A method and system for scheduling and controlling the co-production of bioproteins and ethanol from syngas.

Abstract

This invention discloses a scheduling and control method and system for the co-production of bioproteins and ethanol from syngas, relating to the field of scheduling and control technology. The method includes: calculating the relative biomass bias of ethanol based on the operating data of the syngas co-production fermentation system; calculating a comprehensive evaluation quantity based on the relative biomass bias of ethanol; constructing a set of gas injection path rearrangement actions based on the configuration data of the controllable actuators of the fermentation reactor, and obtaining a structural intervention action library based on the set of gas injection path rearrangement actions; obtaining an updated pheromone matrix based on the comprehensive evaluation quantity and the structural intervention action library, and obtaining a target action sequence based on the updated pheromone matrix. This invention improves the ability to regulate complex gas-liquid distribution states.

Description

Technical Field

[0001] This invention relates to the field of scheduling and control technology, and in particular to a scheduling and control method and system for the co-production of bioproteins and ethanol from syngas. Background Technology

[0002] In industrial plants that co-produce ethanol and biological protein from syngas fermentation, syngas is mainly composed of carbon monoxide, hydrogen, and carbon dioxide. The upstream syngas source may be gasification gas or industrial tail gas. The composition of the inlet gas is prone to fluctuation and may carry trace impurities. At the same time, due to the enlarged scale of the fermentation reactor, microbubble groups, gas content variations, dissolved gas spatial gradients, and foam behavior are common in the tank. This results in significant spatial differences in the distribution and transfer of gas in the fermentation broth. In the co-production mode, ethanol production and cell growth are not strictly in the same direction. When gas utilization occurs more at the bubble interface and does not effectively enter the bulk phase to support cell growth, a structural imbalance of high ethanol and low protein is likely to occur. Unified scheduling and control are needed among adjustable links such as gas injection path, stirring and shearing, and reflux inlet position to maintain stable co-production.

[0003] Existing technologies often employ tracking control methods based on flow rate or component setpoints to address the aforementioned co-production imbalance problem. These methods focus on stabilizing the intake volume, intake components, or single process variables. However, when the bubble interface shell adheres, some gas is preferentially consumed at the interface, creating a gas utilization short circuit. This results in a situation where unit gas consumption no longer corresponds to unit protein formation, and the bias between ethanol and biomass persists. Consequently, control methods relying solely on setpoint tracking struggle to identify spatial separation caused by interface occupancy and fail to provide sequential scheduling strategies for structural interventions such as gas injection paths and shear sequences. Consequently, it becomes difficult to stably achieve the co-production target of ethanol and biological protein. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies in achieving stable co-production of ethanol and biological proteins, and to propose a scheduling and control method and system for co-producing biological proteins and ethanol from syngas.

[0005] To address the problems existing in the prior art, the present invention adopts the following technical solution: A method for controlling the co-production of bioproteins and ethanol using synthetic gas, comprising: S1. Calculate the relative biomass bias of ethanol based on the operating data of the syngas co-production fermentation system; S2. Calculate the comprehensive evaluation quantity based on the relative biomass bias of ethanol; S3. Based on the configuration data of the controllable actuators of the fermentation reactor, construct a set of gas injection path rearrangement actions, and obtain a structural intervention action library based on the set of gas injection path rearrangement actions. S4. Based on the comprehensive evaluation quantity and the structural intervention action library, the updated pheromone matrix is ​​obtained, and the target action sequence is obtained based on the updated pheromone matrix; S5. Based on the target action sequence, schedule and control the fermentation process of syngas co-producing bioproteins and ethanol.

[0006] Preferably, the relative biomass bias of ethanol is calculated based on the operating data of the syngas co-production fermentation system, including: Collect operational data from the syngas co-production fermentation system; Calculate the ethanol concentration increment based on the ethanol concentration in the operational data; Calculate the biomass increment based on the biomass indicators in the operational data; The ratio of the increase in ethanol concentration to the increase in biomass is calculated to obtain the relative biomass bias of ethanol.

[0007] Preferably, the comprehensive evaluation metric is calculated based on the relative biomass bias of ethanol, including: The apparent carbon monoxide conversion rate is calculated based on the volume fraction of carbon monoxide in the intake air and the volume fraction of carbon monoxide in the exhaust air from the operating data. The ratio between apparent carbon monoxide conversion rate and biomass increment was calculated to obtain the correlation quantity. Generate a sequence of correlation values ​​based on the correlation values; Based on the relative biomass bias of ethanol, a relative biomass bias sequence of ethanol was generated. The attachment state of the bubble interface shell was determined based on the ethanol relative biomass bias sequence and correlation sequence. Based on the attachment state of the bubble interface shell, the relative biomass bias sequence and correlation sequence of ethanol were screened to obtain the interface occupation class dataset. The remaining data points in the ethanol relative biomass bias sequence and the corresponding correlation data points were filtered to obtain a dataset of body phase growth. Shell the interface occupancy class data set to obtain a shelled interface marked data set; The body phase growth dataset is labeled with body phase growth characteristics to obtain a body phase growth labeled dataset. Count the shelled interface marker data set to obtain shelled interface statistics; Count the data set of body growth markers to obtain body growth statistics; Calculate the shelling bias coefficient based on the shelling interface statistics and the volume growth statistics; Based on the shelling bias coefficient, correlation analysis was performed on the correlation quantity sequence and the ethanol relative biomass bias sequence to obtain the comprehensive evaluation quantity.

[0008] Preferably, the set of gas injection path rearrangement actions is constructed, including: Obtain the configuration data of the controllable actuators in the fermentation reactor; Based on the set of gas injection path switching actuators in the controllable actuator configuration data, a set of gas injection path rearrangement actions is constructed.

[0009] Preferably, a structural intervention action library is obtained by rearranging the gas injection path action set, including: Based on the set of stirring actuators in the controllable actuator configuration data, a set of shearing sequence actions is constructed. Based on the set of backflow path switching actuators in the controllable actuator configuration data, construct a set of backflow access location rearrangement actions; The sets of gas injection path rearrangement actions, shear sequence actions, and reflux access position rearrangement actions are merged to obtain a structural intervention action library.

[0010] Preferably, the updated pheromone matrix is ​​obtained based on the comprehensive evaluation quantity and the structural intervention action library, including: Construct a pheromone matrix based on a structural intervention action library; Based on the pheromone matrix, generate the candidate action sequence for the current iteration; Based on the calculation rules of the comprehensive evaluation quantity, calculate the sequence evaluation quantity of the candidate action sequence in the current iteration; Based on the sequence evaluation of the candidate action sequence in the current iteration, and combined with the ant colony algorithm, the pheromone matrix is ​​updated to obtain the updated pheromone matrix.

[0011] Preferably, the target action sequence is obtained based on the updated pheromone matrix, including: Based on the updated pheromone matrix, generate the candidate action sequence for the next iteration; Based on the calculation rules of the comprehensive evaluation quantity, calculate the sequence evaluation quantity of the candidate action sequence for the next iteration; Based on the sequence evaluation value of the candidate action sequences in the next iteration, the candidate action sequences in the next iteration are compared and filtered to obtain the target action sequence.

[0012] Preferably, the fermentation process of syngas-co-produced bioproteins and ethanol is scheduled and controlled according to the target action sequence, including: Convert the target action sequence into a control command sequence; The fermentation process of syngas co-producing bioproteins and ethanol is scheduled and controlled according to the sequence of control commands.

[0013] To address the aforementioned problems, the present invention also provides a syngas-based co-production biological protein and ethanol scheduling and control system, the system comprising: The bias calculation module calculates the relative biomass bias of ethanol based on the operating data of the syngas co-production fermentation system. The evaluation quantity generation module calculates the comprehensive evaluation quantity based on the relative biomass bias of ethanol. The action library construction module constructs a set of gas injection path rearrangement actions based on the configuration data of the controllable actuators of the fermentation reactor, and obtains a structural intervention action library based on the set of gas injection path rearrangement actions. The action sequence solving module obtains an updated pheromone matrix based on the comprehensive evaluation quantity and the structural intervention action library, and obtains the target action sequence based on the updated pheromone matrix; The scheduling and control execution module schedules and controls the fermentation process of syngas co-produced bioproteins and ethanol according to the target action sequence.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention constructs a comprehensive evaluation quantity by constructing an ethanol relative biomass bias and combining it with the carbon monoxide conversion relationship, thereby achieving a unified quantitative description of the gas utilization and product distribution state. It can identify the operating state in which gas is preferentially consumed at the interface and does not enter the bulk phase, thus overcoming the problem that relying solely on a single process variable is insufficient to reflect the true fermentation state and improving the ability to identify deviations in the operation of the co-production process.

[0015] 2. This invention constructs a structural intervention action library that includes gas injection path rearrangement, shear sequence adjustment, and reflux access position rearrangement. It combines pheromone matrix and ant colony algorithm to iteratively optimize the action sequence, realizing coordinated scheduling among multiple execution links. This transforms the control method from single-variable regulation to multi-action combination optimization, thereby improving the ability to regulate complex gas-liquid distribution states.

[0016] 3. This invention schedules and controls the fermentation process based on the optimized target action sequence, adjusts the gas entry path and transfer process in terms of spatial distribution, weakens the preferential consumption phenomenon at the interface and promotes the uniform distribution of gas to the bulk phase, so that the correspondence between gas utilization and cell growth is re-established, thereby stably achieving the co-production balance of ethanol and biological protein and improving the overall production stability. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a schematic flowchart of a method for scheduling and controlling the co-production of bioproteins and ethanol using synthetic gas, provided in an embodiment of the present invention. Figure 2This is a functional block diagram of a synthetic gas co-production biological protein and ethanol scheduling and control system provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] This embodiment provides a method for the scheduling and control of synthetic gas-generated bioproteins and ethanol. See [link to relevant documentation]. Figure 1 Specifically, including: S1. Calculate the relative biomass bias of ethanol based on the operating data of the syngas co-production fermentation system; In an embodiment of the present invention, the ethanol relative biomass bias is calculated based on the operating data of the syngas co-production fermentation system, including: Collect operational data from the syngas co-production fermentation system; Specifically, the gas flow meter, pressure transmitter, and temperature transmitter on the syngas inlet main pipe are installed in the stable flow section and their equipment numbers are registered. The carbon monoxide concentration detector, hydrogen concentration detector, and carbon dioxide concentration detector are arranged on the inlet sampling pipeline after water and mist removal treatment and calibrated. The liquid level meter, temperature meter, pH meter, stirring speed feedback device, tank pressure meter, feed flow meter, discharge flow meter, and tail gas component detector in the fermenter are connected to the data acquisition terminal one by one. The online ethanol concentration analyzer is set in the fermentation broth circulation pipeline and the sampling position is ensured to be in the circulation stable zone. The output signals of each detection device are connected to the corresponding input channels of the acquisition terminal according to the signal type, and the correspondence between the channel number and the physical quantity is established. The acquisition terminal periodically reads each input channel according to a unified time reference and records the acquisition time. The acquired raw signals are converted into corresponding physical quantity values ​​according to the range relationship. After time alignment and integrity verification of the data at each moment, they are written into the operation data record table to form the syngas co-production fermentation operation data for subsequent calculations.

[0020] Calculate the ethanol concentration increment based on the ethanol concentration in the operational data; Operational data refers to real-time or historical data collected from various detection units of the syngas co-production fermentation system, including gas phase components, liquid phase components, and the operating status of the reaction system; ethanol concentration refers to the content of ethanol in a unit volume of liquid phase in the fermentation broth, used to reflect the product generation level; ethanol concentration increment refers to the change in ethanol concentration within adjacent scheduling cycles, used to characterize the change in ethanol generation rate.

[0021] Specifically, when calculating the ethanol concentration increment based on the ethanol concentration in the operational data, all ethanol concentration records are read from the operational data table in ascending order of acquisition time. The acquisition time interval between two adjacent records is checked to ensure it is equal to one second. For record segments with an interval not equal to one second, the records are reordered according to the actual time, and records with duplicate times are removed. In the remaining records, the current ethanol concentration value is recorded as the current concentration value, and the ethanol concentration value adjacent to it but acquired one second earlier is recorded as the previous concentration value. The ethanol concentration increment corresponding to the current moment is obtained by subtracting the previous concentration value from the current concentration value. This ethanol concentration increment is then... The unit is determined as the concentration change per liter per second and a corresponding relationship is established with the current time. The subtraction operation is repeated for all adjacent records to form a complete ethanol concentration increment sequence. If there is a missing ethanol concentration, a time jump, or an abnormal acquisition status marking at any time, the ethanol concentration increment at that time is marked as invalid and the reason for invalidity is recorded separately. The valid ethanol concentration increments are written into the result table in the order of acquisition time. Each row in the result table contains the current time, the previous time, the current concentration value, the previous concentration value, the time difference between the two times, and the ethanol concentration increment value. Thus, the ethanol concentration increment calculated based on the running data is obtained.

[0022] Calculate the biomass increment based on the biomass indicators in the operational data; Specifically, the biomass index data corresponding to two adjacent scheduling cycles are read, and the biomass index of the current scheduling cycle is paired with the biomass index of the previous scheduling cycle in chronological order. The difference between the two paired biomass indices is calculated to obtain the biomass change value of the current scheduling cycle relative to the previous scheduling cycle, and this biomass change value is determined as the biomass increment. Here, the biomass index is the data characterizing the microbial cell content in the fermentation broth, and the biomass increment is the change in microbial cell content between two adjacent scheduling cycles.

[0023] The ratio of the increase in ethanol concentration to the increase in biomass is calculated to obtain the relative biomass bias of ethanol.

[0024] Biomass index refers to the content or concentration of microbial cells in the fermentation broth, used to reflect the growth status of microorganisms; biomass increment refers to the change in biomass index within adjacent scheduling cycles, used to characterize the change in microbial growth rate; ethanol relative biomass bias refers to the ratio between the ethanol concentration increment and the biomass increment, used to reflect the relative distribution between product generation and cell growth during fermentation.

[0025] Specifically, the ethanol concentration increment data and biomass increment data corresponding to the current scheduling cycle are read. The ethanol concentration increment is used as the dividend and the biomass increment is used as the divisor to perform a division operation to obtain the ratio of the ethanol concentration increment to the biomass increment. This ratio is determined as the ethanol relative biomass bias. Here, the ethanol concentration increment is the change in ethanol concentration between two adjacent scheduling cycles, and the ethanol relative biomass bias is data used to characterize the relative relationship between changes in ethanol production and changes in microbial growth.

[0026] S2. Calculate the comprehensive evaluation quantity based on the relative biomass bias of ethanol; In embodiments of the present invention, the comprehensive evaluation metric is calculated based on the relative biomass bias of ethanol, including: The apparent carbon monoxide conversion rate is calculated based on the volume fraction of carbon monoxide in the intake air and the volume fraction of carbon monoxide in the exhaust air from the operating data. Specifically, the volume fraction data of carbon monoxide in the intake air and the volume fraction data of carbon monoxide in the exhaust air within the same scheduling cycle are read. The volume fraction of carbon monoxide in the intake air is subtracted from the volume fraction of carbon monoxide in the exhaust air to obtain the difference in carbon monoxide volume fraction. Then, this difference is divided by the volume fraction of carbon monoxide in the intake air to obtain the apparent carbon monoxide conversion rate. Here, the volume fraction of carbon monoxide in the intake air represents the volume proportion of carbon monoxide in the gas entering the fermentation reactor, the volume fraction of carbon monoxide in the exhaust air represents the volume proportion of carbon monoxide in the gas leaving the fermentation reactor, and the apparent carbon monoxide conversion rate represents the relative degree to which carbon monoxide is consumed within the scheduling cycle.

[0027] The ratio between apparent carbon monoxide conversion rate and biomass increment was calculated to obtain the correlation quantity. Specifically, the apparent carbon monoxide conversion rate data and biomass increment data corresponding to the current scheduling cycle are read. The apparent carbon monoxide conversion rate is used as the dividend, and the biomass increment is used as the divisor for division to obtain the ratio of apparent carbon monoxide conversion rate to biomass increment. This ratio is then determined as the correlation quantity. Here, biomass increment represents the change in microbial cell content between two adjacent scheduling cycles, and the correlation quantity represents the correspondence between changes in carbon monoxide consumption and changes in cell growth.

[0028] Generate a sequence of correlation values ​​based on the correlation values; The volume fraction of carbon monoxide in the inlet gas refers to the volume proportion of carbon monoxide in the gas mixture entering the fermenter. The volume fraction of carbon monoxide in the outlet gas refers to the volume proportion of carbon monoxide in the exhaust gas leaving the fermenter. The apparent carbon monoxide conversion rate refers to the degree of carbon monoxide consumption reflected by the changes in carbon monoxide content in the inlet and outlet gas. The biomass increment refers to the change in microbial cell content within adjacent scheduling cycles. The correlation quantity refers to the ratio between the apparent carbon monoxide conversion rate and the biomass increment, used to characterize the degree of correspondence between gas consumption and cell growth. The correlation quantity sequence refers to the data set formed by arranging the correlation quantities in chronological order within multiple scheduling cycles.

[0029] Specifically, the associated quantity data obtained in chronological order within multiple consecutive scheduling cycles is read, and the associated quantities corresponding to each scheduling cycle are arranged sequentially according to the collection time to form an ordered data set that corresponds one-to-one with each scheduling cycle. This ordered data set is then defined as the associated quantity sequence, where the associated quantity sequence represents the continuous arrangement result of the associated quantities changing over time within multiple scheduling cycles.

[0030] Based on the relative biomass bias of ethanol, a relative biomass bias sequence of ethanol was generated. The ethanol relative biomass bias sequence refers to the data set formed by arranging the ethanol relative biomass bias in chronological order over multiple scheduling cycles.

[0031] Specifically, the relative biomass bias data of ethanol obtained in chronological order within multiple consecutive scheduling cycles are read. The relative biomass bias of ethanol corresponding to each scheduling cycle is arranged sequentially according to the collection time to form an ordered data set that corresponds one-to-one with each scheduling cycle. This ordered data set is determined as the relative biomass bias sequence of ethanol. The relative biomass bias of ethanol is the ratio of the increase in ethanol concentration to the increase in biomass. The relative biomass bias sequence of ethanol represents the continuous arrangement of the changes in ethanol production relative to the changes in cell growth within multiple scheduling cycles.

[0032] The attachment state of the bubble interface shell was determined based on the ethanol relative biomass bias sequence and correlation sequence. Specifically, the current data in the ethanol relative biomass bias sequence and the corresponding data in the associated quantity sequence within the same scheduling period are read. The current data and the corresponding data are compared item by item. The operating state in which the ethanol relative biomass bias is greater than or equal to the corresponding associated quantity is determined as the bubble interface shell attachment state, and the operating state in which the ethanol relative biomass bias is less than the corresponding associated quantity is determined as the non-bubble interface shell attachment state. The bubble interface shell attachment state indicates that the microorganisms in the fermentation broth form attachment and aggregation at the bubble gas-liquid interface, and the change in ethanol production is dominant relative to the change in cell growth.

[0033] Based on the attachment state of the bubble interface shell, the relative biomass bias sequence and correlation sequence of ethanol were screened to obtain the interface occupation class dataset. Specifically, the relative biomass bias sequence and the associated quantity sequence of ethanol are read and a one-to-one correspondence is established according to the same scheduling cycle. Based on the determination result of the shell attachment status of the bubble interface, the presence of interface attachment occupation in each scheduling cycle is identified one by one. The relative biomass bias data and associated quantity data of ethanol corresponding to the scheduling cycle that is determined to have interface attachment occupation are extracted at the same time and combined and arranged according to the original time order. All paired data that meet the determination of interface attachment occupation are summarized to form an interface occupation data set composed of relative biomass bias data of ethanol and associated quantity data.

[0034] The remaining data points in the ethanol relative biomass bias sequence and the corresponding correlation data points were filtered to obtain a dataset of body phase growth. Specifically, data from the ethanol relative biomass bias sequence that are not included in the interface occupancy data set are read, and corresponding correlation data in the same scheduling period as the data not included in the interface occupancy data set are extracted from the correlation sequence. The extracted ethanol relative biomass bias data and corresponding correlation data are paired one by one according to the scheduling period to form a volume growth data set. The volume growth data set represents the data combination of ethanol generation changes and gas consumption changes relative to cell growth changes corresponding to the non-bubble interface shell attachment state.

[0035] Shell the interface occupancy class data set to obtain a shelled interface marked data set; Specifically, each set of data in the interface occupancy data set is read, the scheduling cycle corresponding to each set of data is identified, and shelled interface identifiers representing the interface attachment and occupancy relationship are added to the set of data, so that each set of data corresponds to a clear shelled interface mark result. All the data that have completed the shelled interface mark are arranged in the order of the scheduling cycle to form a shelled interface mark data set. The shelled interface mark data set represents the set of data that has been confirmed to belong to the interface attachment and occupancy category.

[0036] The body phase growth dataset is labeled with body phase growth characteristics to obtain a body phase growth labeled dataset. Specifically, each group of data in the somatic growth category dataset is read, the scheduling cycle corresponding to each group of data is identified, and a somatic growth identifier is added to the data group to represent the somatic cell growth relationship, so that each group of data corresponds to a clear somatic growth label result. All data that have completed somatic growth labeling are arranged in the order of scheduling cycles to form a somatic growth label dataset, where the somatic growth label dataset represents the dataset that has been confirmed to belong to the somatic cell growth category.

[0037] The bubble interface shell attachment state refers to the operational state in which some microorganisms in the fermentation broth tend to attach to the bubble gas-liquid interface and form local shell aggregations. The interface occupation data set refers to the data set corresponding to the bubble interface shell attachment state. The volume growth data set refers to the data set that does not show the bubble interface shell attachment state but mainly reflects the cell growth in the main area of ​​the fermentation broth. The shell-marked interface data set refers to the data set formed after shell-marking the interface occupation data set. The volume growth marked data set refers to the data set formed after volume growth marking the volume growth data set.

[0038] Count the shelled interface marker data set to obtain shelled interface statistics; Specifically, all marked data in the shell interface marking data set are read, and the shell interface marking results are identified one by one in the order of the scheduling cycle. Each data with a shell interface marking is accumulated and recorded until all data in the shell interface marking data set is traversed. The number of accumulated data is determined as the shell interface statistic, where the shell interface statistic represents the number of data marked as the interface attachment category.

[0039] Count the data set of body growth markers to obtain body growth statistics; Specifically, all labeled data in the phase growth label data set are read, and the phase growth label results are identified one by one according to the scheduling cycle. Each data with a phase growth label is accumulated and recorded until all data in the phase growth label data set is traversed. The number of accumulated data is determined as the phase growth statistic, where the phase growth statistic represents the number of data labeled as phase bacterial growth category.

[0040] Calculate the shelling bias coefficient based on the shelling interface statistics and the volume growth statistics; The shelling interface statistic refers to the number of data in the shelling interface labeled dataset, the volume growth statistic refers to the number of data in the volume growth labeled dataset, and the shelling bias coefficient refers to the proportional relationship between the shelling interface statistic and the volume growth statistic, which is used to characterize the degree of dominance of interface attachment behavior relative to volume growth behavior.

[0041] Specifically, the shell-forming interface statistics and the bulk cell growth statistics are read, and the shell-forming interface statistics and the bulk cell growth statistics are summed to obtain the total metric. Then, the shell-forming interface statistics are divided by the total metric to obtain the proportion of the shell-forming interface statistics in all statistics. This proportion is determined as the shell-forming bias coefficient, which represents the degree of bias of the interface attachment and occupation behavior relative to the bulk cell growth behavior.

[0042] Based on the shelling bias coefficient, correlation analysis was performed on the correlation quantity sequence and the ethanol relative biomass bias sequence to obtain the comprehensive evaluation quantity.

[0043] The comprehensive evaluation quantity refers to the evaluation result obtained by combining the correlation quantity sequence, the ethanol relative biomass bias sequence, and the shelling bias coefficient, which is used to reflect the comprehensive relationship between gas utilization, ethanol production and cell growth in the current fermentation process.

[0044] Specifically, the shelling bias coefficient data, correlation quantity sequence data, and ethanol relative biomass bias sequence data within the same scheduling range are read. According to the scheduling cycle, each sequence value in the correlation quantity sequence is paired with the corresponding sequence value in the ethanol relative biomass bias sequence one by one. The evaluation value of the corresponding scheduling cycle is obtained by summing the paired data for each set. Then, the shelling bias coefficient is multiplied by the evaluation value of each scheduling cycle one by one to obtain the weighted evaluation result of each scheduling cycle. The weighted evaluation results corresponding to all scheduling cycles are arranged in chronological order and summarized to obtain a comprehensive evaluation quantity that reflects the correspondence between the correlation quantity change, ethanol generation change, and interface attachment ratio.

[0045] It should be noted that the degree of carbon monoxide conversion during fermentation reflects the utilization of gaseous substrates, while the ethanol-to-biomass bias reflects the distribution of carbon flow between product generation and cell synthesis. These two factors correspond to substrate consumption and metabolic distribution, respectively. The occupancy state formed by shell adhesion at the bubble interface alters the gas transport pathway, making carbon monoxide more likely to be consumed preferentially near the interface and changing the metabolic flow direction. Therefore, by combining the correlation quantity with the ethanol-to-biomass bias, the coupling relationship between substrate utilization and metabolic distribution can be characterized simultaneously. By introducing a shell bias coefficient to weight the combined result, the influence of interface adhesion behavior on gas utilization efficiency and product generation bias can be incorporated into a unified evaluation, thereby obtaining an evaluation result that can comprehensively reflect the interaction between gas conversion, cell growth, and product generation.

[0046] It should be noted that when the shell attachment state exists at the microbubble interface, some gas is preferentially consumed at the bubble interface without effectively entering the bulk phase to participate in cell growth, which disrupts the correspondence between gas conversion and protein formation. In this case, relying solely on the correlation coefficient is insufficient to reflect the true bulk phase utilization. While the ethanol relative biomass bias can characterize changes in the direction of metabolite flow, it cannot reflect the distribution of gas utilization pathways alone. Therefore, by introducing the shell attachment bias coefficient to quantify the proportion of interface attachment behavior and using it as a weight to conduct a unified correlation analysis between the correlation coefficient sequence and the ethanol relative biomass bias sequence, we can simultaneously characterize the distribution of gas consumption locations, the direction of metabolic distribution, and the degree of interface occupancy, thereby obtaining a comprehensive evaluation quantity that can reflect the relationship between gas utilization efficiency and product distribution.

[0047] S3. Based on the configuration data of the controllable actuators of the fermentation reactor, construct a set of gas injection path rearrangement actions, and obtain a structural intervention action library based on the set of gas injection path rearrangement actions. In an embodiment of the present invention, constructing a set of gas injection path rearrangement actions includes: Obtain the configuration data of the controllable actuators in the fermentation reactor; A fermentation reactor is a container used for syngas fermentation, which forms a space where gas comes into contact with the fermentation broth and undergoes mass transfer and biochemical transformation. Controllable actuator configuration data refers to the type, quantity, connection relationship, and control interface information of the actuators related to the operation and regulation of the fermentation reactor.

[0048] Specifically, the equipment identification information, installation location data, connection pipeline data, control port data, and action status feedback data corresponding to each actuator connected to the fermentation reactor are read and organized according to the connection relationship of the actuator in the gas inlet pipeline, gas distribution pipeline, and gas injection branch, forming a data set to characterize the controllable range, switchable path, and responsive action of each actuator. Among them, the controllable actuator configuration data includes actuator data that can change the gas flow direction, gas distribution position, or gas passage connection relationship.

[0049] Based on the set of gas injection path switching actuators in the controllable actuator configuration data, a set of gas injection path rearrangement actions is constructed.

[0050] The gas injection path switching actuator set refers to a set of actuators installed on the gas inlet pipeline, gas distribution pipeline, or gas injection branch and capable of changing the gas flow path. These actuators include valves, switches, or drive components related to changes in the gas passage connection state. The gas injection path rearrangement action set refers to a combination of various executable actions generated based on the gas injection path switching actuator set, used to change the path, entry position, or distribution method of gas entering the fermentation reactor, thereby changing the gas distribution state and transmission process in the fermentation broth.

[0051] Specifically, the actuator data related to the gas injection path switching is extracted from the controllable actuator configuration data. The gas inlet position, connecting pipe section and gas passage formed after switching are identified for each actuator. Each actuator action state is matched with the corresponding gas flow path to obtain multiple sets of different gas injection path switching results. The gas injection path switching results are then organized into a gas injection path rearrangement action set according to the correspondence between actuator action and gas path change.

[0052] In embodiments of the present invention, a structural intervention action library is obtained based on a set of gas injection path rearrangement actions, including: Based on the set of stirring actuators in the controllable actuator configuration data, a set of shearing sequence actions is constructed. The set of stirring actuators refers to the set of actuators installed in the fermentation reactor to drive the stirring device. The set of shearing sequence actions refers to the combination of various shearing action changes formed by the stirring actuators at different speeds or operating states, which are used to change the fluid shear intensity and bubble breakage behavior in the fermentation broth.

[0053] Specifically, the equipment identification data, speed control data, action start and stop data, and installation layer data corresponding to each mixing actuator are read. The speed change patterns formed by each mixing actuator under different operating conditions are arranged in chronological order to obtain an action combination composed of multiple mixing actions in sequence. The mixing actuator, action sequence, and speed change content corresponding to each action combination are recorded to form a shearing sequence action set.

[0054] Based on the set of backflow path switching actuators in the controllable actuator configuration data, construct a set of backflow access location rearrangement actions; The reflux path switching actuator set refers to the set of actuators installed in the reflux pipeline to change the flow path of reflux gas or liquid. The reflux access position rearrangement action set refers to the action combination of adjusting the state of the reflux path switching actuator to allow the reflux material to enter different positions in the fermentation reactor.

[0055] Specifically, the device identification data, pipeline connection data, return inlet location data, and action status switching data corresponding to the return path switching actuator are read. The return path and return entry position corresponding to each return path switching actuator in different action states are identified. The return access position change results formed in different action states are matched one-to-one with the corresponding actuator actions to obtain multiple sets of return access position change actions. The sets of return access position change actions are then organized into a set of return access position rearrangement actions.

[0056] The sets of gas injection path rearrangement actions, shear sequence actions, and reflux access position rearrangement actions are merged to obtain a structural intervention action library.

[0057] The structural intervention action library refers to the set of actions formed by merging the sets of gas injection path rearrangement actions, shear sequence actions, and return access position rearrangement actions, which are used to make overall adjustments to gas distribution, liquid flow, and gas-liquid contact state.

[0058] Specifically, each injection action in the injection path rearrangement action set, each shearing action in the shearing sequence action set, and each return action in the return access position rearrangement action set are read. Each action is uniformly numbered according to its source, content, and corresponding actuator. All numbered actions are then summarized and arranged into the same action set to form a structural intervention action library that includes injection path adjustment actions, shearing adjustment actions, and return access position adjustment actions.

[0059] S4. Based on the comprehensive evaluation quantity and the structural intervention action library, the updated pheromone matrix is ​​obtained, and the target action sequence is obtained based on the updated pheromone matrix; In an embodiment of the present invention, an updated pheromone matrix is ​​obtained based on a comprehensive evaluation quantity and a structural intervention action library, including: Construct a pheromone matrix based on a structural intervention action library; Specifically, all action data in the structural intervention action library are read, and the action number, action source, and action execution order of each action are identified. The possible sequential connections between any two actions are listed one by one, and rows and columns in a matrix are established according to the correspondence between the previous and subsequent actions, so that each position in the matrix corresponds to a set of specific action connections. Each set of action connections is then written into the corresponding matrix position to form a data matrix that represents the connection relationships between actions, and this data matrix is ​​determined as the pheromone matrix.

[0060] Based on the pheromone matrix, generate the candidate action sequence for the current iteration; The pheromone matrix refers to a set of data used to record the degree of influence of each action on the fermentation operation results during the historical scheduling process. Each element corresponds to a specific action or the connection between actions and reflects the degree of its selection tendency. The candidate action sequence refers to a combination of actions formed by selecting several actions from the structural intervention action library and arranging them in a certain order. These actions are executed sequentially during a scheduling process to change the fermentation operation state.

[0061] Specifically, the position data of each matrix in the pheromone matrix is ​​read, the starting action is selected from the structural intervention action library according to the action connection relationship, and the next action is determined one by one according to the subsequent action connection relationship corresponding to the starting action in the pheromone matrix. The determined actions are arranged in order to form a set of action sequences. The remaining actions not included in the set of action sequences are read and arranged in the same way to obtain multiple sets of action sequences composed of actions in the structural intervention action library. These multiple sets of action sequences are determined as candidate action sequences for the current iteration.

[0062] Based on the calculation rules of the comprehensive evaluation quantity, calculate the sequence evaluation quantity of the candidate action sequence in the current iteration; The current iteration refers to a round in the process of repeatedly selecting and evaluating action sequences. The sequence evaluation quantity refers to the numerical result obtained by quantifying the impact of the action sequence on gas utilization, ethanol production and cell growth changes during fermentation. The calculation rule of the comprehensive evaluation quantity refers to the method of combining the correlation quantity with the ethanol relative biomass bias and combining it with the shelling bias coefficient to obtain the evaluation result.

[0063] Specifically, the system reads the action execution result data corresponding to each group of candidate action sequences in the current iteration, and reads the correlation data, ethanol relative biomass bias data, and shelling bias coefficient data within the scheduling period corresponding to each candidate action sequence. The system sums the correlation data and ethanol relative biomass bias data corresponding to the same candidate action sequence to obtain the evaluation value for each scheduling period. Then, the system multiplies the evaluation value of each scheduling period with the corresponding shelling bias coefficient to obtain the weighted evaluation result for each scheduling period. Finally, the system summarizes all the weighted evaluation results corresponding to the same candidate action sequence to obtain the sequence evaluation quantity corresponding to the candidate action sequence. The system records the sequence evaluation quantities of all candidate action sequences according to the candidate action sequence number.

[0064] Based on the sequence evaluation of the candidate action sequence in the current iteration, and combined with the ant colony algorithm, the pheromone matrix is ​​updated to obtain the updated pheromone matrix.

[0065] Ant colony optimization refers to an iterative optimization method that simulates multiple path selection processes and strengthens or weakens the selection tendency between each action based on the evaluation results, thereby gradually adjusting the action selection probability to obtain a better action sequence. The updated pheromone matrix refers to the data set obtained after adjusting the degree of tendency of each action according to the sequence evaluation after the current iteration.

[0066] Specifically, all candidate action sequences and their corresponding sequence evaluation values ​​are read within the current iteration. The connection relationships between adjacent actions in each candidate action sequence are identified one by one. All action connections contained in each candidate action sequence are mapped to corresponding positions in the pheromone matrix according to the order of the actions. For each candidate action sequence, its pheromone contribution to the corresponding action connection relationship is determined based on its sequence evaluation value. Candidate action sequences with better sequence evaluation values ​​generate larger pheromone increments at their corresponding positions in the pheromone matrix, while candidate action sequences with poorer sequence evaluation values ​​generate smaller pheromone increments at their corresponding positions in the pheromone matrix. A volatilization process is performed on all original matrix positions in the pheromone matrix, causing the pheromones corresponding to action connections that are not repeatedly reinforced by the current iteration's candidate action sequences to weaken during the update process. Then, the pheromone increments corresponding to each candidate action sequence are superimposed onto the corresponding positions of the volatilized pheromone matrix, completing the cumulative update of the pheromone matrix by all candidate action sequences. The update result of any matrix position is determined by the original pheromone at that matrix position, the volatilization result corresponding to that matrix position, and the pheromone increment formed by all candidate action sequences at that matrix position in the current iteration. Finally, the matrix after all matrix positions have been updated is determined as the updated pheromone matrix.

[0067] In an embodiment of the present invention, the target action sequence is obtained based on the updated pheromone matrix, including: Based on the updated pheromone matrix, generate the candidate action sequence for the next iteration; Specifically, the updated pheromone matrix reads all matrix position data, identifies the connection relationship between the previous and next actions corresponding to each matrix position, selects the starting action from the structural intervention action library based on the action connection strength recorded in each matrix position, and then selects the next action one by one according to the connection relationship of the subsequent actions corresponding to the starting action in the updated pheromone matrix. The selected actions are arranged in sequence to form a set of action sequences, and the connection relationship reading and sequential arrangement are repeated for the remaining actions not included in the set of action sequences to obtain multiple sets of action sequences composed of actions from the structural intervention action library. These multiple sets of action sequences are determined as candidate action sequences for the next iteration.

[0068] Based on the calculation rules of the comprehensive evaluation quantity, calculate the sequence evaluation quantity of the candidate action sequence for the next iteration; The candidate action sequence for the next iteration refers to multiple action permutations formed under the guidance of the updated pheromone matrix, which are used to represent the action schemes that can be executed in the new scheduling round. The sequence evaluation quantity refers to the quantitative result of the impact of each candidate action sequence on gas utilization, ethanol production and cell growth changes during fermentation, which is used to measure the quality of the action sequence.

[0069] Specifically, the action execution result data corresponding to each group of candidate action sequences in the next iteration is read, and the correlation data, ethanol relative biomass bias data, and shelling bias coefficient data corresponding to each candidate action sequence within the scheduling period are read. The correlation data and ethanol relative biomass bias data corresponding to the same candidate action sequence are summed to obtain the evaluation value of each scheduling period. Then, the evaluation value of each scheduling period is multiplied by the corresponding shelling bias coefficient to obtain the weighted evaluation result of each scheduling period. All weighted evaluation results corresponding to the same candidate action sequence are summarized to obtain the sequence evaluation quantity corresponding to the candidate action sequence. The sequence evaluation quantities of all candidate action sequences are recorded one by one according to the candidate action sequence number.

[0070] Based on the sequence evaluation value of the candidate action sequences in the next iteration, the candidate action sequences in the next iteration are compared and filtered to obtain the target action sequence.

[0071] The target action sequence refers to the optimal action arrangement and combination determined by comparing and filtering the sequence evaluation values ​​of each candidate action sequence in the same iteration, which is used as the control basis for actual scheduling and execution.

[0072] Specifically, read all candidate action sequences and the sequence evaluation data corresponding to each candidate action sequence in the next iteration, establish a correspondence table between candidate action sequences and sequence evaluation values ​​according to the candidate action sequence number, compare all sequence evaluation values ​​under the same evaluation caliber, identify the candidate action sequence with the best sequence evaluation value, and extract the action number, action order and action content corresponding to the candidate action sequence. Arrange the extracted action number, action order and action content according to the execution order to form a complete action execution sequence, and determine the action execution sequence as the target action sequence. S5. Based on the target action sequence, schedule and control the fermentation process of syngas co-producing bioproteins and ethanol.

[0073] In embodiments of the present invention, the fermentation process of syngas-co-produced bioproteins and ethanol is scheduled and controlled according to a target action sequence, including: Convert the target action sequence into a control command sequence; Specifically, all action data in the target action sequence is read, and the action number, action sequence, action content, and corresponding actuator for each action are identified. The gas injection path rearrangement action in the target action sequence is converted into the on / off control command or path switching control command of the corresponding gas injection path switching actuator. The shearing sequence action is converted into the speed adjustment command and start / stop control command of the corresponding stirring actuator. The return flow inlet position rearrangement action is converted into the switching control command of the corresponding return flow path switching actuator. The converted control content of each actuator is arranged in sequence according to the action sequence in the target action sequence to form a control command sequence that corresponds one-to-one with each actuator and can be directly received by the actuator.

[0074] The fermentation process of syngas co-producing bioproteins and ethanol is scheduled and controlled according to the sequence of control commands.

[0075] Specifically, the system reads each control command from the control command sequence and, according to their order, sends them sequentially to their corresponding actuators. This causes the gas injection path switching actuator to change the path and entry position of the gas into the fermentation reactor, the stirring actuator to change the stirring action and shear intensity, and the reflux path switching actuator to change the inlet position of the reflux material. Through these actuators, the system coordinates and regulates the gas distribution, liquid flow, and gas-liquid contact state during fermentation, thereby completing the scheduling and control of the syngas-co-produced bioprotein and ethanol fermentation process.

[0076] It should be noted that when the microbubble interface shell is attached, the gas is preferentially consumed at the interface and does not effectively enter the bulk phase to participate in cell growth. This leads to a spatial separation between the gas utilization path and the bulk phase growth path, resulting in a persistently high ethanol production relative to protein formation. Therefore, it is necessary to coordinate the gas injection path, stirring and shearing, and reflux inlet position through a target action sequence to change the gas entry position and delivery path in terms of spatial distribution, weaken the preferential consumption at the interface, and promote the uniform distribution of gas to the bulk phase. This will restore the correspondence between gas utilization and cell growth, and achieve a balance between the co-production of ethanol and biological protein.

[0077] like Figure 2 The diagram shown is a functional block diagram of a synthetic gas co-production biological protein and ethanol scheduling and control system provided in an embodiment of the present invention.

[0078] In this embodiment, the functions of each module / unit are as follows: The bias calculation module calculates the relative biomass bias of ethanol based on the operating data of the syngas co-production fermentation system. The evaluation quantity generation module calculates the comprehensive evaluation quantity based on the relative biomass bias of ethanol. The action library construction module constructs a set of gas injection path rearrangement actions based on the configuration data of the controllable actuators of the fermentation reactor, and obtains a structural intervention action library based on the set of gas injection path rearrangement actions. The action sequence solving module obtains an updated pheromone matrix based on the comprehensive evaluation quantity and the structural intervention action library, and obtains the target action sequence based on the updated pheromone matrix; The scheduling and control execution module schedules and controls the fermentation process of syngas co-produced bioproteins and ethanol according to the target action sequence.

[0079] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for scheduling and controlling the co-production of bioproteins and ethanol using synthetic gas, characterized in that, Includes the following steps: S1. Calculate the relative biomass bias of ethanol based on the operating data of the syngas co-production fermentation system; S2. Calculate the comprehensive evaluation quantity based on the relative biomass bias of ethanol; S3. Based on the configuration data of the controllable actuators of the fermentation reactor, construct a set of gas injection path rearrangement actions, and obtain a structural intervention action library based on the set of gas injection path rearrangement actions. S4. Based on the comprehensive evaluation quantity and the structural intervention action library, the updated pheromone matrix is ​​obtained, and the target action sequence is obtained based on the updated pheromone matrix; S5. Based on the target action sequence, schedule and control the fermentation process of syngas co-producing bioproteins and ethanol.

2. The method for scheduling and controlling the co-production of bioproteins and ethanol using synthetic gas according to claim 1, characterized in that, Based on the operational data of the syngas co-production fermentation system, the relative biomass bias of ethanol was calculated, including: Collect operational data from the syngas co-production fermentation system; Calculate the ethanol concentration increment based on the ethanol concentration in the operational data; Calculate the biomass increment based on the biomass indicators in the operational data; The ratio of the increase in ethanol concentration to the increase in biomass is calculated to obtain the relative biomass bias of ethanol.

3. The method for scheduling and controlling the co-production of bioproteins and ethanol using synthetic gas according to claim 2, characterized in that, Based on the relative biomass bias of ethanol, a comprehensive evaluation metric is calculated, including: The apparent carbon monoxide conversion rate is calculated based on the volume fraction of carbon monoxide in the intake air and the volume fraction of carbon monoxide in the exhaust air from the operating data. The ratio between apparent carbon monoxide conversion rate and biomass increment was calculated to obtain the correlation quantity. Generate a sequence of correlation values ​​based on the correlation values; Based on the relative biomass bias of ethanol, a relative biomass bias sequence of ethanol was generated. The attachment state of the bubble interface shell was determined based on the ethanol relative biomass bias sequence and correlation sequence. Based on the attachment state of the bubble interface shell, the relative biomass bias sequence and correlation sequence of ethanol were screened to obtain the interface occupation class dataset. The remaining data points in the ethanol relative biomass bias sequence and the corresponding correlation data points were filtered to obtain a dataset of body phase growth. Shell the interface occupancy class data set to obtain a shelled interface marked data set; The body phase growth dataset is labeled with body phase growth characteristics to obtain a body phase growth labeled dataset. Count the shelled interface marker data set to obtain shelled interface statistics; Count the data set of body growth markers to obtain body growth statistics; Calculate the shelling bias coefficient based on the shelling interface statistics and the volume growth statistics; Based on the shelling bias coefficient, correlation analysis was performed on the correlation quantity sequence and the ethanol relative biomass bias sequence to obtain the comprehensive evaluation quantity.

4. The method for scheduling and controlling the co-production of bioproteins and ethanol using synthetic gas according to claim 1, characterized in that, Construct a set of gas injection path rearrangement actions, including: Obtain the configuration data of the controllable actuators in the fermentation reactor; Based on the set of gas injection path switching actuators in the controllable actuator configuration data, a set of gas injection path rearrangement actions is constructed.

5. The method for scheduling and controlling the co-production of bioproteins and ethanol using synthetic gas according to claim 1, characterized in that, A structural intervention action library is obtained based on the set of gas injection path rearrangement actions, including: Based on the set of stirring actuators in the controllable actuator configuration data, a set of shearing sequence actions is constructed. Based on the set of backflow path switching actuators in the controllable actuator configuration data, construct a set of backflow access location rearrangement actions; The sets of gas injection path rearrangement actions, shear sequence actions, and reflux access position rearrangement actions are merged to obtain a structural intervention action library.

6. The method for scheduling and controlling the co-production of bioproteins and ethanol using synthetic gas according to claim 1, characterized in that, The updated pheromone matrix, obtained based on comprehensive evaluation metrics and a structural intervention action database, includes: Construct a pheromone matrix based on a structural intervention action library; Based on the pheromone matrix, generate the candidate action sequence for the current iteration; Based on the calculation rules of the comprehensive evaluation quantity, calculate the sequence evaluation quantity of the candidate action sequence in the current iteration; Based on the sequence evaluation of the candidate action sequence in the current iteration, and combined with the ant colony algorithm, the pheromone matrix is ​​updated to obtain the updated pheromone matrix.

7. The method for scheduling and controlling the co-production of bioproteins and ethanol using syngas according to claim 6, characterized in that, The target action sequence is obtained based on the updated pheromone matrix, including: Based on the updated pheromone matrix, generate the candidate action sequence for the next iteration; Based on the calculation rules of the comprehensive evaluation quantity, calculate the sequence evaluation quantity of the candidate action sequence for the next iteration; Based on the sequence evaluation value of the candidate action sequences in the next iteration, the candidate action sequences in the next iteration are compared and filtered to obtain the target action sequence.

8. The method for scheduling and controlling the co-production of bioproteins and ethanol using synthetic gas according to claim 1, characterized in that, Based on the target action sequence, the fermentation process of syngas-co-produced bioproteins and ethanol is scheduled and controlled, including: Convert the target action sequence into a control command sequence; The fermentation process of syngas co-producing bioproteins and ethanol is scheduled and controlled according to the sequence of control commands.

9. A system for scheduling and controlling the synthesis of bioproteins and ethanol using a method according to any one of claims 1-9, characterized in that, The system includes: The bias calculation module calculates the relative biomass bias of ethanol based on the operating data of the syngas co-production fermentation system. The evaluation quantity generation module calculates the comprehensive evaluation quantity based on the relative biomass bias of ethanol. The action library construction module constructs a set of gas injection path rearrangement actions based on the configuration data of the controllable actuators of the fermentation reactor, and obtains a structural intervention action library based on the set of gas injection path rearrangement actions. The action sequence solving module obtains an updated pheromone matrix based on the comprehensive evaluation quantity and the structural intervention action library, and obtains the target action sequence based on the updated pheromone matrix; The scheduling and control execution module schedules and controls the fermentation process of syngas co-produced bioproteins and ethanol according to the target action sequence.

Images