An intelligent feedback feeding control system based on on-line detection of substrate concentration during fermentation

Abstract

The present invention discloses an intelligent feedback feeding control system based on on-line detection of substrate concentration during the fermentation process. The substrate concentration in the bioreactor is detected by a biosensing analyzer, and the detection result is transmitted to the upper computer control system through serial communication. The upper computer control system reads the weight data of the bioreactor provided by the weighing scale, automatically calculates the substrate consumption rate according to the substrate concentration data, predicts the substrate concentration data at the next monitoring time point, and compares it with the preset feeding target concentration. When feeding is required at the next monitoring time point, the system transmits the required feeding liquid volume data to the peristaltic pump, and the peristaltic pump automatically pumps the feeding liquid into the bioreactor according to the instruction, thereby completing the feedback feeding of the fermentation process. This system reduces manual intervention, lowers production costs, overcomes the drawbacks of feeding during the fermentation process based on experience, and improves the efficiency and product quality of microbial fermentation production.

Description

Technical Field

[0001] The present invention relates to the technical field of on-line monitoring and control of microbial fermentation processes, and specifically to a control system that uses an on-line biosensing analyzer to achieve on-line detection of the substrate concentration in a fermentation system, and then performs intelligent feedback feeding on the biological fermentation process. Background Art

[0002] Microbial fermentation is a complex biochemical process that involves the ability of various microorganisms such as bacteria, fungi, yeast, etc. to decompose organic matter or synthesize specific compounds under specific conditions. During the fermentation process, microorganisms use carbon sources, nitrogen sources, and other nutrients in the culture medium for metabolic activities to produce the desired products. These products can be valuable industrial products such as antibiotics, amino acids, enzymes, vitamins, organic acids, biofuels (such as ethanol), etc. During the fermentation process, microorganisms will go through four stages: the lag phase, the logarithmic growth phase, the stationary phase, and the death phase. The lag phase is the stage where microorganisms adapt to the new environment; the logarithmic growth phase is the period when microorganisms rapidly proliferate; the stationary phase is the period when the cell number reaches the maximum and the product accumulation is the most; the death phase is the process where microorganisms start to die after the nutrients are exhausted. The improvement of fermentation efficiency mainly depends on the optimized management of the logarithmic growth phase and the stationary phase. Therefore, on-line monitoring during these two stages is very necessary, which helps to timely adjust the fermentation conditions, maintain the optimal growth state of microorganisms, and thus improve the yield and quality of products. Most existing fermentation processes rely on off-line sampling and analysis. The data obtained by this method has lag, that is, the detected concentration data is actually the result at a certain past moment, which leads to missing the best adjustment opportunity when problems occur in the fermentation process. Therefore, traditional feeding strategies often determine when to feed and the amount of feeding based on empirical values. Such a feeding method is likely to cause an excess or deficiency of nutrients, affect the metabolic activities of microorganisms, and even cause the accumulation of metabolic by-products, further affecting the quality and yield of fermentation products. Moreover, the fermentation process still needs to rely on frequent manual monitoring and adjustment, which not only increases the labor intensity, but also the inconsistency of manual operations will lead to instability of the fermentation process, affecting the consistency and repeatability of products. To overcome these problems, it is necessary to develop a new on-line monitoring and intelligent feeding control system that can detect the changes of key parameters in the fermentation process in real time, predict in advance and timely adjust the feeding strategy, so as to ensure the stability and efficiency of the fermentation process. Summary of the Invention

[0003] Object of the Invention: The object of the present invention is to overcome the disadvantages of large lag in the detection data of the substrate concentration in the existing fermentation process and inaccurate control of the existing feeding control method, and to propose a control system for intelligent feedback feeding that can on-line detect the substrate concentration in the fermentation system and perform predictive feeding according to the detection results.

[0004] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0005] An intelligent feedback feeding control system based on on-line detection of substrate concentration during fermentation. The system forms a closed-loop circuit composed of a biosensing analyzer, a bioreactor, a peristaltic pump, a weighing scale and a host computer control system. The biosensing analyzer and the weighing scale serve as data input components of the control system, and the peristaltic pump serves as a data output component of the control system. The input and output components use serial communication for data interaction to jointly control the substrate concentration in the bioreactor;

[0006] Among them, the biosensing analyzer is used to detect the substrate concentration in the bioreactor and transmit the results to the host computer control system through serial communication. At the same time, the host computer control system reads the weight data of the bioreactor weighed by the weighing scale, automatically calculates the substrate consumption rate according to the substrate concentration data, and predicts the substrate concentration data at the next monitoring time point. The predicted substrate concentration data is compared with the feeding target concentration data. If feeding is required at the next monitoring time point, the required feeding liquid volume data is transmitted to the peristaltic pump, and the peristaltic pump automatically pumps the feeding liquid into the bioreactor, thereby completing the feedback feeding of the fermentation process.

[0007] Specifically, the host computer control system pre-sets the detection period F (h) of the fermentation process, the feedback feeding concentration control target value c r (g / L), the feeding control range coefficient K (%), the substrate feeding liquid concentration c f (g / L), the flow rate u p (L / h) of the feeding peristaltic pump. After setting the initial parameters, the biosensing analyzer will detect the substrate concentration according to the detection period F (h) and transmit the data to the control system. At the same time, the weighing scale will transmit the volume change of the bioreactor when the substrate concentration result comes out to the control system. After the control system receives the two groups of substrate concentration data, it automatically calculates the substrate concentration consumption rate u n . The relationship between the substrate concentration consumption rate and the substrate concentration data during the fermentation process is as follows:

[0008]

[0009] Among them, u n is the substrate consumption rate between two periods, c n-1 is the substrate concentration of the previous period, V n-1 is the volume of the fermentation broth in the bioreactor of the previous period, c n is the substrate concentration of this period, V n is the volume of the fermentation broth in the bioreactor of this period, t n -t n-1 is the time difference between this period and the previous period.

[0010] Specifically, after calculating the consumption rate of the substrate concentration, calculate the predicted value of the substrate concentration for the next cycle, and compare the predicted value of the substrate concentration for the next cycle with the feedback feeding concentration control target value c r (g / L). If the predicted value of the substrate concentration for the next cycle is lower than the feedback feeding concentration control target value c r (g / L), then feeding is required in the next cycle, and it is necessary to start predicting the feeding time point within this feeding cycle; if the predicted value of the substrate concentration for the next cycle is not lower than the feedback feeding concentration control target value c r (g / L), then the biosensing analyzer detects the substrate concentration in the bioreactor in the next cycle, and continues to calculate the consumption rate of the substrate concentration in the most recent two cycles and the predicted value of the substrate concentration in the next cycle and compare it with the feedback feeding concentration control target value c r (g / L) until the predicted value of the substrate concentration in a certain cycle is lower than the feedback feeding concentration control target value c r (g / L).

[0011] Specifically, the relationship between the predicted value of the substrate concentration for the next cycle and the substrate concentration for the previous cycle in the fermentation process is:

[0012]

[0013] Among them, c n+1 is the predicted value of the substrate concentration for the next cycle, c n is the substrate concentration for this cycle, V n is the volume of the fermentation broth in the bioreactor for this cycle, u n is the substrate consumption rate between two cycles, V n+1 is the volume of the fermentation broth in the bioreactor for the next cycle, t n+1 -t n is the time difference between the next cycle and this cycle.

[0014] Specifically, when feeding is required in the next cycle, then this cycle is defined as the feeding cycle, and the feeding time point is calculated within the feeding cycle, that is, the specific time point when the substrate concentration within the cycle is lower than the target concentration. The relationship between the predicted value of the feeding time point and the target feeding value within the feeding cycle in the fermentation process is:

[0015]

[0016] Among them, is the predicted value of the feeding time point within the feeding cycle in the fermentation process, c n is the substrate concentration for this cycle, V n is the volume of the fermentation broth in the bioreactor for this cycle, c r is the feedback feeding concentration control target value, K is the feeding control range coefficient, u nis the substrate consumption rate between two cycles, and t n is the time of this cycle.

[0017] Further, after obtaining the feeding time point within the feeding cycle, the biosensing analyzer detects whether the substrate concentration in the bioreactor is lower than the feedback feeding concentration control target value c r (g / L) at this time point. If it is lower than the feedback feeding concentration control target value c r (g / L), then the feeding program is started. Otherwise, continue to calculate the substrate concentration consumption rate within the most recent two cycles and the predicted value of the substrate concentration in the next cycle and compare it with the feedback feeding concentration control target value c r (g / L) until the predicted value of the substrate concentration in a certain cycle is lower than the feedback feeding concentration control target value c r (g / L), and calculate the feeding time point within this feeding cycle.

[0018] Specifically, the feeding program of the control system is mainly executed by a peristaltic pump. The control system calculates the running time of the peristaltic pump based on the existing data. After the running time of the peristaltic pump ends, the intelligent feedback feeding in the first fermentation process is completed until the detection cycle ends. The relationship between the running time and the existing data is:

[0019]

[0020] Among them, Δt n+1 is the running time of the peristaltic pump, V′ n+1 is the volume of the fermentation broth in the bioreactor at the feeding time point, c′ n+1 is the substrate concentration in the bioreactor at the feeding time point, c r is the feedback feeding concentration control target value, K is the feeding control range coefficient, u′ n+1 is the substrate consumption rate from the feeding time point to this cycle, t f -t′ n+1 is the time difference between the feeding time point and the next cycle, c f is the concentration of the feeding solution, u p is the running flow rate of the feeding peristaltic pump.

[0021] Specifically, when entering the feeding cycle, the control system needs to calculate a new consumption rate based on the substrate concentration measured by the biosensing analyzer. The relationship between the consumption rate and the substrate concentration after the feeding cycle is:

[0022]

[0023] Among them, u′ n+1 is the substrate consumption rate from the feeding time point to this cycle, c n is the substrate concentration of this cycle, V nis the volume of the fermentation broth in the bioreactor for this cycle, V′ n+1 is the volume of the fermentation broth in the bioreactor at the feeding time point, c′ n+1 is the substrate concentration in the bioreactor at the feeding time point, t′ n+1 -t n is the time difference between the feeding time point and the next cycle.

[0024] Specifically, the microbial growth process is divided into 4 stages: lag phase, logarithmic growth phase, stationary phase, and death phase. During the lag phase and the early stage of the logarithmic growth phase of the microorganism, a biosensing analyzer is used to test the substrate concentration data in the bioreactor, and a consumption rate metabolic curve is established. Based on this data, the substrate concentration data for the next cycle is predicted, and it is judged whether it is lower than the feeding target concentration. The peristaltic pump is driven in a timely manner to prevent the microorganism from entering the death phase, so that the microorganism can normally produce the target substance and improve production efficiency.

[0025] Furthermore, the present invention also claims the application of the above control system for automatic feeding in the fermentation process of Saccharomyces cerevisiae.

[0026] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0027] The system of the present invention is a control system for intelligent feedback feeding based on on-line detection of substrate concentration during the fermentation process. Based on the substrate concentration data detected by the biosensing analyzer, a substrate concentration consumption curve is constructed. The control system automatically predicts the feeding time point to perform the feeding control work of fermentation production. Through on-line monitoring and intelligent feedback feeding, the efficiency and product quality of the fermentation process are improved. It has a pre-prediction function and a constant-speed feeding function, with accurate control and good operation repeatability, enabling the fermentation process to be in a good process, reducing manual intervention, lowering production costs, overcoming the drawbacks of feeding during the fermentation process based on experience, thus enabling the microbial fermentation production to proceed efficiently. Through precise control, resource waste is reduced, energy conservation and emission reduction are achieved, and the engineering application level of optimized feeding control in the fermentation process is promoted. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The following further specifically describes the present invention in conjunction with the drawings and specific embodiments, and the above and / or other advantages of the present invention will become clearer.

[0029] Figure 1 is a schematic diagram of the control system for intelligent feedback feeding based on on-line detection of substrate concentration during the fermentation process of the present invention.

[0030] Figure 2 is a flowchart of the control system for intelligent feedback feeding with on-line detection of substrate concentration during the fermentation process of the present invention.

[0031] Figure 3is the glucose concentration of the fermentation broth detected by a biosensor analyzer during the fermentation process. Detailed implementation mode

[0032] The present invention can be better understood according to the following embodiments.

[0033] Combined with Figure 1 And Figure 2 As shown, the control system of intelligent feedback feeding based on on-line detection of substrate concentration during the fermentation process of the present invention mainly forms a closed-loop circuit by a biosensor analyzer, a bioreactor, a peristaltic pump, a weighing scale and a host computer control system. The biosensor analyzer and the weighing scale are used as data input components of the control system, and the peristaltic pump is used as a data output component of the control system. The input and output components use serial communication for data interaction to jointly control the substrate concentration in the bioreactor.

[0034] The biosensor analyzer is used to detect the substrate concentration in the bioreactor and transmit the results to the host computer control system through serial communication. At the same time, the host computer control system reads the weight data of the bioreactor weighed by the weighing scale, automatically calculates the substrate consumption rate according to the substrate concentration data and predicts the substrate concentration data at the next monitoring time point. Compare the predicted substrate concentration data with the feeding target concentration data. If feeding is required at the next monitoring time point, the required feeding liquid volume data will be transmitted to the peristaltic pump, and the peristaltic pump will automatically pump the feeding liquid into the bioreactor, thereby completing the feedback feeding of the fermentation process.

[0035] The host computer control system presets the detection period F (h), the feedback feeding concentration control target value c r (g / L), the feeding control range coefficient K (%), the substrate feeding liquid concentration c f (g / L), the flow rate u p (L / h) of the feeding peristaltic pump. After setting the initial parameters, the biosensor analyzer will detect the substrate concentration according to the detection period F (h) and transmit the data to the control system. At the same time, the weighing scale will transmit the volume change of the bioreactor when the substrate concentration result comes out to the control system. After the control system receives the two groups of substrate concentration data, it automatically calculates the substrate concentration consumption rate u n , the relationship between the substrate concentration consumption rate and the substrate concentration data during the fermentation process is:

[0036]

[0037] Among them, u n is the substrate consumption rate between two cycles, c n-1 is the substrate concentration of the previous cycle, V n-1 is the volume of the fermentation broth in the bioreactor of the previous cycle, c n is the substrate concentration of this cycle, Vn is the volume of the fermentation broth in the bioreactor for this cycle, t n -t n-1 is the time difference between this cycle and the previous cycle.

[0038] After calculating the substrate concentration consumption rate, calculate the predicted value of the substrate concentration for the next cycle, and compare the predicted value of the substrate concentration for the next cycle with the feedback feeding concentration control target value c r (g / L). If the predicted value of the substrate concentration for the next cycle is lower than the feedback feeding concentration control target value c r (g / L), then feeding is required in the next cycle, and it is necessary to start predicting the feeding time point within this feeding cycle; if the predicted value of the substrate concentration for the next cycle is not lower than the feedback feeding concentration control target value c r (g / L), then the biosensing analyzer detects the substrate concentration in the bioreactor in the next cycle, and continues to calculate the substrate concentration consumption rate in the most recent two cycles and the predicted value of the substrate concentration in the next cycle and compare it with the feedback feeding concentration control target value c r (g / L) until the predicted value of the substrate concentration in a certain cycle is lower than the feedback feeding concentration control target value c r (g / L).

[0039] The relationship between the predicted value of the substrate concentration in the next cycle and the substrate concentration in the previous cycle during the fermentation process is:

[0040]

[0041] where c n+1 is the predicted value of the substrate concentration for the next cycle, c n is the substrate concentration for this cycle, V n is the volume of the fermentation broth in the bioreactor for this cycle, u n is the substrate consumption rate between two cycles, V n+1 is the volume of the fermentation broth in the bioreactor for the next cycle, t n+1 -t n is the time difference between the next cycle and this cycle.

[0042] When feeding is required in the next cycle, then define this cycle as the feeding cycle, and calculate the feeding time point within the feeding cycle, that is, the specific time point when the substrate concentration in the cycle is lower than the target concentration. The relationship between the predicted value of the feeding time point and the target feeding value during the feeding cycle of the fermentation process is:

[0043]

[0044] where is the predicted value of the feeding time point during the feeding cycle of the fermentation process, c n is the substrate concentration for this cycle, Vn is the volume of the fermentation broth in the bioreactor for this cycle, c r is the target value of the feedback feeding concentration control, K is the feeding control range coefficient, u n is the substrate consumption rate between two cycles, t n is the time of this cycle.

[0045] After obtaining the feeding time point within the feeding cycle, the biosensing analyzer detects whether the substrate concentration in the bioreactor at this time point is lower than the target value c of the feedback feeding concentration control r (g / L). If it is lower than the target value c of the feedback feeding concentration control r (g / L), the feeding program is started. Otherwise, continue to calculate the substrate concentration consumption rate in the last two cycles and the predicted value of the substrate concentration in the next cycle and compare it with the target value c of the feedback feeding concentration control r (g / L) until the predicted value of the substrate concentration in a certain cycle is lower than the target value c of the feedback feeding concentration control r (g / L), and calculate the feeding time point within this feeding cycle.

[0046] The feeding program of the control system is mainly executed by the peristaltic pump. The control system calculates the running time of the peristaltic pump according to the existing data. After the running time of the peristaltic pump ends, the intelligent feedback feeding of the first fermentation process is completed until the end of the detection cycle. The relationship between the running time and the existing data is:

[0047]

[0048] Among them, Δt n+1 is the running time of the peristaltic pump, V′ n+1 is the volume of the fermentation broth in the bioreactor at the feeding time point, c′ n+1 is the substrate concentration in the bioreactor at the feeding time point, c r is the target value of the feedback feeding concentration control, K is the feeding control range coefficient, u′ n+1 is the substrate consumption rate from the feeding time point to this cycle, t f -t′ n+1 is the time difference between the feeding time point and the next cycle, c f is the concentration of the feeding solution, u p is the running flow rate of the feeding peristaltic pump.

[0049] Entering the feeding cycle, the control system needs to calculate a new consumption rate according to the substrate concentration measured by the biosensing analyzer. The relationship between the consumption rate and the substrate concentration after the feeding cycle is:

[0050]

[0051] Among them, u′n+1 is the substrate consumption rate of the feeding time point from the substrate consumption rate of this cycle, c n is the substrate concentration of this cycle, V n is the volume of the fermentation broth in the bioreactor of this cycle, V′ n+1 is the volume of the fermentation broth in the bioreactor at the feeding time point, c′ n+1 is the substrate concentration in the bioreactor at the feeding time point, t′ n+1 -t n is the time difference between the feeding time point and the next cycle.

[0052] The present invention divides the microbial growth process into four stages: lag phase, logarithmic growth phase, stationary phase, and decline phase. During the lag phase and the early logarithmic growth phase of the microorganism, a biosensing analyzer is used to test the substrate concentration data in the bioreactor, and a consumption rate metabolic curve is established. Based on this data, the substrate concentration data in the next cycle is predicted, and it is judged whether it is lower than the feeding target concentration. The peristaltic pump is driven in time to prevent the microorganism from entering the decline phase, so that the microorganism can normally produce the target substance and improve the production efficiency.

[0053] The control system is used in the measurement and control of the Saccharomyces cerevisiae fermentation process to realize the intelligent regulation of the substrate concentration during the fermentation process. The initial concentration of glucose in the fermentation broth is 78.81 g / L, and the added glucose concentration is 500 g / L. A biosensing analyzer is used to detect the glucose concentration in the fermentation broth. After the detection is completed, the data is transmitted to the control system. At the same time, the weighing scale transmits the weight change of the bioreactor when the glucose concentration result comes out to the control system. After the control system receives the two groups of substrate concentration data, it calculates the substrate consumption rate in this cycle based on the two groups of substrate data, and judges whether feeding is required in the next cycle. If feeding is required, the control system calculates the feeding amount and sends the calculated feeding amount to the injection pump to complete the feeding operation. In the measurement and control of the Saccharomyces cerevisiae fermentation process, the detection frequency of the biosensing analyzer is set to 30 min, the target concentration is 26 g / L, the feeding control range coefficient K(%) is 10, and the feeding pump flow rate is 10 ml / min. The control system for intelligent feedback feeding based on the on-line detection of the substrate concentration during the fermentation process is as Figure 3 is the glucose concentration of the fermentation broth substrate detected by the biosensing analyzer during the fermentation process. From Figure 3It can be seen that the total fermentation time is 399 minutes. Among them, within 0 - 314 minutes, the glucose concentration rapidly decreases. During this period, the biosensing analyzer samples and detects every 30 minutes, and the host computer analyzes the detection results. However, based on the calculation, the host computer does not need to feed as the substrate glucose concentration is above the target concentration during this period. In the subsequent 314 - 399 minutes, it enters the feeding cycle. The host computer accurately predicts that the substrate glucose concentration in the fermentation broth will be lower than the target concentration at 323 minutes based on the sampled data at 284 minutes and 314 minutes, and controls the biosensing analyzer to sample and detect at 323 minutes. The detection result is 25.6 g / L, which is lower than the target concentration, and feeding starts. The host computer calculates the feeding volume to be 30 mL based on the calculation and controls the peristaltic pump to feed. After the feeding is completed, the host computer controls the biosensing analyzer to detect the substrate glucose concentration in the fermentation broth again, and the result is 28.53 g / L, which is within the feeding control range, and the first feeding is completed.

[0054] From Figure 3 It can be seen that after the first feeding is completed, the host computer predicts that the substrate glucose concentration in the fermentation broth will be lower than the target concentration at 351 minutes based on the sugar consumption rate calculated from the sampled detection results at 314 minutes and 323 minutes, and controls the biosensing analyzer to sample and detect at 351 minutes. The detection result is 25.18 g / L, which is lower than the target concentration, and feeding starts. The host computer calculates the feeding volume to be 35.4 mL based on the calculation and controls the peristaltic pump to feed. After the feeding is completed, the host computer controls the biosensing analyzer to detect the substrate glucose concentration in the fermentation broth again, and the result is 28.81 g / L, which is within the feeding control range, and the second feeding is completed. From Figure 3 It can be seen that after the second feeding is completed, the host computer predicts that the substrate glucose concentration in the fermentation broth will be lower than the target concentration at 386 minutes based on the sugar consumption rate calculated from the sampled detection results at 329 minutes and 351 minutes, and controls the biosensing analyzer to sample and detect at 386 minutes. The detection result is 25.18 g / L, which is lower than the target concentration, and feeding starts. The host computer calculates the feeding volume to be 35.7 mL based on the calculation and controls the peristaltic pump to feed. After the feeding is completed, the host computer controls the biosensing analyzer to detect the substrate glucose concentration in the fermentation broth again, and the result is 28.59 g / L, which is within the feeding control range, and the third feeding is completed.

[0055] Table 1 Data of the change of substrate glucose concentration and fermentation broth volume with time in the yeast fermentation experiment

[0056]

[0057] In different fermentation systems, the feeding control range coefficient K(%) needs to be appropriately adjusted according to the substrate consumption rate of the bacterial cells. If the substrate consumption rate is fast, the feeding control range coefficient K(%) needs to be increased to leave sufficient time to complete the "detection - feeding - detection" process. If the substrate consumption rate is slow, the feeding control range coefficient K(%) can be reduced to achieve more precise feeding control.

[0058] Existing feedback feeding control schemes all indirectly reflect the substrate consumption during fermentation based on the pH and dissolved oxygen of the fermentation broth, with a large deviation and unable to meet the requirements of high-precision fermentation feeding control. The intelligent feedback feeding control system based on the online detection of substrate concentration during fermentation described in the present invention adopts an online detection feedback feeding scheme for the substrate, and can adjust the feeding amount according to the substrate consumption rate in each cycle during fermentation, so as to meet the requirements of high-precision fermentation feeding control.

[0059] The intelligent feedback feeding control system based on the online detection of substrate concentration during fermentation automatically corrects the feedback feeding control according to the sugar consumption iteration. Compared with the existing feeding methods based on physical and chemical parameters (such as pH, dissolved oxygen, etc.), it is more direct and efficient, has stronger adaptability to different fermentation systems, can realize unattended operation during the fermentation process, and has broad application prospects in the field of fermentation.

[0060] The present invention provides an idea and method for an intelligent feedback feeding control system based on the online detection of substrate concentration during fermentation. There are many methods and ways to specifically implement this technical solution. The above description is only the preferred implementation mode of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention. Each component not clearly defined in this embodiment can be realized by existing technologies.

Claims

1. An intelligent feedback feeding control system based on on-line detection of substrate concentration during fermentation process, characterized in that, The system consists of a biosensor analyzer, a bioreactor, a peristaltic pump, a dosing scale and a host computer control system to form a closed loop. The biosensor analyzer and the dosing scale serve as data input components of the control system, and the peristaltic pump serves as a data output component of the control system. The input and output components use serial communication to exchange data and coordinately control the substrate concentration in the bioreactor. Among them, a biosensing analyzer is used to detect the substrate concentration in the bioreactor and transmit the results to the upper computer control system through serial communication. At the same time, the upper computer control system reads the weight data of the bioreactor weighed by the weighing scale, automatically calculates the substrate consumption rate according to the substrate concentration data. After completing the calculation of the substrate concentration consumption rate, the calculation of the predicted value of the substrate concentration in the next cycle is carried out, and the predicted value of the substrate concentration in the next cycle is compared with the feedback feeding concentration control target value c r If the predicted value of the substrate concentration in the next cycle is lower than the feedback feeding concentration control target value c r at this time, feeding is required in the next cycle. This cycle is defined as the feeding cycle, and the prediction of the feeding time point in this feeding cycle needs to be started; The feeding time point is calculated within the feeding cycle, that is, the specific time point when the substrate concentration is lower than the target concentration within the cycle. The relationship between the predicted value of the feeding time point within the feeding cycle of the fermentation process and the target feeding value is: Among them, is the predicted value of the feeding time point during the feeding cycle of the fermentation process, c n is the substrate concentration of this cycle, V n is the volume of the fermentation broth in the bioreactor of this cycle, c r is the target value of the feedback feeding concentration control, K is the feeding control range coefficient, u n is the substrate consumption rate between two cycles, t n is the time of this cycle; If feeding is required at the next monitoring time point, the volume data of the required feeding liquid will be transmitted to the peristaltic pump, and the peristaltic pump will automatically pump the feeding liquid into the bioreactor, thereby completing the feedback feeding of the fermentation process; The feeding program of the control system is mainly executed by the peristaltic pump. The control system calculates the running time of the peristaltic pump based on the existing data. After the running time of the peristaltic pump ends, the first fermentation process intelligent feedback feeding is completed until the detection cycle ends. The relationship between the running time and the existing data is: Among them, Δt n+1 is the operation time of the peristaltic pump, V′ n+1 is the volume of the fermentation broth in the bioreactor at the feeding time point, c′ n+1 is the substrate concentration in the bioreactor at the feeding time point, c r is the target value of the feedback feeding concentration control, K is the feeding control range coefficient, u′ n+1 is the substrate consumption rate from the feeding time point to this cycle, t f -t′ n+1 is the time difference between the feeding time point and the next cycle, c f is the concentration of the feeding solution, u p is the operation flow rate of the feeding peristaltic pump.

2. The control system for intelligent feedback feeding based on on-line detection of substrate concentration during fermentation according to claim 1, characterized in that, The upper computer control system preset the detection period F (h) of the fermentation process, the feedback feeding concentration control target value c r (g / L), the feeding control range coefficient K (%), the substrate feeding solution concentration c f (g / L), the flow rate u of the feeding peristaltic pump p (L / h). After setting the initial parameters, the biosensing analyzer will detect the substrate concentration according to the detection period F (h) and transmit the data to the control system. At the same time, the weighing scale will transmit the volume change of the bioreactor when the substrate concentration result comes out to the control system. After receiving the two groups of substrate concentration data, the control system automatically calculates the substrate concentration consumption rate u n . The relationship between the substrate concentration consumption rate and the substrate concentration data during the fermentation process is: Among them, u n is the substrate consumption rate between two cycles, c n-1 is the substrate concentration of the previous cycle, V n-1 is the volume of the fermentation broth in the bioreactor of the previous cycle, c n is the substrate concentration of this cycle, V n is the volume of the fermentation broth in the bioreactor of this cycle, t n -t n-1 is the time difference between this cycle and the previous cycle.

3. The control system for intelligent feedback feeding based on on-line detection of substrate concentration during fermentation according to claim 1, wherein If the predicted value of the substrate concentration in the next cycle is not lower than the feedback feeding concentration control target value c r then, in the next cycle, the biosensing analyzer measures the substrate concentration in the bioreactor, continues to calculate the substrate concentration consumption rate in the most recent two cycles and the predicted value of the substrate concentration in the next cycle, and compares it with the feedback feeding concentration control target value c r until the predicted value of the substrate concentration in a certain cycle is lower than the feedback feeding concentration control target value c r .

4. The control system for intelligent feedback feeding based on on-line detection of substrate concentration during fermentation according to claim 1, characterized in that, Predicted value c of substrate concentration in the next cycle of the fermentation process n+1 The relationship with the substrate concentration in the previous cycle is as follows: Among them, c n+1 is the predicted value of the substrate concentration in the next cycle, c n is the substrate concentration in this cycle, V n is the volume of the fermentation broth in the bioreactor in this cycle, u n is the substrate consumption rate between two cycles, V n+1 is the volume of the fermentation broth in the bioreactor in the next cycle, t n+1 -t n is the time difference between the next cycle and this cycle.

5. The control system for intelligent feedback feeding based on on-line detection of substrate concentration during fermentation according to claim 1, characterized in that, Entering the feeding cycle, the control system needs to calculate the new consumption rate based on the substrate concentration tested by the biosensor analyzer. The relationship between the consumption rate and the substrate concentration after the feeding cycle is: where u′ n+1 is the substrate consumption rate of the feeding time point relative to this cycle, c n is the substrate concentration of this cycle, V n is the volume of the fermentation broth in the bioreactor of this cycle, V′ n+1 is the volume of the fermentation broth in the bioreactor at the feeding time point, c′ n+1 is the substrate concentration in the bioreactor at the feeding time point, t′ n+1 -t n is the time difference between the feeding time point and this cycle.

6. The control system for intelligent feedback feeding based on on-line detection of substrate concentration during fermentation according to claim 5, wherein The microbial growth process is divided into four stages: lag phase, logarithmic growth phase, stable phase and decay phase. In the lag phase and the early logarithmic growth phase of the microorganism, the substrate concentration data in the bioreactor is tested using a biosensor analyzer, and a consumption rate metabolic curve is established. Based on this data, the substrate concentration data in the next cycle is predicted, and it is determined whether it is lower than the target feed concentration. The peristaltic pump is driven in time to feed the feed to prevent the microorganism from entering the decay phase, so that the microorganism can normally produce the target substance and improve production efficiency.

7. Use of the control system according to claim 1 for automatic feeding in a fermentation process of Saccharomyces cerevisiae.

Images