] S1, identifying the subculture generation of the Ophiocordyceps sinensis strain in the current cultivation batch;

[0008] S2, real-time monitoring of environmental parameters and physiological indicators of the fungus in the cultivation cabin;

[0009] S3, according to the identified subculture generation, set the corresponding core culture target; wherein G1 generation takes the activation of the synthesis pathway of cordycepin and adenosine as the target, G2 generation takes the maximization of total nucleotide content and total antioxidant activity as the target, and G3 generation takes the maximization of mycelium biomass and the delay of strain degradation as the target;

[0010] S4, based on the set core target and real-time monitoring data, dynamically adjusting the culture conditions.

[0011] In a preferred embodiment of the present application, identifying the subculture generation comprises: collecting mycelium microscopic morphology images through a machine vision module, extracting mycelium branching density, mycelium diameter uniformity, cell vacuole proportion and mycelium integrity characteristics, and comparing with a pre-stored morphology characteristic-generation matching model to determine the subculture generation.

[0012] In a preferred embodiment of the present application, in step S2, the environmental parameters include temperature, humidity, light intensity and period, carbon dioxide concentration, oxygen concentration, dissolved oxygen concentration and pH value monitored in real time by sensors; and the physiological indicators of the fungus include biomass, metabolite precursor concentration monitored in real time by a special detector, and total nucleotide content and total antioxidant activity predicted by an online spectrum analysis module.

[0013] In a preferred embodiment of the present application, when the identified subculture generation is G1, according to the adenosine concentration feedback by the metabolite precursor detector, the adenosine is dynamically supplemented by a nutrient supplement pump array to maintain its concentration in the range of 0.3g / L to 0.6g / L; the real-time data of the dissolved oxygen detector is received, and the speed of the mechanical stirrer or the shaking table is dynamically adjusted by a fuzzy PID control algorithm to maintain the dissolved oxygen concentration in the culture solution at 60% to 80% saturation; the culture temperature is maintained at 22℃±0.5℃ by a temperature regulator, the relative humidity is maintained at 80%±5% by a humidity regulator, and the pH value is maintained at 6.0±0.2 by a pH regulator.

[0014] In a preferred embodiment of the present application, when the identified subculture generation is G2, the total nucleotide content and total antioxidant activity data predicted by the online spectrum analysis module are received, a multi-objective optimization algorithm based on non-dominated sorting genetic algorithm NSGA-II is run, the maximization of the total nucleotide content and the total antioxidant activity is taken as the optimization target, and the optimal combination of the cultivation cabin temperature, blue light intensity and blue light period parameters is calculated; according to the optimal parameter combination calculated by the multi-objective optimization algorithm, the culture temperature is regulated in the range of 18℃ to 25℃ by a temperature regulator, the blue light intensity is regulated in the range of 100-400μmol·m -2 ·s -1Within the specified range, the blue light cycle is set to 12 to 18 hours per day; the dissolved oxygen concentration is maintained at 70% ± 5% saturation by using a dissolved oxygen detector in conjunction with a mechanical stirrer and a ventilation regulator, and the pH value is maintained at 6.5 ± 0.2 by using a pH detector in conjunction with an acid-base titration system.

[0015] In a preferred embodiment of the present invention, when the identified subculture number is G3, based on the mycelial growth rate fed back by the biomass detector and the real-time data from the carbon source concentration detector, the nutrient supply pump array is adjusted through a model predictive control algorithm to execute a carbon source gradient supply strategy: maintaining the glucose concentration at a certain level during the initial stage of cultivation.

[0016] The total carbon source concentration was maintained at 40 g / L to 50 g / L during the mid-stage of cultivation, with a mixture of glucose and maltose. The total carbon source concentration was maintained at 30 g / L to 40 g / L, and the maltose content was gradually increased to 30%. When the machine vision module detected signs of mycelial degradation, the supply was switched to sucrose, and the total carbon source concentration was maintained at 20 g / L to 30 g / L. When the machine vision module detected signs of mycelial degradation, a short-term low-temperature stress strategy was triggered. The culture temperature was lowered from 24°C to 18°C ​​to 20°C within 24 to 48 hours and maintained for 24 to 48 hours, and then restored to 24°C. The dissolved oxygen concentration was maintained at 85% ± 5% saturation by increasing the aeration rate and stirring speed, and the pH value was maintained at 5.5 ± 0.2 by the pH regulator.

[0017] In a preferred embodiment of the present invention, step S4, dynamically adjusting the culture conditions includes: generating control commands based on the core objective using at least one of a PID control algorithm, a fuzzy PID control algorithm, a model predictive control algorithm, or a multi-objective optimization algorithm, and adjusting the culture conditions through an execution module.

[0018] The present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions, which are used to cause a processor to execute a method for controlling the cultivation of Cordyceps sinensis.

[0019] This invention provides an electronic device, characterized in that it comprises:

[0020] At least one processor; and

[0021] A memory communicatively connected to the at least one processor; wherein,

[0022] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform a method for controlling the cultivation of Cordyceps sinensis.

[0023] This invention provides a Cordyceps sinensis cultivation and control system, comprising:

[0024] The detection module is used to monitor environmental parameters and bacterial physiological indicators in the culture chamber in real time.

[0025] The generation number identification module is used to identify the generation number of the Cordyceps sinensis strain in the current culture batch.

[0026] The data acquisition and preprocessing module is used to acquire and preprocess the data from the detection module and the successive algebra identification module;

[0027] Metabolic model and prediction module, used to predict mycelial growth status and metabolite content based on preprocessed data;

[0028] The multi-objective optimization decision module is used to set core training objectives based on the number of generations and to calculate the optimal combination of environmental parameters through optimization algorithms.

[0029] The control strategy module is used to generate control commands based on the output of the multi-objective optimization decision module;

[0030] The execution module is used to dynamically adjust the culture conditions according to the control instructions;

[0031] The human-computer interaction and alarm module is used for parameter setting, status monitoring and abnormal alarms;

[0032] The data storage module is used to store all data throughout the system process.

[0033] This invention addresses the shortcomings of the prior art and has the following beneficial effects:

[0034] (1) This invention achieves this through precise generational identification and generational core objective setting. The system can accurately identify the generational number of Cordyceps sinensis strains and set core objectives for different physiological characteristics from G1 to G3, namely, activating metabolic pathways, maximizing active substances, ensuring biomass, and delaying degradation. Different generations of strains exhibit significant physiological differences: G1 generation shows no metabolic activation, G2 generation shows peak activity, and G3 generation is prone to degradation. This generational objective setting allows for precise matching of regulation to the strain's current optimal physiological capacity, avoiding a disconnect between objectives and needs. Each generation of strains fully utilizes its own advantages, preventing wasted capacity or unmet needs. Compared to existing technologies that ignore generational differences through static macroscopic control, this system addresses the root cause of regulatory misalignment, further clarifying the direction for subsequent refined regulation and ensuring stable product quality and high yield in industrial production.

[0035] (2) This invention utilizes a multi-dimensional real-time monitoring system. The detection module monitors environmental parameters and bacterial physiological indicators within the cultivation chamber in real time. Environmental parameters include temperature, humidity, light, and gas concentration, while physiological indicators include biomass, metabolic precursor concentration, predicted values ​​of active substances, and mycelial morphology. These dimensions comprehensively cover key factors affecting the growth and metabolism of the strain, dynamically capturing the strain status at different cultivation stages. For example, in the G1 generation, monitoring adenine helps determine precursor supply; in the G2 generation, spectroscopy reveals the accumulation of active substances; and in the G3 generation, machine vision is used to observe signs of degradation, providing comprehensive and accurate data for regulation. The system promptly detects subtle changes, avoiding untimely regulation due to information lag. Compared to existing technologies that only focus on macroscopic environmental monitoring and lack the ability to capture internal bacterial indicators, this system fills a gap, further promoting regulation from passive stabilization to proactive adaptation to metabolic needs, reducing the waste of resources from blind regulation.

[0036] (3) This invention utilizes a differentiated control algorithm. The system employs an adaptive algorithm for the core objectives of different generations: G1 generation uses fuzzy PID to regulate dissolved oxygen and adenine supply; G2 generation uses the NSGA-II algorithm to optimize temperature and blue light parameters; and G3 generation uses model predictive control to execute carbon source gradient supply. The control focus differs for each generation: G1 generation requires stable metabolic conditions, G2 generation requires balancing multiple optimization objectives, and G3 generation requires addressing degradation risks. The adaptive algorithm can accurately calculate the optimal parameter combination, avoiding the limitations of a single algorithm. The culture conditions for each generation are dynamically optimized based on the objectives and real-time data, resulting in flexible and precise control. Compared to existing technologies that use a single, simple algorithm and cannot meet complex metabolic needs, this system effectively overcomes these limitations, further improving resource utilization efficiency, reducing nutrient waste and substandard products, lowering the complexity of manual intervention, and supporting large-scale industrial culture.

[0037] (4) This invention utilizes an active degradation intervention mechanism for G3 generation strains. The system monitors the morphology of G3 generation mycelia in real time using machine vision. Upon identifying signs of degradation, it automatically triggers a carbon source gradient supply and short-term low-temperature stress strategy. The carbon source gradually switches from glucose to sucrose, and the low-temperature stress is achieved through phased cooling and recovery. G3 generation strains are prone to degradation due to decreased metabolism. If intervention is not timely, it will affect biomass and strain utilization. This active intervention can adjust conditions in the early stages of degradation and slow down aging. It effectively delays the degradation process and ensures stable accumulation of G3 generation mycelial biomass. Compared with the shortcomings of existing technologies that lack active monitoring and intervention and only passively adjust, this system achieves early prevention and control of degradation risks, further extends the strain utilization cycle, reduces the cost of mother culture preparation, improves the continuity of industrial production, and provides a high-quality source of mycelia for subsequent preservation and propagation. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a perspective structural diagram of a preferred embodiment of the present invention;

[0040] Figure 2 This is a schematic diagram of the overall process of a preferred embodiment of the present invention;

[0041] Figure 3 This is a preferred embodiment of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein. Therefore, the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0044] Application Overview:

[0045] This application addresses the need for biomimetic cultivation control during the 1st to 3rd generation (G1-G3) subculture of Cordyceps sinensis. These three generations were selected because numerous studies have confirmed that G1 to G3 represent a critical window period for Cordyceps sinensis mycelium to adapt to its environment, reach its metabolic peak, and exhibit growth advantages. Key quality indicators such as mycelial biomass, nucleoside content, and antioxidant activity show regular and significant intergenerational differences during this stage.

[0046] The G1 generation, as the initial generation, exhibits strong mycelial adaptability but unstable product accumulation; the G2 generation typically reaches the peak of antioxidant activity and nucleoside compounds, representing a critical period for optimizing active ingredients; the G3 generation easily achieves maximum biomass but carries the risk of degradation.

[0047] The fundamental limitation of existing biomimetic cultivation chamber control technology is that its regulation logic is based on treating the cultivation process as a whole or making extensive segments, which seriously ignores the generational specificity of the intrinsic physiological characteristics and metabolic pathways of strains at different subgenerations.

[0048] The root cause of the difficulty in achieving breakthroughs in existing technologies lies in the fundamental flaw of their environment-first logic: traditional control focuses on maintaining macro-environmental stability, relying on passive adaptation to preset parameters, and lacks a perception and response mechanism for the dynamic physiological metabolism of strains across generations. There is a significant shift in metabolic pathways between the first and third generations of Cordyceps sinensis: G1 is dominated by mycelial colonization, G2 emphasizes the synthesis of active substances, and G3 shifts towards biomass accumulation. The requirements for parameters such as nutrient ratios and aeration efficiency differ significantly at each stage.

[0049] This application breaks with convention by constructing a biomimetic cultivation chamber control method and system that can accurately identify, target, regulate, and actively optimize the cultivation goals of each generation. This invention does not simply optimize environmental parameters, but redefines the control goals, that is, it sets a priority output goal that best matches the physiological characteristics of each generation, so that the control strategy has a clear biological orientation.

[0050] Exemplary method:

[0051] like Figure 1 , Figure 2 As shown, a method for controlling the cultivation of Cordyceps sinensis includes:

[0052] S1. Identify the generation number of the Cordyceps sinensis strain in the current culture batch;

[0053] S2. Real-time monitoring of environmental parameters and bacterial physiological indicators within the cultivation chamber;

[0054] S3. Based on the identified subgenerations, set the corresponding core culture objectives; among them, the G1 generation aims to activate the synthesis pathways of cordycepin and adenosine, the G2 generation aims to maximize the total nucleoside content and total antioxidant activity, and the G3 generation aims to maximize mycelial biomass and delay strain degradation.

[0055] S4. Based on the set core objectives and real-time monitoring data, dynamically adjust the cultivation conditions.

[0056] In the execution process of the Cordyceps sinensis biomimetic cultivation chamber control method, step S1, as the starting point of the entire control process, is fundamentally about providing precise input for subsequent differentiated control strategies through a systematic generation identification mechanism. This step determines the generation number of the current cultivation batch in real time by integrating a human-machine interface or automatically retrieving historical culture records of strains from the data storage module. This ensures that the control system can quickly load the preset parameter set and optimization target that matches the generation, thus laying the foundation for generational fine-grained control.

[0057] In step S1, the number of subcultures refers to the number of times a Cordyceps sinensis strain, starting from the original mother strain (marked as G0), undergoes cyclical subcultures of inoculation, culture, isolation, and re-inoculation during artificial culture. Each additional subculture is recorded as one generation, i.e., G1, G2, G3, and so on. The physiological characteristics of strains with different subcultures differ significantly. G1 strains have metabolic pathways that need to be activated, G2 strains have a strong ability to accumulate active substances, and G3 strains are prone to degradation. This is the core basis for formulating differentiated control strategies.

[0058] The physiological and metabolic characteristics of Cordyceps sinensis strains change significantly with each generation, specifically:

[0059] The G1 generation strain has just been isolated from the parent strain and its metabolic pathways have not been fully activated. It is necessary to specifically regulate precursor supply and dissolved oxygen, with the core objective being to activate metabolism.

[0060] The G2 generation strain reaches its peak physiological activity, with the strongest total nucleoside synthesis capacity and antioxidant activity. It is necessary to regulate parameters such as temperature and blue light to maximize the accumulation of active substances. The core objective is quality optimization.

[0061] The G3 generation strain began to show signs of degradation, and it is necessary to focus on carbon source gradient supply and low temperature stress to delay degradation, while maximizing biomass. The core objectives are biomass protection and degradation delay.

[0062] If the number of generations cannot be obtained, the system will be unable to match the corresponding refined control strategy and will only be able to use a one-size-fits-all, coarse parameter. This will result in the target products of different generations of strains failing to meet the standards, and may even accelerate strain degradation, which violates the core design concept of generational targeted regulation of this invention. Therefore, the number of generations is the core parameter of step S1 and is the starting point for all subsequent regulatory logic.

[0063] In conjunction with the Cordyceps sinensis biomimetic cultivation chamber control system described in this invention, the generation number is obtained through the following two methods to ensure accuracy and convenience;

[0064] Firstly, the key information of the strain to be cultured can be manually entered by the operator through the human-computer interaction interface, including the strain number, parent culture source, and subculture number. After input, the information is transmitted to the subculture number identification module of the control unit in real time. This method is suitable for new strains or cases without historical records.

[0065] Secondly, the generation identification module of the control unit accesses the data storage module through a high-speed data interface and automatically retrieves the historical database based on the strain number or culture batch number input by the operator.

[0066] The historical records pre-stored in the data storage module include: strain number, culture batch, subculture number, and related information of culture results;

[0067] After the generation number identification module matches the corresponding record, it automatically extracts the generation number of the current strain, eliminating the need for manual re-entry and reducing operational errors.

[0068] Furthermore, the generation number is determined by recognizing the microscopic morphology of mycelia through the machine vision module already equipped in the system.

[0069] By acquiring microscopic morphological images of hyphae using a machine vision module and performing feature analysis, automatic identification of the number of generations can be achieved. This method does not rely on a historical database and can be used independently as a supplementary solution for obtaining the number of generations. It is especially suitable for scenarios such as database failure, missing historical records, or cross-system transfer of strains.

[0070] The specific steps are as follows:

[0071] After the operator completes the culture medium preparation and strain inoculation, they can find and click the image recognition generation function button in the human-machine interface to trigger the start of the generation image recognition process.

[0072] After receiving the image recognition generation start command, the control unit sends a working command to the machine vision module to perform the following operations:

[0073] Turn on the adjustable ring light source and adjust the wavelength of the light source to the 450-500nm blue light band; then, control the microscope head with a magnification of 500-1000X to move to the preset mycelial sampling area in the cultivation chamber, ensuring that the microscope head is aligned with the center of the sampling area; finally, control the high-resolution industrial camera to continuously acquire 3-5 mycelial micro images according to preset parameters.

[0074] The image acquisition card receives raw image data captured by the industrial camera in real time and transmits the raw image data completely to the processing unit related to successive algebra recognition in the control unit, waiting for subsequent analysis and processing;

[0075] The algebraic recognition related processing unit preprocesses the received raw image data, converting the color image into an 8-12 bit grayscale image to complete the image grayscale processing; then, a Gaussian filter with a size of 5×5 is used to filter the grayscale image to remove noise interference in the image; finally, the filtered image is processed by the Otsu threshold segmentation method to separate the mycelial region from the background region such as culture medium impurities and bubbles, generating a binary mycelial image.

[0076] After image preprocessing, the processing unit proceeds to the feature extraction stage. It uses an algorithm to extract four core morphological features that are strongly correlated with generation number. The specific extraction method is as follows:

[0077] The number of hyphal branching points within a unit field of view is counted to obtain the hyphal branching density.

[0078] The diameter of the hyphal trunk was calculated using a skeletal algorithm, and the uniformity of the hyphal diameter was then calculated using the standard deviation formula.

[0079] By using image segmentation technology to distinguish between cytoplasm and vacuolar regions, and by calculating the proportion of vacuolar region area to total cytoplasm area, the vacuolar proportion of the cell is obtained.

[0080] The proportion of broken hyphal fragments within the field of view to the total number of hyphal fragments is used to determine hyphal integrity.

[0081] After feature extraction is completed, the processing unit enters the feature matching and judgment stage, which includes the following operations:

[0082] The pre-stored morphological feature-generation matching model is invoked, and the four types of feature parameters extracted in real time are input into the model and compared one by one with the standard feature thresholds of generations G1, G2, and G3.

[0083] If the branch density is 3-5 per field of view, the standard deviation of hyphal diameter is ≤0.5μm, the proportion of cell vacuoles is ≤5%, and the proportion of hyphal integrity breaks is ≤10%, then the current strain is determined to be of generation G1.

[0084] If the branch density is 8-12 per field of view, the standard deviation of hyphal diameter is ≤0.3μm, the proportion of cell vacuoles is ≤3%, and the proportion of hyphal integrity breaks is ≤5%, then the current strain is determined to be the G2 generation.

[0085] If the branch density is 1-2 per field of view, the standard deviation of hyphal diameter is ≥0.8μm, the proportion of cell vacuoles is ≥15%, and the proportion of broken hyphae is ≥30%, then the current strain is determined to be the G3 generation.

[0086] Among them, the morphological feature-generation matching model was generated by training a support vector machine algorithm with 1,000+ sets of mycelial images of different generations in the early stage, and the generation determination accuracy was ≥92%.

[0087] The control unit will clearly display the final determined generation number on the human-machine interface, and automatically retrieve the generation control parameter set corresponding to that generation from the pre-stored parameter library to complete the parameter loading operation.

[0088] Step S1, serving as the starting point for the biomimetic cultivation control of Cordyceps sinensis, focuses on accurately determining the generation number of the current culture batch through human-computer interaction input or automatic retrieval from the data storage module, combined with the identification of mycelial micromorphological features by the machine vision module. This step provides crucial information for subsequent differentiated regulation. Because the physiological characteristics of strains at different generations differ significantly, only by clearly identifying the generation can targeted monitoring dimensions and control strategies be matched in subsequent steps, avoiding the problem of substandard target products caused by extensive regulation, thus naturally connecting to the real-time monitoring stage in step S2.

[0089] In step S2, environmental parameters and bacterial physiological indicators within the cultivation chamber are monitored in real time.

[0090] Specifically, environmental parameters refer to the physical and chemical factors that affect mycelial growth within the cultivation chamber, including temperature, humidity, light, and gas composition. These parameters directly regulate the metabolic activities and growth rate of the mycelium.

[0091] Biological parameters that reflect the internal state of hyphae and metabolic output include biomass, precursor concentration, metabolite content and morphological characteristics, and are used to assess strain health and productivity.

[0092] Environmental parameter monitoring includes:

[0093] The temperature detector is used to obtain the temperature of the culture medium or the ambient temperature in the cultivation chamber in real time, with a measurement range of 0-50℃ and an accuracy of ±0.1℃.

[0094] The relative humidity inside the cultivation chamber is acquired in real time using a humidity detector, with a measurement range of 10%-99%RH and an accuracy of ±2%RH.

[0095] The light intensity inside the cultivation chamber was acquired in real time using a light detector, ranging from 0 to 1000 μmol·m⁻¹. -2 ·s -1 Spectral distribution and illumination period;

[0096] The carbon dioxide concentration in the cultivation chamber, ranging from 0 to 5000 ppm, and the oxygen concentration, ranging from 0 to 25% O2, were obtained in real time using a ventilation detector.

[0097] The dissolved oxygen concentration in the culture medium was obtained in real time using a dissolved oxygen detector, ranging from 0 to 20 mg / L, with an accuracy of ±0.05 mg / L.

[0098] The pH value of the culture medium is obtained in real time using a pH detector, with a measurement range of 2-10 and an accuracy of ±0.01.

[0099] The monitoring of bacterial physiological indicators includes:

[0100] Biomass detectors are used to acquire mycelial biomass indicators in real time based on either dielectric constant measurement or optical density principles. When based on dielectric constant measurement, the dielectric properties of the mycelial cell membrane are utilized to reflect cell density by detecting changes in the dielectric constant of the culture medium. When based on optical density, the turbidity OD600 value is obtained by detecting the degree of absorption of light of a specific wavelength by the mycelium. Both methods can accurately characterize the biomass accumulation during mycelial growth, providing a quantitative basis for subsequent nutrient supply and environmental parameter adjustments.

[0101] The metabolite precursor detector uses an online high-performance liquid chromatography (HPLC) module as its core detection component, enabling real-time monitoring of precursor substances such as adenine in the culture medium. The detector boasts high sensitivity, with a detection limit of 0.01 mg / L for precursor substances like adenine. It accurately captures concentration fluctuations of precursor substances in the culture medium, ensuring timely detection of insufficient or excessive precursor substances during mycelial metabolism. This provides crucial data support for precisely controlling the precursor replenishment rate and ensuring the synthesis of target products such as cordycepin and adenosine.

[0102] The online spectral analysis module is constructed using a near-infrared spectrometer, covering a spectral range of 900-1700 nm, enabling real-time acquisition of spectral data from the culture medium. After acquiring the spectral data, the module invokes a pre-set chemometric model, specifically a partial least squares regression (PLS) model. Through analysis and calculation of the spectral data, it can accurately predict the content of key metabolites in the culture medium, such as total nucleoside content and total antioxidant activity, without the need for offline sampling and detection, thus achieving real-time monitoring of mycelial metabolite accumulation.

[0103] The machine vision module consists of a high-resolution industrial camera, a microscope lens, an adjustable light source, and an image acquisition card. It can periodically acquire macroscopic and microscopic morphological images of mycelia. After image acquisition, the module runs dedicated image processing algorithms, including skeletonization algorithms and fractal dimension analysis algorithms. These algorithms process and analyze the images to extract multiple morphological features of the mycelia, specifically including branch point density, mycelial diameter uniformity, cell vacuolation degree, and mycelial integrity. This allows for the assessment of mycelial growth status and the presence of signs of degradation, providing visual data support for interventions to slow down mycelial degradation.

[0104] Real-time monitoring of environmental parameters and bacterial physiological indicators is a core prerequisite for achieving precise generational regulation. Strains of different generations exhibit significantly different physiological characteristics: the G1 generation requires activation of metabolic pathways, necessitating monitoring of adenine concentration and dissolved oxygen to ensure precursor supply and aerobic metabolism; the G2 generation requires maximizing the accumulation of bioactive substances, thus requiring near-infrared spectroscopy to predict total nucleosides and antioxidant activity to optimize temperature and light; the G3 generation requires delaying degradation, thus requiring machine vision monitoring of hyphal morphology changes to trigger timely interventions. Without this monitoring, the control system will be unable to capture generational specific needs, resulting in a crude regulatory strategy that fails to maximize target products or effectively delay degradation.

[0105] Cordyceps sinensis strains exhibit intergenerational physiological differences during subculture, and macro-environmental stability alone cannot meet the optimization goals of each generation.

[0106] Through multimodal real-time monitoring, the system can dynamically sense changes in the state of the strain and make precise interventions in combination with generational control strategies. This enables the system to activate metabolism in the G1 generation, improve quality in the G2 generation, and ensure biomass and delay degradation in the G3 generation, ultimately solving the problems of extensive control strategies and lack of generational targeting in existing technologies.

[0107] In step S3, the Cordyceps sinensis strain exhibits significant differences in physiological characteristics across different generations. If a specific core culture objective is not set according to the generation, the regulatory direction will not match the strain's capabilities. Specifically: if the core objective for the G1 generation is biomass maximization, the strain's metabolic pathways are not activated, and nutrients such as carbon sources will be prioritized for basic growth rather than metabolic pathway construction, thus inhibiting the synthesis of active substances in subsequent generations; if the core objective for the G2 generation is to delay degradation, the peak capacity for active substance accumulation will be wasted, leading to a reduction in medicinal value; if the core objective for the G3 generation is active substance accumulation, the strain has already shown signs of degradation and decreased metabolic capacity, making it difficult to increase the content of active substances and accelerating strain aging.

[0108] Only by setting core culture objectives based on generation can the regulatory strategy focus on the optimal capabilities of the strain at the current stage, fully leverage the physiological advantages of each generation, avoid its inherent shortcomings, provide a clear direction for subsequent precise adjustments to culture conditions, and ensure that each generation can achieve the optimal output for the corresponding stage.

[0109] The core culture objectives are set based on the differences in physiological characteristics of strains across generations, the pre-set generational target system, and the needs of industrial production.

[0110] Among them, the differences in physiological characteristics between generations of strains are the fundamental basis for setting targets.

[0111] The G1 generation strains, having just been isolated from the parent strain, have not yet fully activated the cordycepin and adenosine synthesis pathways, and their physiological capabilities are concentrated on metabolic system construction. Therefore, the core objective should be to activate these pathways. The G2 generation strains reach peak physiological activity, with the strongest total nucleoside synthase activity and antioxidant production capacity. The core objective should be to maximize the accumulation of these active substances. The G3 generation strains begin to decline in metabolic activity, and the hyphae are prone to degenerative phenomena such as breakage and vacuolation. The core objective should be to balance biomass accumulation and delaying degeneration.

[0112] The control unit pre-stores a target system based on a large amount of experimental data, which includes core targets, related indicators and parameter thresholds for each generation.

[0113] Target setting must balance technical feasibility and industrial value. The G1 generation activates metabolism to lay the foundation for high-quality production in subsequent generations, avoiding yield fluctuations caused by inactive metabolic pathways. The G2 generation focuses on active substances because these components are the core of Cordyceps sinensis' medicinal value and directly affect the product's economic benefits. The G3 generation ensures biomass and delays degradation to extend the strain's lifespan, reduce the cost of mother culture preparation, and improve the continuity of industrial production.

[0114] The core culture objective refers to the highest priority culture direction set for the physiological and metabolic characteristics of Cordyceps sinensis strains at different generations. It is the core guide for subsequent dynamic adjustment of culture conditions, such as metabolic pathway activation in the G1 generation, accumulation of active substances in the G2 generation, and biomass assurance and degradation delay in the G3 generation.

[0115] The generational target matching logic refers to the rule system pre-stored within the control system that associates successive generations with core culture objectives. This logic is constructed based on a large amount of strain culture experimental data to ensure that the objectives of each generation match the physiological capabilities of the strain, avoiding a disconnect between the objectives and the characteristics of the strain.

[0116] The specific operations of step S3 include:

[0117] When the generation number is G1, the core objective is to initiate the synthesis of cordycepin and adenosine, that is, to activate the relevant metabolic pathways and lay the foundation for the accumulation of metabolites in subsequent generations. The G1 generation strain has just been isolated from the parent species, and the metabolic pathways are not fully activated. Targeted regulation is needed to provide conditions for the construction of metabolic pathways and avoid the inhibition of the synthesis of active substances in subsequent generations.

[0118] Dissolved oxygen concentration in the culture medium is obtained in real time by a dissolved oxygen detector to ensure that the monitoring data reflects the aerobic metabolic state of the mycelium; adenine concentration in the culture medium is detected in real time by a metabolite precursor detector. Adenine is a key precursor for the synthesis of cordycepin and adenosine, and its supply needs to be monitored in real time; mycelial growth is monitored in real time by a biomass detector to determine whether the basic growth of mycelium is normal, providing a basic guarantee for the activation of metabolic pathways.

[0119] In the early stage of mycelial growth, adenine was added to the culture medium via a nutrient supply pump array as a biological precursor for the synthesis of cordycepin and adenosine. The initial addition concentration was set at 0.3-0.5 g / L. During subsequent culture, the adenine concentration was maintained in the range of 0.3 g / L to 0.6 g / L based on the adenine concentration feedback from the metabolite precursor detector to ensure a sufficient supply of precursors.

[0120] The peristaltic pump flow rate is precisely adjusted using a PID control algorithm. The PID controller parameters are set to a proportional coefficient of 0.8, an integral coefficient of 0.1, and a derivative coefficient of 0.05, with a control cycle of 5 minutes. The algorithm responds to the deviation between the online detected adenine concentration and the target concentration, ensuring a continuous and sufficient supply of precursor substances.

[0121] The control unit receives real-time data from the dissolved oxygen detector and internally runs a fuzzy PID control algorithm, using dissolved oxygen concentration as the control variable. The optimal dissolved oxygen concentration range is preset to 60% to 80% saturation. The inputs to the fuzzy PID control algorithm include the dissolved oxygen concentration deviation and the rate of change of deviation, while the output is the control signal for the variable frequency motor of the mechanical stirrer or shaker. The fuzzy rule base has multiple preset rules, constructed based on expert knowledge and previous experimental data, which can effectively handle dissolved oxygen concentration fluctuations and hysteresis. The variable frequency motor operates at a frequency range of 50Hz to 150Hz, corresponding to a shaker speed of 80rpm to 200rpm. By dynamically adjusting the stirring rate, the dissolved oxygen concentration is precisely controlled, optimizing the aerobic metabolism of the mycelium and promoting the activation of metabolic pathways.

[0122] Other environmental parameter controls include:

[0123] The culture temperature was set at 22℃±0.5℃, relative humidity at 80%±5%, and pH at 6.0±0.2. Temperature, humidity, and pH regulators were maintained stable through independent PID control loops. The temperature PID controller parameters were set to a proportional coefficient of 1.5, an integral coefficient of 0.05, and a derivative coefficient of 0.1, with a control cycle of 30 seconds. Each regulator adjusted in real time based on feedback data from its corresponding detector, providing a stable environment for mycelial growth and metabolic activation.

[0124] The metabolic pathways of the G1 generation strain were not activated. Adenine is a key precursor for the synthesis of cordycepin and adenosine, and maintaining its concentration range can ensure sufficient metabolic raw materials. Dissolved oxygen affects the aerobic metabolic efficiency of mycelia. Fuzzy PID control can respond quickly to concentration changes and avoid metabolic obstruction due to hypoxia. Stable control of temperature, humidity and pH can reduce the interference of environmental fluctuations on the initial growth of mycelia and create suitable conditions for the activation of metabolic pathways.

[0125] When the generation number is G2, the core targets are set on total nucleoside content and total antioxidant activity. Total nucleosides include adenosine, guanosine, uridine, etc., aiming to maximize the medicinal value of Cordyceps sinensis. The physiological activity of the G2 generation strain reaches its peak, with the strongest total nucleoside synthesis capacity and antioxidant activity. It is necessary to focus on this advantage to achieve efficient accumulation of active substances.

[0126] The near-infrared spectral data of the culture medium is acquired in real time by the online spectral analysis module. The control unit calls the pre-established partial least squares regression model or support vector regression model to predict the total nucleoside content and related antioxidant activity indicators in real time from the spectral data. The temperature in the culture chamber is monitored in real time by the temperature detector, and the blue light intensity is monitored in real time by the light detector, providing data support for subsequent parameter optimization.

[0127] The control unit internally runs a multi-objective optimization algorithm based on the non-dominated sorting genetic algorithm NSGA-II. The optimization objective is to maximize the predicted total nucleoside content and total antioxidant activity, while constraining the moderate growth of mycelial biomass to avoid excessive nutrient consumption that could affect the synthesis of active substances. The decision variables of the algorithm include the culture chamber temperature, blue light intensity, and blue light cycle. The temperature optimization range is 18℃ to 25℃, and the blue light intensity range is 100-400 μmol·m⁻¹. -2 ·s -1 The blue light cycle is set to 12 to 18 hours per day; the algorithm population size is set to 100, the maximum number of iterations is 200, the crossover probability is 0.9, and the mutation probability is 0.1. The optimal parameter combination is determined by generating non-dominated solutions and according to preset priorities.

[0128] The multi-objective optimization function is

[0129] Maximize F(T,I,P)=[f1(T,I,P),f2(T,I,P)];

[0130] in, This is the predicted value for total nucleoside content; The total antioxidant activity is the predicted value; T is the culture temperature; I is the blue light intensity; P is the blue light period.

[0131] Constraints: in, θ represents the mycelial biomass growth rate; θ is the biomass growth rate threshold.

[0132] Based on the optimal combination of temperature and blue light irradiation parameters calculated by the multi-objective optimization algorithm, the environment inside the cultivation chamber is precisely adjusted in real time through temperature and light regulators. The online spectral analysis module continuously predicts the total nucleoside content and total antioxidant activity indicators, and uses the prediction results as real-time feedback to the multi-objective optimization algorithm to dynamically adjust the optimal parameters and ensure that the core target indicators reach their peak values.

[0133] Environmental parameter control includes:

[0134] Temperature-metabolic response relationship:

[0135] Among them, R T T is the temperature response coefficient. minT opt T max These are the minimum, optimal, and maximum temperature thresholds, respectively.

[0136] Blue light intensity - product synthesis relationship: Among them, S I K is the blue light-promoted product synthesis coefficient. I β is the half-saturation constant; β is the light suppression coefficient.

[0137] Other environmental parameter controls include: maintaining the dissolved oxygen concentration in the culture medium at 70% ± 5% saturation, achieved through closed-loop control using a dissolved oxygen detector in conjunction with a mechanical stirrer and aeration regulator; maintaining the pH value of the culture medium at 6.5 ± 0.2, regulated through a pH detector in conjunction with an acid-base titration system; and maintaining the aeration rate at 1.0 vvm, stabilizing the gas supply through a ventilation detector in conjunction with a mass flow controller, thus providing a stable metabolic environment for the accumulation of active substances.

[0138] The G2 generation strain is at its peak physiological activity, with the highest activity of enzymes related to total nucleoside synthesis and antioxidant activity. Real-time monitoring of core indicators and dynamic optimization of key parameters are necessary to maximize the accumulation of active substances. Near-infrared spectroscopy analysis can achieve non-destructive real-time detection, avoiding interference from offline sampling in the culture process. The non-dominated sorting genetic algorithm II can find the optimal parameter combination under multi-objective constraints, taking into account both total nucleoside content and antioxidant activity. Stable environmental parameters can reduce external interference and ensure the continuous and efficient operation of metabolic pathways.

[0139] like Figure 3 As shown, when the generation number is G3, the core objectives are to maximize mycelial biomass and effectively delay strain degradation. The G3 generation strain begins to show signs of degradation, manifested as slower mycelial growth, reduced branching, and increased cell vacuolation. Therefore, intervention measures are needed to delay degradation and extend the strain's lifespan while ensuring biomass accumulation.

[0140] The machine vision module periodically acquires microscopic morphological images of mycelia every 4-6 hours. The control unit runs image processing algorithms, including skeletonization, fractal dimension analysis, branch point counting, mycelial diameter measurement, and cell wall integrity analysis algorithms, to extract mycelial morphological features from the image data and determine whether there are signs of degradation, such as increased mycelial breakage rate, decreased branch point density, and cell vacuolation exceeding 20%. The biomass detector monitors the mycelial growth rate in real time, and the carbon source concentration detector monitors the carbon source concentration in the culture medium in real time, providing a basis for carbon source replenishment.

[0141] The control unit implements a gradient supply of carbon source through a nutrient supply pump array based on the mycelial growth rate fed back by the biomass detector and the real-time data from the carbon source concentration detector.

[0142] Initially, a high concentration of glucose was used, with an initial addition concentration of 50 g / L, which was maintained in the range of 40 g / L to 50 g / L to meet the carbon source requirements for rapid mycelial growth.

[0143] During the mid-term, when mycelial growth reaches a plateau, the supply of glucose and maltose is switched to a mixture of glucose and maltose, with the total carbon source concentration maintained in the range of 30 g / L to 40 g / L. The proportion of maltose gradually increases to 30%, providing a more sustained energy supply.

[0144] Later, when the machine vision module determined that the mycelium showed signs of degradation, the carbon source was switched to sucrose, and the total concentration was maintained in the range of 20 g / L to 30 g / L. The slow hydrolysis characteristics of sucrose can reduce the metabolic burden on the mycelium.

[0145] The carbon source replenishment rate is precisely adjusted through a model predictive control algorithm. This algorithm is based on a mycelial growth kinetics model and a carbon source consumption model. The prediction time domain is 24 hours, and the control time domain is 3-4 hours. The optimization objective is to minimize the deviation between the carbon source concentration and the target value, while avoiding osmotic pressure stress caused by excessively high concentrations.

[0146] When the machine vision module detects signs of mycelial degradation, the control unit immediately triggers a short-term low-temperature stress strategy. The temperature regulator lowers the temperature in the cultivation chamber from the normal culture temperature of 24°C to 18°C ​​to 20°C for 24 to 48 hours, and then restores it to 24°C. Short-term low-temperature stress can slow down the mycelial metabolic rate, reduce the production of reactive oxygen species and cell damage, thereby effectively delaying strain degradation.

[0147] Other environmental parameter controls include: maintaining the dissolved oxygen concentration in the culture medium at 85% ± 5% saturation, ensuring sufficient oxygen supply by increasing the aeration rate and stirring speed to support high biomass growth; maintaining the pH of the culture medium at 5.5 ± 0.2, as the slightly acidic environment can inhibit the growth of miscellaneous bacteria to a certain extent, while also meeting the metabolic needs of G3 generation mycelia; and maintaining the aeration rate at 1.5 vvm, with precise control of gas flow rate through an aeration regulator to ensure a continuous and sufficient supply of oxygen.

[0148] G3 generation strains are prone to degradation. The machine vision module can provide early warning of degradation through changes in morphological characteristics, avoiding a significant drop in biomass due to delayed detection. The carbon source gradient supplementation adjusts the type and concentration of carbon sources according to the needs of different mycelial growth stages. High glucose in the early stage supports rapid growth, mixed carbon sources maintain metabolism in the middle stage, and sucrose slows down degradation in the later stage. Low temperature stress reduces cell damage by lowering the metabolic rate and is a key intervention to delay degradation. High dissolved oxygen and high aeration rate can meet the oxygen requirements of high biomass growth, and a slightly acidic pH can improve the stability of the culture environment.

[0149] Quantitative model for mycelial degradation:

[0150] Comprehensive Degradation Index

[0151] Among them, B t V represents the mycelial branching point density at time t, and B0 represents the initial branching point density. t Vt represents the cell vacuoleization ratio at time t, and V0 represents the initial vacuoleization ratio. max The maximum permissible vaping ratio; F t Let F0 be the hyphal integrity at time t, and w1, w2, and w3 be the initial hyphal integrity.

[0152] The objective function for predictive control of the carbon source replenishment model is:

[0153] Where, N p For prediction in the time domain; N c To control the time domain; C target The target carbon source concentration; C t+k Predict the carbon source concentration at time t+k; Δu t+k λ represents the change in carbon source replenishment rate at time t+k; λ is the control weight coefficient.

[0154] The G3 generation optimizes the carbon source replenishment strategy through model prediction and control, and triggers low-temperature stress intervention by combining real-time monitoring of the degradation index, effectively delaying strain degradation while ensuring biomass accumulation.

[0155] Step S4, based on the core cultivation objectives set in step S3 and the real-time monitoring data obtained in step S2, dynamically adjusts the cultivation conditions in the cultivation chamber through the execution module, forming a closed-loop control of perception-decision-execution.

[0156] The control unit's integrated decision engine receives real-time data streams from various monitoring modules, performing data fusion and status assessment at a frequency of 10-100 times per second. Based on the core objectives of the current generation, the decision engine invokes the corresponding control algorithm to generate a precise set of control commands.

[0157] G1 generation metabolic activation control uses precursor concentration and dissolved oxygen as key regulatory variables to generate adenine supply rate commands and stirring speed commands.

[0158] G2 generation quality optimization control uses temperature and blue light parameters as optimization variables to generate temperature setpoint instructions and light parameter instructions;

[0159] G3 generation biomass protection control focuses on carbon source type and concentration, generating carbon source replenishment strategy instructions and temperature stress trigger instructions;

[0160] The execution module receives control commands via an industrial Ethernet bus, and each actuator responds and executes the corresponding action within 10 milliseconds, ensuring the real-time performance and accuracy of the control.

[0161] Exemplary system:

[0162] A Cordyceps sinensis cultivation and control system includes: a detection module, a generation number identification module, a data acquisition and preprocessing module, a metabolic model and prediction module, a multi-objective optimization decision-making module, a control strategy module, an execution module, a human-computer interaction and alarm module, and a data storage module.

[0163] The detection module is used to collect environmental parameters and physiological indicators of the fungus that affect the growth of Cordyceps sinensis in the cultivation chamber in real time, providing multi-dimensional data input for subsequent regulation and decision-making. It is the core sensing layer for generational fine regulation.

[0164] Environmental parameter monitoring achieves comprehensive coverage through dedicated sensors. Temperature detectors acquire real-time temperature of the culture medium and the chamber environment, humidity detectors collect relative humidity, light detectors monitor the specific spectral distribution of photosynthetically active radiation and the photoperiod, ventilation detectors acquire carbon dioxide and oxygen concentrations, and dissolved oxygen and pH detectors collect dissolved oxygen and pH of the culture medium, respectively. Monitoring of bacterial physiological indicators quantifies biological state through dedicated detection components. Biomass detectors characterize mycelial cell density and activity in real time, metabolite precursor detectors monitor the concentration of precursor substances such as adenine, online spectral analysis modules predict total nucleoside content and total antioxidant activity using specific models, and machine vision modules acquire macroscopic and microscopic images of mycelia for morphological feature analysis. The detection module also performs preliminary filtering and calibration on the collected raw data, transmitting it in real-time to the data acquisition and preprocessing module via a specific bus to ensure data timeliness and accuracy.

[0165] The generation number identification module is responsible for accurately determining the generation number of the current batch of Cordyceps sinensis strains, providing a core decision-making basis for generational control strategies and avoiding a one-size-fits-all, crude control approach. This module supports three complementary identification paths to cover different application scenarios: manual input identification is suitable for new strains or scenarios without historical records; operators input the strain number, parent strain origin, and generation number through a human-machine interface, and the information is synchronized to the module in real time; historical data retrieval accesses the data storage module through a high-speed data interface, automatically matching pre-stored strain generational culture records with the input strain number or batch number to extract the generation number and reduce human error; machine vision automatic identification serves as an independent supplementary solution for database failures or cross-system transfer scenarios, controlling the ring-shaped blue light after receiving a start command. Once the source is activated, the microscope lens moves to the sampling area and continuously acquires multiple microscopic images of hyphae. After converting the color images to grayscale, Gaussian filtering is used for noise reduction. The Otsu threshold segmentation method is used to separate the hyphae from the background to generate a binary image. Four core morphological features—branch density, diameter uniformity, cell vacuolar ratio, and hyphal integrity—are extracted using an algorithm. A pre-trained support vector machine morphological feature generation matching model is called to compare the feature thresholds and determine the generation. The generation recognition module synchronizes the determination results to the multi-objective optimization decision module and the refined control strategy module, and displays them on the human-computer interaction interface, while simultaneously triggering the loading of the pre-stored generation parameter set.

[0166] The data acquisition and preprocessing module serves as the system's data hub, responsible for integrating data from various modules and standardizing it to ensure data quality meets the needs of subsequent modeling and decision-making. This module receives raw sensor data from the detection module, generation results from the algebraic identification module, and feedback data from the execution module in real time via an industrial bus, forming a unified data inflow channel. In the data processing stage, statistical methods are used to remove abnormal data caused by sensor malfunctions or interference. Sampling data from different sensors are aligned based on timestamps to avoid time-series bias. Data of different dimensions is converted into standardized values ​​for easier model calculation. The processed data is also encapsulated in a specific format and supplemented with metadata. The data acquisition and preprocessing module outputs the processed data in two paths: one path is stored in real-time in the data storage module, and the other path is transmitted to the metabolic model and prediction module as input for prediction calculations. Simultaneously, a data processing log is recorded to ensure data traceability.

[0167] The metabolic model and prediction module, based on real-time monitoring data, predicts mycelial growth status and target metabolite content through preset models, enabling proactive regulation and avoiding quality loss due to delayed regulation. This module integrates three specialized prediction models to cover the core needs of different generations. For G2 generation total nucleosides and antioxidant activity, partial least squares regression or support vector regression models are used, inputting near-infrared spectral data and environmental parameters to output real-time predicted values. For G3 generation biomass, a mycelial growth kinetic model is used, inputting carbon source concentration, dissolved oxygen, and temperature data to predict the mycelial growth rate and peak biomass for the next 24 hours. For G3 generation degradation risk, morphological features extracted by machine vision are input, and a logistic regression model calculates the degradation probability, triggering an early warning signal when the probability reaches a specific threshold. The metabolic model and prediction module periodically calls historical culture data from the data storage module, using gradient descent to update model parameters and improve prediction accuracy. It also supports manual import of new batches of data for model fine-tuning to adapt to changes in strain characteristics. The module transmits prediction results to the multi-objective optimization decision-making module in real time, while displaying prediction trend curves on the human-computer interaction interface to provide intuitive reference for decision-making.

[0168] The multi-objective optimization decision module sets the core training objectives for each generation based on the successive generations and prediction results, and calculates the optimal combination of environmental parameters through optimization algorithms, serving as the decision-making center of the system. In terms of target setting, the core objectives are dynamically matched based on generation number to ensure consistency with the strain's physiological characteristics. The core objective of generation G1 is to activate the cordycepin and adenosine synthesis pathways; generation G2 has dual objectives of maximizing total nucleoside content and total antioxidant activity; and generation G3 has dual objectives of maximizing biomass and minimizing degradation rate. Each generation has corresponding constraints. In terms of optimization algorithm selection, dedicated algorithms are used for different objectives to ensure optimization efficiency and accuracy. Generations G1 and G2 use a non-dominated sorting genetic algorithm, setting the population size and number of iterations. Decision variables include temperature, blue light intensity, and adenine replenishment rate, outputting a Pareto optimal parameter set. Generation G3 uses a model predictive control algorithm, setting the prediction time domain and control time domain. Decision variables include carbon source replenishment rate and temperature, with the objective function being to minimize carbon source deviation and degradation probability. The multi-objective optimization decision module selects the final parameter combination from the optimal parameter set according to industrial production priorities, encapsulates it as a parameter executor mapping instruction, and transmits it to the refined control strategy module.

[0169] Based on the optimization decision results, the control strategy module selects an appropriate control algorithm to generate precise execution instructions, ensuring that each actuator achieves closed-loop control according to the generational requirements. This is the core link in the implementation of the decision. In terms of control algorithm adaptation, dedicated algorithms are adopted for different control targets to improve control accuracy. The G1 generation adenine precursor supply adopts a PID control algorithm, and the peristaltic pump flow rate is output by the input adenine concentration deviation. The dissolved oxygen control of the G1 and G2 generations adopts a fuzzy PID control algorithm, and the stirrer speed is output by the input dissolved oxygen deviation and deviation change rate through preset fuzzy rules. The G3 generation carbon source gradient supply adopts a model predictive control algorithm, and the different carbon source supply rates of the nutrient pump array are output by the input deviation between the predicted value and the target value of the carbon source concentration, so as to realize the gradient switching of sucrose in the early stage of high sugar, the middle stage of mixing, and the late stage. The low temperature stress control of the G3 generation degradation intervention adopts a time-series control algorithm, which generates a step-type temperature command after receiving the degradation warning, and adjusts the dissolved oxygen to adapt to the low temperature metabolic demand. The fine control strategy module converts the result of the control algorithm calculation into a command format that the actuator can recognize, and sends it to the execution module through industrial Ethernet. At the same time, it receives the feedback data of the execution module in real time, calculates the control deviation and dynamically adjusts the command to form a closed loop control.

[0170] The execution module is the system's execution terminal, implementing control commands through various specialized actuators to alter the culture conditions within the culture chamber and ensure the achievement of core objectives. The actuator configuration covers all control dimensions to achieve full-scenario control. The temperature regulator consists of a heater, a semiconductor cooling array, and a circulating water bath; upon receiving temperature commands, it maintains the culture medium temperature accuracy through PID closed-loop control. The light regulator consists of a blue LED array, adjusting light intensity via specific signals and supporting adjustable light cycles. The ventilation regulator consists of a mass flow controller, an air pump, and a carbon dioxide cylinder, adjusting the ventilation rate and oxygen / carbon dioxide ratio according to commands. The nutrient supply pump array consists of multiple precision peristaltic pumps, each independently receiving supply rate commands for different nutrients. The stirrer or shaker is driven by a variable frequency motor, adjusting the culture medium mixing rate to meet dissolved oxygen control requirements upon receiving speed commands. The pH regulator consists of two peristaltic pumps, precisely adding acid or alkali solutions to maintain pH accuracy upon inputting pH deviation commands. Each actuator has a built-in status sensor that collects execution parameters in real time and feeds them back to the data acquisition and preprocessing module via an industrial bus, forming a command execution feedback closed loop.

[0171] The human-machine interface and alarm module serves as the interface between the system and operators, responsible for parameter setting status monitoring and anomaly warnings, ensuring system usability and security. In terms of interactive functions, this module provides a graphical user interface supporting multi-dimensional operation. Operators can manually input strain information, generation number, target parameters, and thresholds. Batch import and export of parameters is supported, and the interface displays core data in real-time, presented as dashboard trend curves and other graphs, with a refresh rate of once per second. It also supports retrieving historical culture records from the data storage module, viewing detailed data reports, and exporting to Excel. Regarding alarm functions, it implements multi-level anomaly warnings. When environmental parameters or physiological indicators exceed thresholds, audible and visual alarms are triggered, along with pop-up notifications on the interface. When actuators report abnormalities, the module automatically locates the faulty equipment and displays fault information, simultaneously sending alarm SMS messages to operators via Ethernet. When the G3 generation degradation probability reaches a specific threshold, a yellow warning is triggered, with the interface displaying a suggestion to activate low-temperature stress and providing a shortcut button for one-click control commands.

[0172] The data storage module is responsible for persistently storing all data in the system, supporting session interruption recovery and historical data analysis, and is the core carrier of the system's data assets. In terms of storage content, it covers all dimensions of system data, including basic data such as strain information, culture batch information, and equipment parameters; process data such as data after preprocessing of raw data from the detection module, generational results, prediction model output, control command records, and actuator feedback data; and result data such as final culture results, quality assessment reports, alarm logs, and data processing logs for each generation. The storage architecture adopts a dual backup mode of local and cloud to ensure data security. Local storage uses an industrial-grade SSD array to store nearly a year's worth of real-time data, supporting millisecond-level data reading to meet the fast access requirements of real-time control. Cloud storage periodically uploads historical data to a cloud database through a specific network, supporting long-term storage of massive amounts of data and providing a remote access interface for data aggregation and analysis across multiple plants. In terms of functional support, the data storage module provides data access and recovery capabilities, supporting multi-condition combined queries based on timestamp, batch number, and data type, with a response time of less than one second. When the system unexpectedly loses power and restarts, it automatically reads the last stored session data and restores the control state before the interruption to avoid interruption of the culture process. At the same time, it opens data interfaces to external analysis systems, supporting the optimization of model parameters based on historical data to continuously improve the system's control accuracy.

[0173] Exemplary device:

[0174] The Cordyceps sinensis biomimetic cultivation chamber control device provided in this application includes a memory, a main controller, and a computer program stored in the memory and capable of running on the main controller. When the main controller executes the program, it implements the Cordyceps sinensis biomimetic cultivation chamber control method described in any of the above embodiments.

[0175] The cultivation chamber control device may also include a memory; the main controller in the device may be one or more, with one main controller being used as an example in the figure; the memory is used to store one or more programs; the one or more programs are executed by the one or more main controllers, so that the one or more main controllers implement the Cordyceps sinensis biomimetic cultivation chamber control method as described in any of the above embodiments.

[0176] The cultivation chamber control equipment also includes: a sensing input module and an execution output module.

[0177] The main controller, memory, sensing input module, and execution output module in the cultivation chamber control equipment can be connected via industrial bus or Ethernet. The diagram shows an example of connection via industrial bus.

[0178] The sensing input module can receive environmental parameters and physiological index data collected by the sensor array, as well as control commands input by the operator, and generate signal inputs related to the operating status and parameter adjustment of the cultivation chamber. The execution output module may include various control actuators (heaters, coolers, peristaltic pumps, etc.), status indicator lights, and audible and visual alarm devices.

[0179] The memory, as a computer-readable storage medium, can be configured to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the Cordyceps sinensis biomimetic cultivation chamber control method described in the embodiments of this application. The memory may include a program storage area and a data storage area. The program storage area may store the operating system, cultivation strategy algorithms for each generation, and sensor calibration programs; the data storage area may store time-series data of environmental parameters generated by the operation of the cultivation chamber, records of strain physiological indicators, actuator action logs, historical cultivation batch archives, etc. Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as at least one solid-state drive or flash memory device. In some instances, the memory may further include memory remotely configured relative to the main controller, and these remote memories can be connected to the cultivation chamber control device via industrial Ethernet. Examples of the aforementioned networks include, but are not limited to, local area networks, industrial IoT, mobile communication networks, and combinations thereof.

[0180] Various implementations of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), systems-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable controller, which may be a dedicated or general-purpose programmable controller, capable of receiving data and instructions from a storage system, a sensing input module, and transmitting data and instructions to the storage system and an execution output module.

[0181] Based on the above embodiments, this embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by the main controller, implements the Cordyceps sinensis biomimetic cultivation chamber control method in any of the above embodiments of the present invention.

[0182] Computer programs used to implement the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the controller, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a local controller, partially locally, as a standalone software package partially locally and partially on a remote server, or entirely on a remote server.

[0183] In the context of this invention, a computer-readable storage medium stores computer instructions that, when executed by a controller, implement the Cordyceps sinensis biomimetic cultivation chamber control method provided by this invention. The computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. The computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0184] To provide interaction with the operator, the systems and techniques described herein can be implemented on a control device having: a display device for displaying operating status (e.g., a Liquid Crystal Display, LCD monitor); and an operation panel, touchscreen, and control buttons through which the operator can provide input to the control device. Other types of devices can also be used to provide interaction with the operator; for example, feedback provided to the operator can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the operator can be received in any form (including physical button input, touch input, and voice input).

[0185] The systems and technologies described herein can be implemented in computing systems that include back-end components (e.g., as data servers), or middleware components (e.g., application servers), or front-end components (e.g., local control cabinets with graphical user interfaces through which operators can interact with implementations of the systems and technologies described herein), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., industrial Ethernet, wireless sensor networks). Examples of communication networks include local area networks (LANs), industrial internet of things (IIoT), and the internet.

[0186] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, used to centrally manage the operational data of multiple cultivation chambers, providing remote monitoring, strategy optimization, and historical data analysis functions, thus overcoming the shortcomings of traditional single-machine control, such as data silos and low management efficiency.

[0187] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0188] Example 1:

[0189] Example 1 describes the specific process and results of subculturing G1, G2, and G3 generation Cordyceps sinensis strains using a Cordyceps sinensis culture control method and system of the present invention.

[0190] Strains preparation: Cordyceps sinensis strain CS1007 was activated in PDA medium and then inoculated into a 10L glass bioreactor containing 5L of culture medium at a volume percentage of 10%.

[0191] The basic culture medium consisted of 30 g / L glucose, 5 g / L yeast extract, 5 g / L peptone, 1 g / L KH2PO4, and 0.5 g / L MgSO4·7H2O, with a pH of 6.0.

[0192] G1 generation culture, metabolic activation phase, culture cycle 10 days;

[0193] The core objective is to improve the utilization efficiency of cordycepin and adenosine precursor (adenine).

[0194] Precise adenine replenishment:

[0195] After 24 hours of cultivation, the adenine concentration was measured to be 0.05 g / L using online HPLC (LC-20A). The system then started the nutrient supply pump (flow rate accuracy ±0.001 mL / min) to pump in an adenine solution with a concentration of 100 g / L. The adenine concentration in the chamber was increased to 0.5 g / L over a period of 120 minutes. Over the next 9 days, the adenine concentration was measured every 2 hours. When the concentration was below 0.35 g / L, it was replenished at a rate of 0.025 mL / min. When it was above 0.55 g / L, the replenishment was stopped. The final average replenishment rate stabilized at 0.02 mL / min, and the adenine concentration fluctuated between 0.35 and 0.52 g / L throughout the entire process.

[0196] Dissolved oxygen and stirring synergistic control:

[0197] The target dissolved oxygen (DO) level was set at 60-80% saturation, and the stirring speed (range 80-150 rpm) was adjusted using a fuzzy PID algorithm. On day 3 of cultivation, when DO dropped to 58%, the stirring speed was increased from 100 rpm to 120 rpm, and DO recovered to 65% after 30 minutes. On day 7, when DO rose to 82%, the stirring speed was reduced to 95 rpm, and DO stabilized at 78% after 20 minutes. The average stirring speed throughout the process was 110 rpm, and the DO fluctuation range was ≤ ±3%.

[0198] Temperature, humidity and pH stability control:

[0199] The temperature was maintained at 22.0℃ using a high-precision temperature controller (accuracy ±0.05℃), with the highest daily temperature being 22.08℃ and the lowest being 21.92℃; the relative humidity was controlled at 80%±2% using a humidity regulator (accuracy ±1%RH), with an actual monitoring range of 78.2-81.5%; and the pH was maintained at 6.0±0.05 using an automatic titration pump (titration accuracy ±0.01mL), with daily fluctuations ≤±0.03.

[0200] G2 generation cultivation, quality optimization stage, cultivation cycle 12 days;

[0201] Take dried G1 generation mycelia and reconstitute them with sterile culture medium to a concentration of 1×10⁻⁶. 8 CFU / mL, inoculated at a 10% (v / v) inoculation rate (500mL) into a new 10L culture chamber, the system identified it as generation G2, the core objective of which is to increase total nucleoside content and total antioxidant activity.

[0202] Mycelial components were analyzed every 4 hours using an online near-infrared spectrometer (model NIR-6500) to predict total nucleoside content and total antioxidant activity in real time (DPPH method). On day 5 of cultivation, the spectroscopic prediction showed a total nucleoside content of 3.8 mg / g and a total antioxidant activity of 61.2%. The optimal parameters calculated by the NSGA-II algorithm were: temperature 20.5℃ and blue light intensity 280 μmol·m². -2·s -1 (LED peak wavelength 450nm, 6 LEDs evenly distributed on the inner wall of the cultivation chamber), blue light cycle 16h / day (8:00-24:00 illumination); on the 7th day after adjustment, the predicted value rose to 4.9mg / g, 64.8%, and the parameters were maintained throughout the entire process until the end of the cultivation.

[0203] The target DO value is 70% ± 3%. By adjusting the ventilation rate (range 0.8-1.2 vvm) and stirring speed (100-130 rpm) in a coordinated manner, the actual DO range is 67.5-72.8%. The ventilation rate is stabilized at 1.0 vvm (i.e., 10 L of sterile air is introduced per minute), and the gas flow meter monitoring error is ≤ ± 0.05 vvm.

[0204] The target pH value is 6.5±0.08, which is adjusted by an automatic titration pump (the titrant is 1 mol / L HCl or NaOH). The actual monitoring range is 6.43-6.56, with daily fluctuations ≤±0.04.

[0205] G3 generation culture, biomass maximization combined with degradation delay phase, culture cycle 20 days;

[0206] Reconstitute G2 generation mycelium to 1×10 8 CFU / mL, inoculated into a new culture chamber at a 10% (v / v) inoculation rate, the system identified it as generation G3, the core objective of which is to increase biomass and delay mycelial degradation.

[0207] During the initial cultivation phase (days 0-6), the glucose concentration was maintained at 45 g / L, and the carbon source concentration was monitored daily. On day 6, the concentration was 43.2 g / L. On day 7, the biomass detector (model BM-800) showed that the mycelial growth rate decreased from 0.8 g / (L·d) to 0.5 g / (L·d), and the carbon source concentration decreased to 15 g / L. The MPC algorithm was then activated to initiate mixed feeding: 1.4 L of 500 g / L glucose solution and 0.35 L of 500 g / L maltose solution were pumped in, bringing the total carbon source concentration to 35 g / L (20% maltose). On day 14, the machine vision system (resolution 2048×1536) detected that the mycelial branching point density decreased from 32 / mm². 2 Reduced to 27.2 per mm 2 When the vacuolization level reached 25%, degradation intervention was initiated: 1.25L of 500g / L sucrose solution and 0.5L of 500g / L glucose solution were pumped in to maintain a total carbon source of 25g / L (sucrose accounting for 50%). Subsequently, the rate of decrease in branch point density slowed to 0.2 branches / (mm²). 2 ·d).

[0208] Low-temperature stress was initiated on day 14, with the temperature decreasing from 24.0℃ to 19.0℃ at a rate of 1℃ / h and maintained for 48 hours (12:00 on day 14 to 12:00 on day 16), and then restored to 24.0℃ at a rate of 0.5℃ / h. After stress, the degree of mycelial vaping decreased from 25% to 18%, and the oxidative damage index (MDA content) decreased from 12.5 nmol / g to 8.3 nmol / g.

[0209] The target DO value is 85% ± 4%, which is achieved by increasing the ventilation rate to 1.5 vvm and the stirring speed to 130 rpm. The actual DO range is 81.2-88.5%. 0.2% (v / v) oxygen is added during ventilation to ensure high DO requirements.

[0210] The target pH value is 5.5±0.1, and the actual monitoring range is 5.42-5.58. The pH is precisely adjusted by a titration pump to avoid the acidic environment inhibiting mycelial growth.

[0211] Comparative example:

[0212] To further highlight the advantages of the control method and system of this invention, this comparative example adopts a traditional liquid deep fermentation culture method for Cordyceps sinensis based on fixed parameter settings, without introducing generational fine-tuning, online monitoring, predictive optimization and degradation early warning strategies.

[0213] Culture method: The strain, inoculum size, and basal culture medium are the same as in the example.

[0214] Culture parameters:

[0215] Generations G1, G2, and G3 all used fixed parameters: temperature 23.0℃, relative humidity 85%, pH 6.0, dissolved oxygen 70% saturation, stirring speed 120 rpm, aeration rate 1.0 vvm, and light intensity 200 μmol·m². -2 ·s -1 (White light), illumination cycle 12 hours / day.

[0216] Nutritional supplementation: Add glucose 50g / L once at the beginning of culture, and do not provide gradient supplementation of adenine or carbon source during culture.

[0217] No online monitoring or feedback: During cultivation, there was no online monitoring of metabolite precursors, spectral analysis, or machine vision; only offline sampling and analysis were performed. No intervention measures were implemented based on signs of hyphal degeneration.

[0218] Table 1 Comparison of key indicators between Example 1 and the comparative example.

[0219]

[0220] The G1 generation example outperformed the comparative example in all indicators. The core reason is that the example precisely matched the physiological characteristics of the G1 generation mycelial metabolic pathways, which were not fully activated. The example provided adenine through dynamic supplementation, offering a continuous precursor for the synthesis of cordycepin and adenosine. At the same time, it used a fuzzy PID algorithm to stabilize dissolved oxygen, ensuring efficient aerobic metabolism. In contrast, the comparative example did not supplement adenine, and the fixed stirring speed resulted in large fluctuations in dissolved oxygen, lacking both metabolic raw materials and a stable metabolic environment, ultimately inhibiting mycelial growth and the synthesis of active ingredients.

[0221] The G2 generation example exhibits superior total nucleoside content, antioxidant activity, and mycelial dry weight due to its ability to capture the peak physiological activity of G2 generation mycelia. This example utilizes online spectral monitoring to track component changes in real time, and combines algorithms to optimize temperature and blue light parameters, precisely activating pathways related to the synthesis of active substances. In contrast, the comparative example uses fixed white light and temperature, which cannot meet the peak requirements for active substance synthesis in the G2 generation, and unstable dissolved oxygen affects nutrient conversion, resulting in lagging performance in key indicators.

[0222] The advantages of the G3 generation example in terms of biomass and resistance to degradation are attributed to targeted carbon source regulation and degradation intervention. Based on the changes in the G3 generation growth rate, the example adjusted the type and concentration of carbon source in stages to meet the needs of different growth stages. Furthermore, when signs of mycelial degradation appeared, low-temperature stress was initiated to mitigate metabolic damage. In contrast, the comparative example added carbon source all at once, resulting in nutrient deficiency and no degradation intervention in the later stages. This led to mycelial aging and autolysis, limited biomass accumulation, and severe degradation.

[0223] Overall, the core advantage of the embodiment is generational targeted regulation, which designs regulation strategies based on the unique physiological characteristics of each generation of mycelium and combines real-time monitoring to dynamically optimize conditions; the comparison embodiment ignores generational differences and uses fixed parameters for extensive regulation, which can never match the real needs of mycelium, so the key indicators of each generation are significantly lower than those of the embodiment.

[0224] Based on the preferred embodiments of the present invention described above, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A method for controlling the cultivation of Cordyceps sinensis, characterized in that, Including the following steps: S1. Identify the generation number of the Cordyceps sinensis strain in the current culture batch; S2. Real-time monitoring of environmental parameters and bacterial physiological indicators within the cultivation chamber; S3. Based on the identified subgenerations, set the corresponding core culture objectives; among them, the G1 generation aims to activate the synthesis pathways of cordycepin and adenosine, the G2 generation aims to maximize the total nucleoside content and total antioxidant activity, and the G3 generation aims to maximize mycelial biomass and delay strain degradation. S4. Based on the set core objectives and real-time monitoring data, dynamically adjust the cultivation conditions.

2. The method for controlling the cultivation of Cordyceps sinensis according to claim 1, characterized in that: Identifying the number of generations involves: acquiring microscopic morphological images of hyphae, extracting features such as hyphal branching density, hyphal diameter uniformity, cell vacuolar ratio, and hyphal integrity, and comparing them with a pre-stored morphological feature-generation matching model to determine the number of generations.

3. The method for controlling the cultivation of Cordyceps sinensis according to claim 1, characterized in that: In step S2, the environmental parameters include: temperature, humidity, light intensity and period, carbon dioxide concentration, oxygen concentration, dissolved oxygen concentration and pH value; the bacterial physiological indicators include: biomass, concentration of metabolite precursors, and predicted total nucleoside content and total antioxidant activity.

4. The method for controlling the cultivation of Cordyceps sinensis according to claim 1, characterized in that: When the identified subculture number is G1, adenine concentration is maintained within the range of 0.3 g / L to 0.6 g / L by dynamically supplementing adenine based on the adenine concentration fed back by the metabolite precursor detector; real-time dissolved oxygen data is received, and the speed of the mechanical stirrer or shaker is dynamically adjusted through a fuzzy PID control algorithm to maintain the dissolved oxygen concentration in the culture medium at 60% to 80% saturation; the culture temperature is maintained at 22℃±0.5℃, the relative humidity at 80%±5%, and the pH value at 6.0±0.

2.

5. The method for controlling the cultivation of Cordyceps sinensis according to claim 1, characterized in that: When the identified generation number is G2, the predicted total nucleoside content and total antioxidant activity data are received. A multi-objective optimization algorithm based on the non-dominated sorting genetic algorithm NSGA-II is run to maximize the total nucleoside content and total antioxidant activity. The optimal combination of culture chamber temperature, blue light intensity, and blue light cycle parameters is calculated. Based on the optimal parameter combination calculated by the multi-objective optimization algorithm, the culture temperature is adjusted to the range of 18℃ to 25℃, and the blue light intensity is adjusted to 100-400 μmol·m⁻¹. -2 ·s -1 Within the specified range, the blue light cycle was set to 12 to 18 hours per day; the dissolved oxygen concentration was maintained at 70% ± 5% saturation, and the pH value was maintained at 6.5 ± 0.2 in conjunction with a pH detector and an acid-base titration system.

6. The method for controlling the cultivation of Cordyceps sinensis according to claim 1, characterized in that: When the identified subculture number is G3, a gradient carbon source supplementation strategy is implemented based on real-time data of mycelial growth rate and carbon source concentration: In the early stage of cultivation, the glucose concentration is maintained at 40 g / L to 50 g / L; in the middle stage, a mixed glucose and maltose supplementation is used, maintaining the total carbon source concentration at 30 g / L to 40 g / L, with the maltose percentage gradually increasing to 30%; when signs of mycelial degeneration are detected, sucrose supplementation is switched, maintaining the total carbon source concentration at 20 g / L to 30 g / L; when signs of mycelial degeneration are detected, a short-term low-temperature stress strategy is triggered, lowering the cultivation temperature from 24°C to 18°C ​​to 20°C within 24 to 48 hours using a temperature regulator and maintaining this temperature for 24 to 48 hours, then restoring it to 24°C; by increasing the aeration rate and stirring speed, the dissolved oxygen concentration is maintained at 85% ± 5% saturation, and the pH value is maintained at 5.5 ± 0.2 using a pH regulator.

7. The method for controlling the cultivation of Cordyceps sinensis according to claim 1, characterized in that: In step S4, dynamically adjusting the cultivation conditions includes: generating control commands based on the core objective using at least one of PID control algorithm, fuzzy PID control algorithm, model predictive control algorithm, or multi-objective optimization algorithm, and adjusting the cultivation conditions through the execution module.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute and implement the Cordyceps sinensis cultivation control method according to any one of claims 1-7.

9. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform a method for controlling the cultivation of Cordyceps sinensis according to any one of claims 1-7.

10. A Cordyceps sinensis cultivation control system, based on the Cordyceps sinensis cultivation control method according to any one of claims 1-7, characterized in that, include: The detection module is used to monitor environmental parameters and bacterial physiological indicators in the culture chamber in real time. The generation number identification module is used to identify the generation number of the Cordyceps sinensis strain in the current culture batch. The data acquisition and preprocessing module is used to acquire and preprocess the data from the detection module and the successive algebra identification module; Metabolic model and prediction module, used to predict mycelial growth status and metabolite content based on preprocessed data; The multi-objective optimization decision module is used to set core training objectives based on the number of generations and to calculate the optimal combination of environmental parameters through optimization algorithms. The control strategy module is used to generate control commands based on the output of the multi-objective optimization decision module; The execution module is used to dynamically adjust the culture conditions according to the control instructions; The human-computer interaction and alarm module is used for parameter setting, status monitoring and abnormal alarms; The data storage module is used to store all data throughout the system process.

Images