Biological fermentation preparation system and process of 5-hydroxytryptophan

Abstract

The invention provides a biological fermentation preparation system process of 5-hydroxytryptophan, the system comprises a fermentation culture medium preparation and inoculation module, a fermentation tank, a sensor data acquisition module, a data processing unit and a single-chip microcomputer, and the process comprises strain optimization, fermentation, fermentation completion and product extraction. Wherein during fermentation, an efficient monitoring and control system is established, the fermentation environment is accurately controlled, changes of various key parameter data in the fermentation process are obtained, all the key parameter data are gathered through a data acquisition module and transmitted to a single-chip microcomputer, and the single-chip microcomputer automatically adjusts the working state of corresponding equipment according to fermentation parameters set by an algorithm. Through combination of genetic engineering strain optimization, an intelligent control system, membrane filtration and a reverse osmosis purification technology, the yield and purity of 5-HTP are remarkably improved, the production cost is effectively reduced, and the production period is shortened. Compared with the traditional process, the method has obvious advantages in multiple aspects, and has wide industrial application potential.

Description

Technical Field

[0001] The invention relates to the technical field of amino acid preparation, and in particular to a biological fermentation preparation system and process for 5-hydroxytryptophan. Background Art

[0002] 5-Hydroxytryptophan (5-HTP) is a metabolite of tryptophan and is widely used in the fields of medicine, health products and cosmetics. As a natural neurotransmitter precursor, 5-HTP has the effect of enhancing the level of serotonin in the brain, so it has important application prospects in the treatment of diseases such as depression, anxiety and insomnia. The traditional preparation method of 5-HTP is usually through chemical synthesis or extraction from natural plants, but the chemical synthesis method requires the use of a large amount of harmful chemical reagents and high-energy consumption reaction conditions, the cost of the preparation process is high, and it may cause environmental pollution. The process of extracting 5-HTP from natural plants is inefficient and is limited by the growth environment and seasonality of the plants. Therefore, the extraction amount is insufficient and cannot meet the needs of large-scale production. Whether it is chemical synthesis or natural extraction, the purification and separation process of 5-HTP is difficult, resulting in low product purity and high technical requirements.

[0003] With the development of biotechnology, the production of 5-HTP using microbial fermentation technology has become a potential environmentally friendly, efficient and sustainable preparation method. In recent years, the research on the production of 5-HTP using microbial strains such as yeast and Escherichia coli has gradually increased, but the existing biological fermentation process still faces some technical difficulties, especially the low metabolic efficiency of microorganisms and the difficulty of controlling the fermentation process. The microbial strains commonly used in 5-HTP production have low metabolic efficiency, resulting in low yields, and the by-products produced during the fermentation process have an inhibitory effect on the production of 5-HTP. In addition, the existing biological fermentation process is prone to instability in temperature, pH value and dissolved oxygen during the fermentation process, resulting in fluctuations in bacterial growth and metabolite synthesis, thereby affecting the yield and quality of 5-HTP. Summary of the invention

[0004] The present invention proposes an optimized 5-hydroxytryptophan biological fermentation preparation system and process, aiming to solve the problems of low yield, high cost, long cycle and the like in the prior art.

[0005] To achieve the above object, the present invention proposes the following technical solution: a biological fermentation preparation process of 5-hydroxytryptophan, comprising the following steps:

[0006] Step 1, strain optimization;

[0007] Step 1.1, selecting and transforming strains, selecting microbial strains with high metabolic capacity as the basic strains for 5-HTP production, using the genetic engineering technology CRISPR-Cas9 to perform targeted transformation on the selected strains, introducing key enzyme genes of tryptophan hydroxylase (TPH) and aromatic amino acid decarboxylase (AADC), and enhancing the metabolic pathway of tryptophan conversion to 5-HTP, so as to obtain the transformed strains;

[0008] Step 1.2, in the modified strain, adjusting the expression level of the key enzyme by optimizing the gene expression system;

[0009] Step 1.3, modifying the tolerance of the inhibitory substances produced during the fermentation process so that the strain can withstand higher concentrations of substrates and products;

[0010] Step 2, fermentation;

[0011] Step 2.1, medium preparation and inoculation, prepare the fermentation medium, the medium includes substrate, nitrogen source and carbon source to meet the growth requirements of the bacteria, and before inoculation, pre-culture the selected modified strain in step 1 to ensure its activity and metabolic capacity, and prepare to inoculate it into the fermentation tank;

[0012] Step 2.2, establish an efficient monitoring and control system to accurately control the fermentation environment. Arrange temperature sensors, pH sensors, dissolved oxygen sensors and nutrient concentration monitoring equipment in the fermenter to obtain changes in key parameter data during the fermentation process. All key parameter data are summarized and transmitted to the single-chip microcomputer through the data acquisition module. The single-chip microcomputer automatically adjusts the working status of the corresponding equipment according to the fermentation parameters set by the algorithm;

[0013] Step 3, fermentation end and product extraction, the fermentation completion time is determined by monitoring and controlling the system, and whether the fermentation is completed is determined based on the cumulative amount of 5-HTP and the bacterial growth curve. At the end of the fermentation, oxygen supply and nutrient addition are stopped to maintain a stable fermentation environment;

[0014] 5-HTP is extracted from the fermentation broth using liquid-liquid extraction or centrifugal separation, and 5-HTP is further purified using membrane filtration and reverse osmosis to remove impurities and improve the purity of 5-HTP. The extracted and purified 5-HTP is then subjected to quality testing, including purity, content, and sterility testing.

[0015] Furthermore, in the present invention, in the step 1.1, the microbial strains are Escherichia coli BL21 and yeast CEN.PK 113-11C.

[0016] Furthermore, in the present invention, in step 1.2, the Gal4 / UAS system or the Tet-On / Tet-Off system is used to achieve precise regulation of specific enzyme genes, and an enhancer-controlled binary promoter system is adopted, combined with a heat-inducible promoter and a non-toxic activation system, to achieve timely regulation of key enzyme genes during the fermentation process.

[0017] Furthermore, in the present invention, in step 1.3, one or more of the ethanol-resistant ADH2 gene, the amino acid-resistant gene, and the acid-resistant gene are used for gene integration, and the amino acid-resistant gene and the acid-resistant gene are drug-resistant genes.

[0018] Furthermore, in the present invention, the temperature sensor in step 2.2 is used to monitor the temperature change of the fermentation liquid in real time, the pH sensor monitors the pH of the fermentation liquid and automatically adjusts it through a feedback system, the dissolved oxygen sensor monitors the concentration of dissolved oxygen during the fermentation process and helps adjust the oxygen supply system, and the nutrient concentration monitoring equipment monitors the concentration of the substrate, nitrogen source, and carbon source through online chromatography or spectroscopy.

[0019] Furthermore, in the present invention, the single chip microcomputer automatically adjusts the working state of the corresponding equipment according to the fermentation parameters set by the algorithm, including the following process:

[0020] According to the optimal growth temperature range of the selected strain, a precise temperature control system is used to ensure temperature stability during the fermentation process. The temperature control system maintains a constant temperature through a heat exchange system or a cooling system;

[0021] Set the ideal pH range during the fermentation process, and measure the pH value in the fermentation broth in real time through automated pH monitoring equipment. If it deviates from the set range, adjust it by adding acid or alkaline solution to ensure the activity of enzymes and the growth conditions of microorganisms during the fermentation process;

[0022] Use a dissolved oxygen detector to monitor the dissolved oxygen content in the fermentation liquid and adjust the oxygen supply according to demand. Use an oxygen ventilation system to adjust the dissolved oxygen concentration in the fermentation liquid;

[0023] Monitor and adjust the concentration of substrate nitrogen source, carbon source and trace elements during the fermentation process. When the system detects that the concentration of substrate or nutrients is lower than the set value, it automatically starts the supplement mechanism to add the required nutrients.

[0024] Furthermore, in the present invention, the single chip microcomputer automatically adjusts the working state of the corresponding equipment according to the fermentation parameters set by the algorithm. The specific process is as follows:

[0025] Step 1: Data collection and preprocessing, denoising, use filters to remove noise from sensor data. Assuming the sensor signal is s(t) and the noise is n(t), the true value of the signal can be estimated through the filter:

[0026] in, is the filtered signal, K(t)K is the Kalman gain, and y(t) is the observed value;

[0027] Standardization was used to normalize the data to make them in the same range, and when some data were missing, the Lagrange interpolation method was used to fill in the missing values;

[0028] Step 2: Feature extraction and selection: feature extraction of sensor data to enable the machine learning model to process effectively. Feature extraction includes time series features and environmental change features. Time series features include the time trend, periodicity, average value, and standard deviation of temperature, pH value, and dissolved oxygen. Environmental change features include external factors that affect fermentation in combination with external environmental conditions, such as external temperature, humidity, and air pressure. Then, the fermentation stage is marked: according to the different stages of the fermentation process, specific stage features are marked. Through principal component analysis (PCA) or correlation analysis, the features that have the greatest impact on 5-HTP production are selected to simplify the model and improve computational efficiency.

[0029] Step 3: Establish a machine learning model. The model is a random forest model. By constructing multiple decision trees and predicting the production of 5-HTP through a voting mechanism, each tree only uses part of the features during training and improves the prediction accuracy through weighted voting. The formula is as follows:

[0030] Among them, Y pred is the predicted value, h i (X) is the output of the i-th tree, X is the input feature, and N is the number of trees;

[0031] The gradient extreme gradient boosting was introduced to improve the prediction accuracy. The gradient extreme gradient was used to perform regression prediction on 5-HTP production. The formula is: Among them, α i is the weight of each tree, f i (X) is the output of the i-th tree, and N is the total number of trees;

[0032] The accuracy of the model was evaluated using the validation set, and the evaluation indicators included mean square error (MSE) and coefficient of determination (R 2 ), R 2 Indicators are used to measure the model fit:

[0033] Among them, Y true，i is the actual value, Y pred，i is the predicted value, is the mean of the actual values, and k-fold cross validation is used to further optimize the model to ensure the generalization ability of the model;

[0034] Step 4: Data fusion and environmental change adjustment. For each sensor input, different weights are assigned based on its reliability. Suppose there are multiple sensor data X1, X2, ..., X n , the fused output is:

[0035] Among them, w i is the weight of each sensor, which is adjusted according to the accuracy and historical performance of the sensor, and the sum of the weights is 1, that is:

[0036] Based on the external environmental data, an environmental change prediction model is established. Combined with the internal factors of the fermentation process, the changes that occur during the fermentation process are predicted. Multiple linear regression or deep neural network (DNN) models are used to achieve the following:

[0037] Y = f(X internal , X environment ), where X internal It is an internal factor, which includes temperature, dissolved oxygen, pH value, etc. environment is the environmental factor, Y is the final 5-HTP production prediction;

[0038] Step 5: Real-time adjustment of the fermentation process. Connect the prediction model with the automatic adjustment system. According to the output of the model, the output is the expected 5-HTP production and bacterial growth status, to adjust the fermentation process in real time, and use PID control algorithm or fuzzy control algorithm to adjust the temperature, pH, dissolved oxygen and nutrient supply.

[0039] Furthermore, in the present invention, when establishing an environmental change prediction model, in addition to using a deep neural network (DNN), a convolutional neural network (CNN) is added to extract spatial features of environmental data, thereby predicting the impact of external factors on the fermentation process. The CNN model formula is: Among them, X i is the input data, W i is the weight, f(X i ) is the convolution operation.

[0040] Furthermore, in the present invention, the specific process of step 5 is as follows:

[0041] Obtain real-time monitoring data, including temperature, pH, dissolved oxygen, and nutrient concentration. For each control parameter, calculate the deviation between the current value and the target value:

[0042] e(t)=setpoint-measured(t);

[0043] Where: e(t) is the error value at the current moment, setpoint is the set target value, and measured(t) is the real-time measurement value;

[0044] The PID control algorithm is used for regulation. The PID control algorithm calculates a control output based on the error value to ensure that the system maintains the target state. The PID algorithm is based on three terms: proportional (P), integral (I) and differential (D), as shown in the following formula:

[0045]

[0046] Where: u(t) is the control quantity, K p is the proportional gain, K i is the integral gain, K d is the differential gain, e(t) is the error value at the current moment;

[0047] Implement regulation, the control system adjusts the corresponding equipment according to the control quantity u(t) calculated by PID, adjusts the temperature of the fermentation tank through the heat exchange system or the cooling system, adjusts the pH value of the fermentation liquid by adding acid or alkali, and changes the dissolved oxygen content by adjusting the ventilation volume or using the gas distribution system;

[0048] Monitor and adjust in real time. The real-time monitoring system continuously feeds back the adjusted control effect and further adjusts the control parameters when necessary. According to the difference between the model prediction and the actual data, the PID parameters are automatically adjusted to optimize the control effect.

[0049] Gradually optimize the control parameters by adjusting the PID parameter K p ,K i ,K d Continuous adjustment and optimization make the control system more accurate and stable.

[0050] Further, in the present invention, a fermentation medium preparation and inoculation module, a fermentation tank, a sensor data acquisition module, a data processing unit and a single-chip microcomputer are included, wherein the fermentation tank is used for biological fermentation of 5-hydroxytryptophan, and the sensors include a temperature sensor, a pH sensor, a dissolved oxygen sensor and a nutrient concentration monitoring device;

[0051] The single-chip microcomputer is used to coordinate and control the work of each module, receive and process sensor data, and automatically adjust fermentation parameters according to the set algorithm and control model to implement PID or fuzzy control algorithm to adjust the fermentation environment in real time. All sensor data are collected in the single-chip microcomputer, and the real-time data is transmitted to the central control system through the communication interface. The single-chip microcomputer makes adjustments based on the data, and the actuator receives the control signal and performs real-time adjustment;

[0052] The fermentation medium preparation and inoculation module includes an automatic liquid adding mechanism and an inoculation component. The automatic liquid adding mechanism is used to automatically mix the substrate, nitrogen source, and carbon source components in proportion and add them to the fermentation tank. The inoculation component is used to control the inoculation amount of the strain and ensure the sterility of the inoculation. The automatic liquid adding mechanism and the inoculation component are connected to and controlled by the single chip microcomputer.

[0053] The sensor data acquisition module includes temperature, pH, dissolved oxygen, and nutrient concentration sensors. All sensors are connected to the data acquisition module. The data processing unit filters, standardizes, and interpolates the collected data. The data acquisition module transmits the processed data to the single-chip microcomputer or control computer for model prediction and fermentation environment adjustment.

[0054] Beneficial effects: The technical solution of this application has the following technical effects:

[0055] 1. The present invention can increase the yield of 5-HTP, optimize the strain through genetic engineering, and precisely regulate the expression of key enzyme genes to enhance the metabolic pathway of tryptophan conversion to 5-HTP, thereby significantly improving the production efficiency of 5-HTP. The CRISPR-Cas9 technology is used to carry out targeted transformation of the strain so that it can withstand high concentrations of substrates and products, further optimizing the fermentation process.

[0056] 2. The fermentation process is more intelligent and adjustable. The present invention uses a highly integrated sensor network, data acquisition module and single-chip system to achieve accurate monitoring of temperature, pH, dissolved oxygen and nutrient concentration during the fermentation process. Through real-time data feedback and PID or fuzzy control algorithms, the fermentation environment can be automatically adjusted to ensure that all key parameters remain within the optimal range, thereby improving production efficiency and fermentation stability.

[0057] 3. The optimized fermentation process, combined with a variety of machine learning models, including random forests and deep neural networks, can identify and respond to potential environmental changes in advance by analyzing and predicting the multi-dimensional characteristics of the fermentation process. Further use of data fusion technology to ensure the reliability of sensor data, thereby achieving more accurate and real-time process adjustments and reducing human intervention and errors. It can reduce production costs and improve purity. By accurately controlling various parameters in the fermentation process, the formation of by-products caused by changes in temperature, pH, etc. is reduced, and the waste of energy and raw materials in the production process is reduced. At the same time, the extraction and purification of 5-HTP uses membrane filtration and reverse osmosis technology to further improve the purity of the final product and meet high standards of quality requirements. With an automated and standardized production process, through automated fermentation medium preparation and inoculation modules, sensor data acquisition modules and automatic control systems, the entire fermentation process can be stably operated without relying on a large number of manual operations, reducing the risks brought by manual operations, and ensuring the repeatability and standardization of each production.

[0058] 4. Application of high-tolerance strains: The present invention enhances the tolerance of strains through genetic modification, especially the tolerance to intermediates such as 5-HTP and tryptophan, so that the strains can still maintain efficient metabolic capacity in high-concentration substrate and product environments, thereby improving production stability and yield. It can enhance environmental adaptability. By integrating external environmental data (such as temperature, humidity, air pressure, etc.) and combining CNN models for spatial feature extraction, the present invention further enhances the adaptability of the fermentation system to environmental changes. The prediction and regulation of external environmental factors help reduce the interference of the external environment on the fermentation process and ensure the smooth progress of the fermentation process.

[0059] It should be appreciated that all combinations of the foregoing concepts, as well as additional concepts described in greater detail below, may be considered to be part of the inventive subject matter of the present disclosure, provided such concepts are not mutually inconsistent.

[0060] The foregoing and other aspects, embodiments and features of the present invention can be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the present invention, such as the features and / or beneficial effects of the exemplary embodiments, will be apparent from the following description or learned from the practice of the specific embodiments according to the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown in various figures may be represented by the same reference numeral. For clarity, not every component is labeled in every figure. Embodiments of various aspects of the present invention will now be described by way of example and with reference to the accompanying drawings, in which:

[0062] Figure 1 Schematic diagram of the system of the present invention. DETAILED DESCRIPTION

[0063] In order to better understand the technical content of the present invention, specific embodiments are cited and described as follows in conjunction with the accompanying drawings. Various aspects of the present invention are described in this disclosure with reference to the accompanying drawings, in which many illustrative embodiments are shown. The embodiments of the present disclosure are not necessarily defined to include all aspects of the present invention. It should be understood that the various concepts and embodiments introduced above, as well as those described in more detail below, can be implemented in any of many ways, because the concepts and embodiments disclosed in the present invention are not limited to any implementation. In addition, some aspects disclosed in the present invention can be used alone, or in any appropriate combination with other aspects disclosed in the present invention.

[0064] Example 1

[0065] A biological fermentation preparation process for 5-hydroxytryptophan comprises the following steps:

[0066] Step 1, strain optimization;

[0067] Step 1.1, select and transform strains, select microbial strains with high metabolic capacity as the basic strains for 5-HTP production, the microbial strains are Escherichia coli BL21 and yeast CEN.PK 113-11C, use the genetic engineering technology CRISPR-Cas9 to carry out directed transformation of the selected strains, introduce key enzyme genes of tryptophan hydroxylase (TPH) and aromatic amino acid decarboxylase (AADC), and enhance its metabolic pathway for converting tryptophan to 5-HTP to obtain a modified strain.

[0068] Specifically, the strains were cultured, BL21 and CEN.PK 113-11C strains were inoculated in liquid culture medium, and standard LB medium (E. coli) and YPD medium (yeast) were used for preliminary culture. During genome extraction and sequencing, the genomes of BL21 and CEN.PK 113-11C were extracted, and the genomes of the strains were comprehensively analyzed by high-throughput sequencing technology to ensure their stability and provide a basis for subsequent genetic modification. Basic metabolic capacity detection, using an HPLC analyzer to determine the metabolic characteristics of BL21 and CEN.PK 113-11C strains at different substrate (such as tryptophan) concentrations. Their growth curves at different tryptophan concentrations were recorded, and their metabolic capacity was determined.

[0069] When the tryptophan concentration was 1 mM, the growth rate of CEN.PK 113-11C was 0.25 h-1, while the growth rate of BL21 was 0.18 h-1, indicating that CEN.PK 113-11C performed better at this concentration.

[0070] The genetic modification is as follows: construct an expression vector, construct TPH and AADC gene expression vectors in Escherichia coli or yeast, and use pET series expression vectors (Escherichia coli) or pGPD series expression vectors (yeast). The constructed expression vectors were transformed into BL21 and CEN.PK 113-11C, respectively, and the transformants were obtained by antibiotic screening. qPCR analysis: Real-time quantitative PCR (qPCR) analysis of the expression of TPH and AADC genes was performed using specific primers to ensure the successful expression of the target genes. The total protein of the modified strain was extracted, and Western Blot detection was performed using anti-TPH and anti-AADC antibodies to ensure the expression of the enzyme. The metabolites of the modified strain were analyzed by HPLC or LC-MS / MS to verify the generation of 5-HTP. The 5-HTP production of BL21 and CEN.PK113-11C strains was recorded and compared with the unmodified strains.

[0071] The 5-HTP production of the genetically modified BL21 strain was 50% higher than that of the unmodified strain, and the 5-HTP production of CEN.PK113-11C was 60% higher.

[0072] Step 1.2, in the modified strain, the expression of key enzymes is regulated by optimizing the gene expression system, using the Gal4 / UAS system or the Tet-On / Tet-Off system to achieve precise regulation of specific enzyme genes, and using an enhancer-controlled binary promoter system, combined with a heat-inducible promoter and a non-toxic activation system, to achieve time-effective regulation of key enzyme genes during the fermentation process;

[0073] Specifically, a regulatory expression vector was constructed, and the Gal4 / UAS system (yeast) or Tet-On / Tet-Off system (E. coli) was added before the promoter of the TPH and AADC genes to ensure that the gene was expressed only when needed. Transformation and screening: The constructed vector was transformed into BL21 and CEN.PK 113-11C, and the transformant was obtained by antibiotic screening. In the gene expression induction experiment, the target gene was induced by different concentrations of tetracycline (Tet-On system) or induction temperature (Gal4 / UAS system). Western Blot and qPCR were used to analyze the gene expression after induction to ensure the successful operation of the expression system. Enzyme activity test: The enzyme activity of TPH and AADC was determined by colorimetry or fluorescence, and the effect of different expression systems on enzyme activity was analyzed.

[0074] After induction with the Tet-On system, the TPH activity of Escherichia coli BL21 increased by 45%, while the AADC activity of yeast CEN.PK 113-11C increased by 55% after induction with the Gal4 / UAS system.

[0075] The binary promoter system allows the gene to be expressed only at the desired time and under the required conditions, avoiding excessive or insufficient expression and optimizing the production of 5-HTP. It has higher enzyme activity and can precisely regulate the expression of TPH and AADC through an optimized expression system, thereby improving the efficiency of tryptophan conversion to 5-HTP.

[0076] Step 1.3, tolerance modification of inhibitory substances (such as tryptophan and 5-HTP) produced during the fermentation process is performed to enable the strain to withstand higher concentrations of substrates and products, and one or more of the ethanol-resistant ADH2 gene, amino acid-resistant gene, and acid-resistant gene are used for gene integration, and the amino acid-resistant gene and acid-resistant gene are drug-resistant genes;

[0077] Specifically, first carry out tolerance gene screening, select suitable gene ADH2 from the gene library of known acid resistance or substrate concentration tolerance, and introduce it into BL21 and CEN.PK 113-11C by gene cloning technology. Then carry out gene introduction and screening, use standard transformation method (such as electric shock transformation) to introduce tolerance gene into target strain, and obtain transformant by antibiotic screening. Finally, carry out tolerance test, culture the modified strain in the culture medium containing high concentration of tryptophan and 5-HTP, and observe its growth and metabolic capacity. By measuring the production rate of bacterial biomass, tryptophan consumption rate and 5-HTP, the effect of tolerance transformation is evaluated.

[0078] High-density fermentation experiments were conducted in fermenters, and an automated culture system was used to monitor the growth of strains in high-concentration substrates (such as tryptophan) and products (such as 5-HTP). The modified BL21 strain grew well in a high-concentration 5-HTP environment, and the 5-HTP production increased by 70%, while the 5-HTP production of the unmodified strain only increased by 30%.

[0079] The modified strain can grow stably in an environment with high concentrations of tryptophan and 5-HTP, which increases the yield and production efficiency of the strain and significantly improves tolerance. The modified strain can withstand high concentrations of substrates and products under large-scale fermentation conditions, ensuring the stability and controllability of the fermentation process and adapting to large-scale production.

[0080] Step 2, fermentation;

[0081] Step 2.1, medium preparation and inoculation, prepare fermentation medium, the medium includes substrate, nitrogen source and carbon source to meet the growth requirements of bacteria, the substrate is tryptophan, the nitrogen source is amino acid, and the carbon source is glucose. Before inoculation, pre-culture the selected modified strain in step 1 to ensure its activity and metabolic capacity, and prepare to inoculate it into the fermentation tank;

[0082] Step 2.2, establish an efficient monitoring and control system to accurately control the fermentation environment. Arrange temperature sensors, pH sensors, dissolved oxygen sensors and nutrient concentration monitoring equipment in the fermenter to obtain changes in key parameter data during the fermentation process. All key parameter data are summarized and transmitted to the single-chip microcomputer through the data acquisition module. The single-chip microcomputer automatically adjusts the working status of the corresponding equipment according to the fermentation parameters set by the algorithm;

[0083] Step 3, fermentation end and product extraction, the fermentation completion time is determined by monitoring and controlling the system, and whether the fermentation is completed is determined based on the cumulative amount of 5-HTP and the bacterial growth curve. At the end of the fermentation, oxygen supply and nutrient addition are stopped to maintain a stable fermentation environment;

[0084] 5-HTP is extracted from the fermentation broth using liquid-liquid extraction or centrifugal separation, and 5-HTP is further purified using membrane filtration and reverse osmosis to remove impurities and improve the purity of 5-HTP. The extracted and purified 5-HTP is then subjected to quality testing, including purity, content, and sterility testing.

[0085] In this embodiment, the temperature sensor in step 2.2 is used to monitor the temperature change of the fermentation liquid in real time, the pH sensor monitors the pH of the fermentation liquid and automatically adjusts it through a feedback system, the dissolved oxygen sensor monitors the concentration of dissolved oxygen during the fermentation process and helps adjust the oxygen supply system, and the nutrient concentration monitoring equipment monitors the concentration of the substrate, nitrogen source, and carbon source through online chromatography or spectroscopy technology.

[0086] In this embodiment, the single chip microcomputer automatically adjusts the working state of the corresponding equipment according to the fermentation parameters set by the algorithm, including the following process:

[0087] According to the optimal growth temperature range of the selected strain, a precise temperature control system is used to ensure temperature stability during the fermentation process. The temperature control system maintains a constant temperature through a heat exchange system or a cooling system;

[0088] Set the ideal pH range during the fermentation process, and measure the pH value in the fermentation broth in real time through automated pH monitoring equipment. If it deviates from the set range, adjust it by adding acid or alkaline solution to ensure the activity of enzymes and the growth conditions of microorganisms during the fermentation process;

[0089] Use a dissolved oxygen detector to monitor the dissolved oxygen content in the fermentation liquid and adjust the oxygen supply according to demand. Use an oxygen ventilation system to adjust the dissolved oxygen concentration in the fermentation liquid;

[0090] Monitor and adjust the concentration of substrate nitrogen source, carbon source and trace elements during the fermentation process. When the system detects that the concentration of substrate or nutrients is lower than the set value, it automatically starts the supplement mechanism to add the required nutrients.

[0091] In this embodiment, the single chip microcomputer automatically adjusts the working state of the corresponding equipment according to the fermentation parameters set by the algorithm. The specific process is as follows:

[0092] Step 1: Data collection and preprocessing, denoising, use filters to remove noise from sensor data. Assuming the sensor signal is s(t) and the noise is n(t), the true value of the signal can be estimated through the filter:

[0093] in, is the filtered signal, K(t)K is the Kalman gain, and y(t) is the observed value;

[0094] The data is normalized by standardization to make it in the same range. When some data is missing, the Lagrange interpolation method is used to fill the missing values. In this step, during the data acquisition and preprocessing stage, the noise suppression effect of the filter in a dynamic environment makes the sensor data more accurate and stable, adapting to the variable factors in the complex fermentation process. This is impossible with traditional methods because it can dynamically adjust parameters to adapt to different interferences. High-quality data processed by filtering can significantly improve the accuracy of subsequent feature extraction and machine learning model training, making model predictions more accurate and robust, especially when sensors are often affected by environmental changes during the actual fermentation process, these interference factors can be effectively eliminated.

[0095] Step 2, feature extraction and selection, feature extraction of sensor data so that the machine learning model can effectively process, feature extraction includes time series features, environmental change features, time series features including temperature, pH value, dissolved oxygen time trend, periodicity, average value, standard deviation, environmental change features including external factors affecting fermentation in combination with external environmental conditions, external temperature, humidity, air pressure, and then fermentation stage marking: according to the different stages of the fermentation process, mark specific stage features, and select the most influential features on 5-HTP production through principal component analysis (PCA) or correlation analysis to simplify the model and improve computational efficiency; in this step, in addition to time series features, Fourier transform is used to analyze the frequency domain characteristics of key parameters in the fermentation process, and combined with external environmental factors (such as meteorological data) for prediction. This method combines the long-term trend of external changes with the short-term dynamics of the fermentation process, improving the multidimensionality and depth of feature extraction. This enables the feature selection process to fully capture the complex factors affecting 5-HTP production. Traditional feature extraction methods mostly rely only on time domain features and ignore the periodic effects in the frequency domain. Combined with the frequency domain characteristics after Fourier transformation, it is possible to identify and utilize periodic changes in the fermentation process (such as temperature fluctuations, pH changes, etc.), thereby improving the prediction accuracy of 5-HTP production, especially playing a key role in the stable control of the periodic fermentation process.

[0096] Step 3: Establish a machine learning model. The model is a random forest model. By constructing multiple decision trees and predicting the production of 5-HTP through a voting mechanism, each tree only uses part of the features during training and improves the prediction accuracy through weighted voting. The formula is as follows:

[0097] Among them, Y pred is the predicted value, h i (X) is the output of the i-th tree, X is the input feature, and N is the number of trees;

[0098] The gradient extreme gradient boosting was introduced to improve the prediction accuracy. The gradient extreme gradient was used to perform regression prediction on 5-HTP production. The formula is: Among them, α i is the weight of each tree, f i (X) is the output of the i-th tree, and N is the total number of trees;

[0099] The accuracy of the model was evaluated using the validation set, and the evaluation indicators included mean square error (MSE) and coefficient of determination (R 2 ), R 2 Indicators are used to measure the model fit:

[0100] Among them, Y true，i is the actual value, Y pred，i is the predicted value, is the mean of the actual values, and k-fold cross validation is used to further optimize the model to ensure the generalization ability of the model;

[0101] Random forest and XGBoost methods were selected to predict the production of 5-HTP, and the k-fold cross-validation optimization model was combined. This method is different from the traditional simple regression model and can effectively improve the accuracy of prediction and the generalization ability of the model. Ensemble learning improves the modeling ability of complex fermentation processes by constructing multiple decision trees and weighted voting, especially when dealing with multivariate and nonlinear data. The XGBoost method can better capture the complex nonlinear relationships that may occur in the 5-HTP fermentation process, and improve the adaptability and accuracy of the model under different fermentation conditions. This technology can accurately predict possible changes in the fermentation process, and then guide the optimization of the fermentation process, thereby significantly improving the yield and quality of 5-HTP.

[0102] Step 4: Data fusion and environmental change adjustment. For each sensor input, different weights are assigned based on its reliability. Suppose there are multiple sensor data X1, X2, ..., X n , the fused output is:

[0103]

[0104] Among them, w i is the weight of each sensor, which is adjusted according to the accuracy and historical performance of the sensor, and the sum of the weights is 1, that is:

[0105] Based on the external environmental data, an environmental change prediction model is established. Combined with the internal factors of the fermentation process, the changes that occur during the fermentation process are predicted. Multiple linear regression or deep neural network (DNN) models are used to achieve the following:

[0106] Y = f(X internal , X environment ), where X internal It is an internal factor, which includes temperature, dissolved oxygen, pH value, etc. environment is the environmental factor, Y is the final 5-HTP production prediction;

[0107] Step 5: Real-time adjustment of the fermentation process. Connect the prediction model with the automatic adjustment system. According to the output of the model, the output is the expected 5-HTP production and bacterial growth status, to adjust the fermentation process in real time, and use PID control algorithm or fuzzy control algorithm to adjust the temperature, pH, dissolved oxygen and nutrient supply.

[0108] In this embodiment, when establishing an environmental change prediction model, in addition to using a deep neural network (DNN), a convolutional neural network (CNN) is added to extract spatial features of environmental data to predict the impact of external factors on the fermentation process. The CNN model formula is: Among them, X i is the input data, W i is the weight, f(X i ) is the convolution operation.

[0109] Combining Kalman filtering with CNN to model environmental changes and achieve multi-level data fusion, especially weighted fusion processing of multi-sensor data, not only improves the reliability of data, but also enhances the ability to predict external environmental changes (such as temperature and humidity changes, air pressure fluctuations). The CNN model can further process and optimize the impact of environmental changes on the fermentation process. Through deep learning models (such as CNN), the system can more accurately predict the impact of changes in environmental factors on the fermentation process, further optimize the adjustment of fermentation parameters, so that the fermentation process can adapt to a more complex external environment, avoid the negative impact of environmental fluctuations, and improve the stability of the production process and 5-HTP production.

[0110] In this embodiment, the specific process of step 5 is as follows:

[0111] Obtain real-time monitoring data, including temperature, pH, dissolved oxygen, and nutrient concentration. For each control parameter, calculate the deviation between the current value and the target value:

[0112] e(t)=setpoint-measured(t);

[0113] Where: e(t) is the error value at the current moment, setpoint is the set target value, and measured(t) is the real-time measurement value;

[0114] The PID control algorithm is used for regulation. The PID control algorithm calculates a control output based on the error value to ensure that the system maintains the target state. The PID algorithm is based on three terms: proportional (P), integral (I) and differential (D), as shown in the following formula:

[0115]

[0116] Where: u(t) is the control quantity, K p is the proportional gain, K i is the integral gain, K d is the differential gain, e(t) is the error value at the current moment;

[0117] Implement regulation, the control system adjusts the corresponding equipment according to the control quantity u(t) calculated by PID, adjusts the temperature of the fermentation tank through the heat exchange system or the cooling system, adjusts the pH value of the fermentation liquid by adding acid or alkali, and changes the dissolved oxygen content by adjusting the ventilation volume or using the gas distribution system;

[0118] Monitor and adjust in real time. The real-time monitoring system continuously feeds back the adjusted control effect and further adjusts the control parameters when necessary. According to the difference between the model prediction and the actual data, the PID parameters are automatically adjusted to optimize the control effect.

[0119] Gradually optimize the control parameters by adjusting the PID parameter K p ,K i ,K d Continuous adjustment and optimization make the control system more accurate and stable.

[0120] The above process, based on the PID control algorithm, introduces adaptive PID control and fuzzy PID control methods to make the real-time adjustment of the fermentation process more flexible and accurate. This dynamic adjustment mechanism automatically optimizes the PID parameters according to real-time monitoring data, avoids the fixed parameter problem in traditional PID control, and improves the system's ability to cope with complex changes in the fermentation process. The combination of adaptive PID and fuzzy PID control algorithms enables the fermentation process to respond quickly and accurately under high dynamic changes, thereby optimizing key fermentation parameters such as temperature, pH, and dissolved oxygen. This adjustment strategy can avoid the overshoot phenomenon in traditional PID control, and enables the system to maintain a high degree of stability and efficiency when dealing with complex working conditions, ensuring high production of 5-HTP.

[0121] In this embodiment, a fermentation medium preparation and inoculation module, a fermentation tank, a sensor data acquisition module, a data processing unit and a single-chip microcomputer are included. The fermentation tank is used for biological fermentation of 5-hydroxytryptophan. The sensors include a temperature sensor, a pH sensor, a dissolved oxygen sensor and a nutrient concentration monitoring device.

[0122] The single-chip microcomputer is used to coordinate and control the work of each module, receive and process sensor data, and automatically adjust fermentation parameters according to the set algorithm and control model to implement PID or fuzzy control algorithm to adjust the fermentation environment in real time. All sensor data are collected in the single-chip microcomputer, and the real-time data is transmitted to the central control system through the communication interface. The single-chip microcomputer makes adjustments based on the data, and the actuator receives the control signal and performs real-time adjustment;

[0123] The fermentation medium preparation and inoculation module includes an automatic liquid adding mechanism and an inoculation component. The automatic liquid adding mechanism is used to automatically mix the substrate, nitrogen source, and carbon source components in proportion and add them to the fermentation tank. The inoculation component is used to control the inoculation amount of the strain and ensure the sterility of the inoculation. The automatic liquid adding mechanism and the inoculation component are connected to and controlled by the single chip microcomputer.

[0124] The sensor data acquisition module includes temperature, pH, dissolved oxygen, and nutrient concentration sensors. All sensors are connected to the data acquisition module. The data processing unit filters, standardizes, and interpolates the collected data. The data acquisition module transmits the processed data to the single-chip microcomputer or control computer for model prediction and fermentation environment adjustment.

[0125] Compared with the traditional 5-HTP biological fermentation method, the present embodiment realizes technological innovation in data processing, model prediction, fermentation parameter optimization, etc., especially through a variety of algorithms (such as Kalman filtering, XGBoost, fuzzy PID) and deep learning technology, it is possible to greatly improve production efficiency and product quality. The optimized fermentation process can reduce energy consumption, improve production stability, reduce by-product generation, and ultimately improve production efficiency and the output of 5-HTP. Especially in large-scale production, the improvement of automation and intelligent level makes the method more economically competitive. With the increase in demand for 5-HTP as a health product such as anti-depression and improved sleep, the biological fermentation technology with efficient and stable production process will have huge market potential. The above method and system have significantly improved the production efficiency, stability and product quality of 5-HTP by innovatively combining intelligent algorithms with traditional fermentation processes, especially in the application of large-scale production environments. There are outstanding advantages. These technical effects not only solve multiple problems in traditional fermentation processes, but also promote technological progress in the field of biopharmaceuticals, and have significant industrial application value.

[0126] In order to verify the effect of the embodiment, the following experiment is given:

[0127] 1. Purpose of the experiment

[0128] The effect of using the genetically modified strains, intelligent control system and membrane filtration technology of the present invention on improving the yield, purity, production cycle and cost of 5-HTP was verified.

[0129] 2. Experimental Grouping

[0130] Experimental group: The full set of technologies of this embodiment was used, including genetically engineered strains, automated and intelligent control systems in large-scale fermentation processes, and membrane filtration and reverse osmosis purification technologies.

[0131] Control group: using traditional fermentation technology, non-genetically modified strains, manually controlled fermentation parameters, and ordinary purification methods.

[0132] 3. Experimental strains

[0133] Experimental group: genetically modified Escherichia coli BL21 or yeast CEN.PK 113-11C, after genetic modification, tryptophan hydroxylase (TPH) and aromatic amino acid decarboxylase (AADC) genes were introduced to enhance their metabolic capacity.

[0134] Control group: original Escherichia coli BL21 or yeast CEN.PK 113-11C.

[0135] 4. Fermentation conditions

[0136] Culture medium: contains nitrogen source, carbon source, substrate tryptophan, vitamins, etc. to ensure the growth and metabolism of the strain.

[0137] Inoculation size: standard inoculation size based on strain activity (OD600=0.05).

[0138] Temperature, pH, and dissolved oxygen control: They were set to the optimal growth conditions respectively, and the experimental group made real-time adjustments through the intelligent control system.

[0139] Fermentation period: 48 hours.

[0140] 5. Data Collection

[0141] Main parameters monitored: temperature, pH, dissolved oxygen, substrate concentration, and 5-HTP concentration.

[0142] 5-HTP production: was measured regularly using high performance liquid chromatography (HPLC).

[0143] Fermentation endpoint: determined by bacterial growth curve and 5-HTP accumulation.

[0144] Experimental method steps

[0145] 1. Culture and inoculation of strains

[0146] The experimental strains were genetically modified to ensure that they could efficiently express TPH and AADC enzymes. Unmodified E. coli or yeast were used as controls. The fermentation medium was prepared, and the substrate (tryptophan) concentration was adjusted to 10 g / L, the nitrogen source was amino acids and nitrogen compounds, and the carbon source was glucose.

[0147] 2. Control of the fermentation process

[0148] By monitoring parameters such as temperature, pH, and dissolved oxygen, the experimental group used an automated control system to adjust the temperature (30-37°C), pH (5.8-6.2), and dissolved oxygen concentration (>40%). The control group used manual control to monitor and manually adjust various fermentation parameters.

[0149] 3. Extraction and purification of 5-HTP

[0150] After fermentation, 5-HTP is extracted by liquid-liquid extraction. The extract is purified by membrane filtration and reverse osmosis technology to further remove impurities and ensure the high purity of 5-HTP.

[0151] 4. Data Recording and Analysis

[0152] The changes in 5-HTP concentration at each time point during the experiment were recorded, and the total fermentation yield was calculated. The purity of the samples was regularly analyzed using HPLC to ensure the consistency of the experimental results.

[0153] Experimental data

[0154] Table 1 Changes in 5-HTP concentration during fermentation

[0155] Time (hours) 5-HTP concentration in control group (g / L) 5-HTP concentration in experimental group (g / L) 0 0 0 6 0.5 1.2 12 1.2 3.0 18 2.0 4.2 24 2.5 5.5 30 3.0 6.3 36 3.5 7.0 42 3.7 7.5 48 3.8 8.0

[0156] Conclusion: The production rate of 5-HTP in the experimental group was significantly higher than that in the control group. Within 48 hours, the 5-HTP concentration in the experimental group reached 8.0 g / L, while that in the control group was only 3.8 g / L.

[0157] Table 2 Purity analysis

[0158] Sample type purity(%) Control group (traditional purification method) 85.3％ Experimental Group (Membrane Filtration and Reverse Osmosis Purification) 97.2％

[0159] Conclusion: The 5-HTP purified by membrane filtration and reverse osmosis technology in the experimental group had a higher purity of 97.2%, which was significantly higher than the 85.3% of the traditional purification method.

[0160] Table 3 Cost analysis

[0161] project Control group (traditional technology) Experimental group (process of the present invention) Fermentation medium cost (yuan) 500 520 Energy consumption (yuan) 200 150 Purification cost (yuan) 300 200 Total cost (yuan) 1000 870

[0162] Conclusion: Compared with the traditional process, the experimental group saved energy consumption and reduced purification costs by using intelligent control and membrane filtration reverse osmosis technology, and the final total cost was reduced by 13%.

[0163] Table 4 Production cycle

[0164] Technology Fermentation cycle (hours) Total production cycle (hours) Control group (traditional technology) 48 72 Experimental group (process of the present invention) 48 60

[0165] Through intelligent control systems and automated operations, the experimental group was able to shorten the time of the purification process, reducing the production cycle by 12 hours compared to traditional processes.

[0166] The results are as follows, analyzing the yield: the yield of 5-HTP in the experimental group was significantly higher than that in the control group, an increase of about 2 times. The combination of genetic modification and intelligent control helps to improve metabolic efficiency and increase the synthesis of 5-HTP. Purity: Through the combination of membrane filtration and reverse osmosis technology, the purity of 5-HTP in the experimental group increased by nearly 12%, ensuring that the product quality meets higher standards. Cost-effectiveness: Although the cost of the fermentation medium in the experimental group increased slightly, the overall production cost was reduced by 13% compared with the control group due to energy and purification cost savings. Production cycle: Due to the optimization of the intelligent control system, the experimental group can quickly purify after the fermentation is completed, and the overall production cycle is shortened by 12 hours, which improves production efficiency.

[0167] In summary, the present invention significantly improves the yield and purity of 5-HTP, effectively reduces production costs and shortens production cycles by combining genetic engineering strain optimization, intelligent control systems, membrane filtration and reverse osmosis purification technologies. Experimental data show that the experimental group has obvious advantages in many aspects compared to traditional processes and has wide potential for industrial application.

[0168] This embodiment has the following advantages: 1. It has multiple genetically modified strain optimization. The embodiment uses CRISPR-Cas9 technology to carry out directional genetic modification of the strain and regulates the expression of key enzymes through the gene expression system. This scheme has not been widely used in traditional fermentation processes. In this way, not only the metabolic capacity of the strain is improved, but also its tolerance to fermentation intermediates is enhanced, effectively increasing the yield of 5-HTP.

[0169] 2. Combination of intelligent and automated control. The embodiment has made innovations in intelligent control, combining real-time monitoring, machine learning models and PID control algorithms to achieve precise adjustment of various parameters and environmental optimization during the fermentation process. In particular, in the data acquisition and processing process, filtering, standardization and interpolation methods are used to process sensor data, providing reliable input for subsequent machine learning models. This system can dynamically adjust the fermentation process, greatly improving the level of intelligence in the fermentation process.

[0170] 3. The environmental prediction of the combination of deep neural network and convolutional neural network is given. By introducing deep neural network (DNN) and convolutional neural network (CNN), the present invention can predict the potential problems of the fermentation process by using the changes in the external environment while predicting the internal changes in the fermentation process. This combination not only optimizes the adaptability to environmental changes, but also improves the accuracy and generalization ability of the model.

[0171] 4. PID or fuzzy control algorithm based on real-time data. In the embodiment, the PID control algorithm or fuzzy control algorithm can adjust various parameters in the fermentation tank according to real-time data, such as temperature, pH, dissolved oxygen, etc. The automatic adjustment and feedback mechanism of this system can quickly respond to deviations in the fermentation process, achieve precise control, and reduce human errors.

[0172] 5. The combined application of membrane filtration and reverse osmosis technology greatly improves the purification effect and reduces production costs during the extraction and purification of 5-HTP. The application of this combined technology has significant advantages, especially in large-scale production, and can improve the purity of the product and ensure its quality.

[0173] Through the above innovations, the present invention has significant advantages in increasing output, reducing costs, ensuring product quality, etc., and has strong application potential in the fermentation industry.

[0174] Although the present invention has been disclosed as above with preferred embodiments, it is not intended to limit the present invention. A person with ordinary knowledge in the technical field to which the present invention belongs may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be determined by the definition of the claims.

Claims

1. A biological fermentation process for preparing 5-hydroxytryptophan, characterized in that: The steps include: Step 1, strain optimization; Step 1.1, selecting and transforming strains, selecting microbial strains with high metabolic capacity as the basic strains for 5-HTP production, using the genetic engineering technology CRISPR-Cas9 to perform targeted transformation on the selected strains, introducing key enzyme genes of tryptophan hydroxylase (TPH) and aromatic amino acid decarboxylase (AADC), and enhancing the metabolic pathway of tryptophan conversion to 5-HTP, so as to obtain the transformed strains; Step 1.2, in the modified strain, adjusting the expression level of the key enzyme by optimizing the gene expression system; Step 1.3, modifying the tolerance of the inhibitory substances produced during the fermentation process so that the strain can withstand higher concentrations of substrates and products; Step 2, fermentation; Step 2.1, medium preparation and inoculation, prepare the fermentation medium, the medium includes substrate, nitrogen source and carbon source to meet the growth requirements of the bacteria, and before inoculation, pre-culture the selected modified strain in step 1 to ensure its activity and metabolic capacity, and prepare to inoculate it into the fermentation tank; Step 2.2, establish an efficient monitoring and control system to accurately control the fermentation environment. Arrange temperature sensors, pH sensors, dissolved oxygen sensors and nutrient concentration monitoring equipment in the fermenter to obtain changes in key parameter data during the fermentation process. All key parameter data are summarized and transmitted to the single-chip microcomputer through the data acquisition module. The single-chip microcomputer automatically adjusts the working status of the corresponding equipment according to the fermentation parameters set by the algorithm; Step 3, fermentation end and product extraction, the fermentation completion time is determined by monitoring and controlling the system, and whether the fermentation is completed is determined based on the cumulative amount of 5-HTP and the bacterial growth curve. At the end of the fermentation, oxygen supply and nutrient addition are stopped to maintain a stable fermentation environment; 5-HTP is extracted from the fermentation broth using liquid-liquid extraction or centrifugal separation, and 5-HTP is further purified using membrane filtration and reverse osmosis to remove impurities and improve the purity of 5-HTP. The extracted and purified 5-HTP is then subjected to quality testing, including purity, content, and sterility testing.

2. The biological fermentation process for preparing 5-hydroxytryptophan according to claim 1, characterized in that: In the step 1.1, the microbial strains are Escherichia coli BL21 and yeast CEN.PK 113-11C.

3. The biological fermentation process for preparing 5-hydroxytryptophan according to claim 1, characterized in that: In step 1.2, the Gal4 / UAS system or the Tet-On / Tet-Off system is used to achieve precise regulation of specific enzyme genes, and an enhancer-controlled binary promoter system is used in combination with a heat-inducible promoter and a non-toxic activation system to achieve time-effective regulation of key enzyme genes during the fermentation process.

4. The biological fermentation process for preparing 5-hydroxytryptophan according to claim 1, characterized in that: In the step 1.3, one or more of the ethanol-resistant ADH2 gene, the amino acid-resistant gene, and the acid-resistant gene are used for gene integration, and the amino acid-resistant gene and the acid-resistant gene are drug-resistant genes.

5. The biological fermentation process for preparing 5-hydroxytryptophan according to claim 1, characterized in that: In step 2.2, the temperature sensor is used to monitor the temperature change of the fermentation liquid in real time, the pH sensor monitors the pH of the fermentation liquid and automatically adjusts it through a feedback system, the dissolved oxygen sensor monitors the concentration of dissolved oxygen during the fermentation process and helps adjust the oxygen supply system, and the nutrient concentration monitoring equipment monitors the concentration of the substrate, nitrogen source, and carbon source through online chromatography or spectroscopy technology.

6. The biological fermentation process for preparing 5-hydroxytryptophan according to claim 5, characterized in that: The single chip microcomputer automatically adjusts the working state of the corresponding equipment according to the fermentation parameters set by the algorithm, including the following process: According to the optimal growth temperature range of the selected strain, a precise temperature control system is used to ensure temperature stability during the fermentation process. The temperature control system maintains a constant temperature through a heat exchange system or a cooling system; Set the ideal pH range during the fermentation process, and measure the pH value in the fermentation broth in real time through automated pH monitoring equipment. If it deviates from the set range, adjust it by adding acid or alkaline solution to ensure the activity of enzymes and the growth conditions of microorganisms during the fermentation process; Use a dissolved oxygen detector to monitor the dissolved oxygen content in the fermentation liquid, and adjust the oxygen supply according to demand, and use an oxygen ventilation system to adjust the dissolved oxygen concentration in the fermentation liquid; Monitor and adjust the concentration of substrate nitrogen source, carbon source and trace elements during the fermentation process. When the system detects that the concentration of substrate or nutrients is lower than the set value, it automatically starts the supplement mechanism to add the required nutrients.

7. The biological fermentation process for preparing 5-hydroxytryptophan according to claim 6, characterized in that: The single chip microcomputer automatically adjusts the working state of the corresponding equipment according to the fermentation parameters set by the algorithm. The specific process is as follows: Step 1: Data collection and preprocessing, denoising, use filters to remove noise from sensor data. Assuming the sensor signal is s(t) and the noise is n(t), the true value of the signal can be estimated through the filter: in, is the filtered signal, K(t)K is the Kalman gain, and y(t) is the observed value; Standardization was used to normalize the data to make them in the same range, and when some data were missing, the Lagrange interpolation method was used to fill in the missing values; Step 2: Feature extraction and selection: feature extraction of sensor data to enable the machine learning model to process effectively. Feature extraction includes time series features and environmental change features. Time series features include the time trend, periodicity, average value, and standard deviation of temperature, pH value, and dissolved oxygen. Environmental change features include external factors that affect fermentation in combination with external environmental conditions, such as external temperature, humidity, and air pressure. Then, the fermentation stage is marked: according to the different stages of the fermentation process, specific stage features are marked. Through principal component analysis (PCA) or correlation analysis, the features that have the greatest impact on 5-HTP production are selected to simplify the model and improve computational efficiency. Step 3: Establish a machine learning model. The model is a random forest model. By constructing multiple decision trees and predicting the production of 5-HTP through a voting mechanism, each tree only uses part of the features during training and improves the prediction accuracy through weighted voting. The formula is as follows: Among them, Y pred is the predicted value, h i (X) is the output of the i-th tree, X is the input feature, and N is the number of trees; The gradient extreme gradient boosting was introduced to improve the prediction accuracy. The gradient extreme gradient was used to perform regression prediction on 5-HTP production. The formula is: Among them, α i is the weight of each tree, f i (X) is the output of the i-th tree, and N is the total number of trees; The accuracy of the model was evaluated using the validation set, and the evaluation indicators included mean square error (MSE) and coefficient of determination (R 2 ), R 2 Indicators are used to measure the model fit: Among them, Y true，i is the actual value, Y pred，i is the predicted value, is the mean of the actual values, and k-fold cross validation is used to further optimize the model to ensure the generalization ability of the model; Step 4: Data fusion and environmental change adjustment. For each sensor input, different weights are assigned based on its reliability. Suppose there are multiple sensor data X1, X2, ..., X n , the fused output is: Among them, w i is the weight of each sensor, which is adjusted according to the accuracy and historical performance of the sensor, and the sum of the weights is 1, that is: Based on the external environmental data, an environmental change prediction model is established. Combined with the internal factors of the fermentation process, the changes that occur during the fermentation process are predicted. Multiple linear regression or deep neural network (DNN) models are used to achieve the following: Y = f(X internal , X environment ), where X internal It is an internal factor, which includes temperature, dissolved oxygen, pH value, etc. environment is the environmental factor, Y is the final 5-HTP production prediction; Step 5: Real-time adjustment of the fermentation process. Connect the prediction model with the automatic adjustment system. According to the output of the model, the output is the expected 5-HTP production and bacterial growth status, to adjust the fermentation process in real time, and use PID control algorithm or fuzzy control algorithm to adjust the temperature, pH, dissolved oxygen and nutrient supply.

8. The biological fermentation process for preparing 5-hydroxytryptophan according to claim 7, characterized in that: When establishing an environmental change prediction model, in addition to using a deep neural network (DNN), a convolutional neural network (CNN) is added to extract spatial features from environmental data to predict the impact of external factors on the fermentation process. The CNN model formula is: Among them, X i is the input data, W i is the weight, f(X i ) is the convolution operation.

9. The biological fermentation process for preparing 5-hydroxytryptophan according to claim 7, characterized in that: The specific process of step 5 is as follows: Obtain real-time monitoring data, including temperature, pH, dissolved oxygen, and nutrient concentration. For each control parameter, calculate the deviation between the current value and the target value: e(t) = setpoint - measure(t); Where: e(t) is the error value at the current moment, setpoint is the set target value, and measured(t) is the real-time measurement value; The PID control algorithm is used for regulation. The PID control algorithm calculates a control output based on the error value to ensure that the system maintains the target state. The PID algorithm is based on three terms: proportional (P), integral (I) and differential (D), as shown in the following formula: Where: u(t) is the control quantity, K p is the proportional gain, K i is the integral gain, K d is the differential gain, e(t) is the error value at the current moment; Implement regulation, the control system adjusts the corresponding equipment according to the control quantity u(t) calculated by PID, adjusts the temperature of the fermentation tank through the heat exchange system or the cooling system, adjusts the pH value of the fermentation liquid by adding acid or alkali, and changes the dissolved oxygen content by adjusting the ventilation volume or using the gas distribution system; Monitor and adjust in real time. The real-time monitoring system continuously feeds back the adjusted control effect and further adjusts the control parameters when necessary. According to the difference between the model prediction and the actual data, the PID parameters are automatically adjusted to optimize the control effect. Gradually optimize the control parameters by adjusting the PID parameter K p ,K i ,K d Continuous adjustment and optimization make the control system more accurate and stable.

10. The biological fermentation system for preparing 5-hydroxytryptophan according to claim 1, characterized in that: It includes a fermentation medium preparation and inoculation module, a fermentation tank, a sensor data acquisition module, a data processing unit and a single-chip microcomputer. The fermentation tank is used for biological fermentation of 5-hydroxytryptophan. The sensors include a temperature sensor, a pH sensor, a dissolved oxygen sensor and a nutrient concentration monitoring device. The single-chip microcomputer is used to coordinate and control the work of each module, receive and process sensor data, and automatically adjust fermentation parameters according to the set algorithm and control model to implement PID or fuzzy control algorithm to adjust the fermentation environment in real time. All sensor data are collected in the single-chip microcomputer, and the real-time data is transmitted to the central control system through the communication interface. The single-chip microcomputer makes adjustments based on the data, and the actuator receives the control signal and performs real-time adjustment; The fermentation medium preparation and inoculation module includes an automatic liquid adding mechanism and an inoculation component. The automatic liquid adding mechanism is used to automatically mix the substrate, nitrogen source, and carbon source components in proportion and add them to the fermentation tank. The inoculation component is used to control the inoculation amount of the strain and ensure the sterility of the inoculation. The automatic liquid adding mechanism and the inoculation component are connected to and controlled by the single chip microcomputer. The sensor data acquisition module includes temperature, pH, dissolved oxygen, and nutrient concentration sensors. All sensors are connected to the data acquisition module. The data processing unit filters, standardizes, and interpolates the collected data. The data acquisition module transmits the processed data to the single-chip microcomputer or control computer for model prediction and fermentation environment adjustment.

Images