Stirring control method and stirring mechanism for biological fermentation
Through sampling and prediction technology, the biofermentation stirring control parameters are optimized, and the problems of lack of mobility and low decision-making efficiency in traditional technology are solved, and efficient industrial production of the biofermentation process is achieved.
Patent Information
- Application Number
- CN202411561694.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-05
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In traditional biofermentation technology, the control parameters of the agitator are based on empirical settings, lack mobility and low decision-making efficiency, so that industrial production cannot be achieved.
By receiving the stirring control optimization request, healthy fermentation state sampling is performed, the expected threshold timing information of pH and dissolved oxygen content is obtained, the stirring control factor set is traversed, and several stirring control factor value results are generated, and the fermentation state prediction and control parameter search are carried out. Optimal, generate optimization parameters.
Adaptive optimization of stirring control parameters based on specific fermentation conditions is achieved, fermentation efficiency and product quality are improved, and the technical effect of industrial production is achieved.
Smart Images

Figure CN119061209A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of material mixing, and in particular to a stirring control method and a stirring mechanism for biological fermentation. Background Art
[0002] During the biological fermentation process, the stirring system provides power through the motor and gearbox to operate the stirring paddle and stirring shaft, thereby mixing the materials in the tank better, which is conducive to full contact between solids and nutrients, facilitating nutrient absorption and dispersion of metabolites. It can also disperse the air entering the tank evenly, increase the gas-liquid contact area in the fermentation tank, and facilitate the mixing of oxygen and fermentation liquid.
[0003] However, in traditional bio-fermentation technology, the agitator is usually simply controlled based on control parameters set based on experience. The control parameters set based on experience are not transferable and have low decision-making efficiency, making them impossible to realize industrialization. Summary of the invention
[0004] The present application provides a stirring control method and a stirring mechanism for biological fermentation, which are used to solve the technical problems in the prior art that the traditional biological fermentation technology is based on experience to set control parameters, has no portability and low decision-making efficiency, and cannot be industrialized.
[0005] The first aspect of the present application provides a stirring control method for biological fermentation, the method comprising: receiving a stirring control optimization request through a user terminal, wherein the stirring control optimization request includes a fermentation agent type, a fermentation substrate type, a fermentation agent ratio, and a fermentation substrate ratio; sampling a healthy fermentation state according to the fermentation agent type, the fermentation substrate type, the fermentation agent ratio, and the fermentation substrate ratio, and obtaining a pH expected threshold timing information and a dissolved oxygen content expected threshold timing information; obtaining a biological fermentation tank constraint condition and a stirring paddle constraint condition, wherein the biological fermentation tank constraint condition includes a fermentation tank diameter and a fermentation tank height, and the stirring paddle constraint condition includes a stirring paddle diameter and an identified stirring length; traversing a stirring control factor set, configuring a stirring control factor sub-value constraint interval set; according to the stirring control factor value constraint interval set, generate several first stirring control factor value result sets; with the fermentation agent type, the fermentation substrate type, the fermentation agent ratio, the fermentation substrate ratio, the fermentation tank diameter, the fermentation tank height, the stirring paddle diameter and the identified stirring length as constraints, perform fermentation state prediction on the several first stirring control factor value result sets to obtain several fermentation state prediction results; according to the several fermentation state prediction results, combine the pH expected threshold time series information and the dissolved oxygen content expected threshold time series information, perform optimization on the several first stirring control factor value result sets, generate stirring control optimization parameters, and send them to the user end.
[0006] The second aspect of the present application provides a stirring mechanism for biological fermentation, which is used to implement the above-mentioned stirring control method for biological fermentation.
[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The stirring control method for biological fermentation provided in the present application relates to the field of material mixing technology. By receiving a stirring control optimization request, sampling of healthy fermentation status is performed, time series information of expected pH threshold values and time series information of expected dissolved oxygen content threshold values are obtained, a set of stirring control factors is traversed, and several first stirring control factor value result sets are generated. Fermentation status prediction and control parameter optimization are performed to generate stirring control optimization parameters. The technical problem that traditional biological fermentation technology in the prior art sets control parameters based on experience, has no portability and low decision-making efficiency, and cannot be industrialized is solved. The method realizes adaptive optimization of stirring control parameters based on specific fermentation conditions and requirements, improves fermentation efficiency and product quality, and realizes the technical effect of industrialized production. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0009] Figure 1 A schematic flow chart of a stirring control method for biological fermentation provided in an embodiment of the present application; Figure 2 A schematic diagram of a process for obtaining desired threshold timing information of pH value and desired threshold timing information of dissolved oxygen content in a stirring control method for biological fermentation provided in an embodiment of the present application; Figure 3 A schematic diagram of a flow chart for obtaining several fermentation state prediction results in a stirring control method for biological fermentation provided in an embodiment of the present application. DETAILED DESCRIPTION
[0010] The present application provides a stirring control method for biological fermentation, which is used to solve the technical problem that the traditional biological fermentation technology in the prior art is based on experience to set control parameters, has no portability and low decision-making efficiency, and cannot be industrialized.
[0011] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0012] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Embodiment 1
[0013] like Figure 1 As shown, the present application provides a stirring control method for biological fermentation, the method comprising: P10: Receive a stirring control optimization request through a user terminal, wherein the stirring control optimization request includes a fermentation agent type, a fermentation substrate type, a fermentation agent ratio, and a fermentation substrate ratio.
[0014] Specifically, a stirring control optimization request is received through a user terminal, and the stirring control optimization request contains key information about the biological fermentation process, including the fermentation agent type, the fermentation substrate type, the fermentation agent ratio, and the fermentation substrate ratio. The fermentation agent type refers to the type of microorganism or enzyme used for biological fermentation. Different fermentation agent types may have different requirements for stirring control because their growth and metabolic characteristics may be different. The fermentation substrate type refers to the raw material or substrate used in the fermentation process. The type and characteristics of the substrate (such as fermentability, concentration, etc.) will also affect the demand for stirring control.
[0015] At the same time, the fermentation agent ratio refers to the relative amount of the fermentation agent in the whole fermentation system. The difference in the fermentation agent ratio may affect the fermentation speed, the type and quantity of the product, so corresponding stirring control is needed to optimize these effects. The fermentation substrate ratio refers to the relative amount of the substrate in the fermentation system. The difference in the substrate ratio will affect the consumption rate of the substrate, the generation rate of the product, etc., so it is also necessary to optimize it through stirring control. The stirring control optimization request can be used to guide the stirring operation in the biological fermentation process to ensure the efficiency, stability and safety of the fermentation process.
[0016] P20: Sampling the healthy fermentation state according to the fermentation agent type, the fermentation substrate type, the fermentation agent ratio and the fermentation substrate ratio, and obtaining the expected threshold time series information of pH value and the expected threshold time series information of dissolved oxygen content.
[0017] Further, such as Figure 2 As shown, step P20 of the embodiment of the present application also includes: P21: Receive historical fermentation logs uploaded by the user; P22: sampling the healthy fermentation state of the historical fermentation log according to the fermentation agent type, the fermentation substrate type, the fermentation agent ratio and the fermentation substrate ratio, and obtaining a pH record time series information set and a dissolved oxygen content record time series information set; P23: Fusion analysis is performed on the pH recording time series information set and the dissolved oxygen content recording time series information set to generate the pH expected threshold time series information and the dissolved oxygen content expected threshold time series information.
[0018] It should be understood that the healthy fermentation state is sampled according to the fermentation agent type, fermentation substrate type, fermentation agent ratio and fermentation substrate ratio provided by the user to obtain the expected threshold time series information of pH value and the expected threshold time series information of dissolved oxygen content. Specifically, firstly, a historical fermentation log uploaded by the user end that is similar to or the same as the current fermentation condition is received, and the historical fermentation log contains real-time records of various parameters (such as pH value, dissolved oxygen content, etc.) in the past fermentation process.
[0019] Further, the healthy fermentation state of the historical fermentation log is sampled according to the fermentation agent type, the fermentation substrate type, the fermentation agent ratio and the fermentation substrate ratio, and the data segments matching the current fermentation conditions are screened out from the historical fermentation log, and then, the records of pH and dissolved oxygen content are extracted from these data segments to form a pH record time series information set and a dissolved oxygen content record time series information set. These time series information sets reflect the change trend of pH and dissolved oxygen content over time under similar fermentation conditions.
[0020] Furthermore, the pH recording time series information set and the dissolved oxygen content recording time series information set are fused and analyzed, for example, by statistical analysis, trend prediction and other methods, to find out the expected range of changes or thresholds of pH and dissolved oxygen content under similar fermentation conditions, and generate the pH expected threshold time series information and the dissolved oxygen content expected threshold time series information as an important reference for subsequent optimization of stirring control.
[0021] Furthermore, step P23 of the embodiment of the present application also includes: P23-1: traverse the pH record time series information set to perform time domain hierarchical analysis to generate a pH reconstruction time series information set; P23-2: traverse the dissolved oxygen content record time series information set to perform time domain hierarchical analysis to generate a dissolved oxygen content reconstruction time series information set; P23-3: Performing simultaneous domain mode fitting on the pH reconstruction time series information set to generate the pH expected threshold time series information; P23-4: Performing simultaneous domain mode fitting on the dissolved oxygen content reconstructed time series information set to generate the dissolved oxygen content expected threshold time series information.
[0022] Specifically, the specific process of generating the expected threshold time series information of pH and the expected threshold time series information of dissolved oxygen content by in-depth analysis of the pH record time series information set and the dissolved oxygen content record time series information set can be to traverse the pH record time series information set and perform time domain hierarchical analysis on each data point in the pH record time series information set, including identifying periodic changes, trends, and outliers in the data. Through time domain hierarchical analysis, a smoother pH reconstruction time series information set with noise and outliers removed is reconstructed, which can more accurately reflect the change pattern of pH over time. Similarly, referring to the above method, the dissolved oxygen content record time series information set is traversed to perform time domain hierarchical analysis to generate a dissolved oxygen content reconstruction time series information set.
[0023] Furthermore, the pH reconstruction time series information set is processed using the mode fitting method in the same time domain. Mode fitting is a statistical method used to find the values with the highest frequency of occurrence in a specific time domain, and these values are regarded as the expected thresholds in the time domain. In this way, a time series information reflecting the expected range of pH change is generated, that is, the pH expected threshold time series information. Similarly, the dissolved oxygen content reconstruction time series information set is subjected to mode fitting in the same time domain to generate the dissolved oxygen content expected threshold time series information.
[0024] In summary, based on the data in the historical fermentation log, accurate time series information of the expected threshold values of pH and dissolved oxygen content is generated for the current fermentation conditions, which provides the ideal range of pH and dissolved oxygen content under similar fermentation conditions for subsequent optimization of stirring control.
[0025] Furthermore, step P23-1 of the embodiment of the present application also includes: P23-11: Configure pH deviation threshold; P23-12: Extracting a first time domain pH record value and a second time domain pH record value according to the pH record time series information set, wherein the first time domain and the second time domain are neighbors of each other; P23-13: When the pH record value deviation between the pH record value of the first time domain and the pH record value of the second time domain is less than or equal to the pH deviation threshold, merging the first time domain and the second time domain into a third time domain; P23-14: wherein the third time domain pH record value is equal to the average of the first time domain pH record value and the second time domain pH record value; P23-15: Cyclic analysis: when the deviation of the pH recording values in any adjacent time domain is greater than the pH deviation threshold, the pH reconstruction timing information set is output.
[0026] Optionally, the specific process of performing time domain hierarchical analysis on the pH record time series information set may be, first, configuring the pH deviation threshold according to the characteristics of the fermentation process and historical data, to determine when adjacent time domains can be merged into a larger time domain during the time domain hierarchical analysis process, to ensure that the merged time domain can reflect the stable change trend of the pH.
[0027] Further, the pH record values of two adjacent time domains (the first time domain and the second time domain) are sequentially extracted from the pH record time series information set to obtain the first time domain pH record value and the second time domain pH record value. When the pH record value deviation of the first time domain pH record value and the second time domain pH record value is small, that is, less than or equal to the pH deviation threshold, it is considered that the pH change in the two time domains is stable, and the first time domain and the second time domain can be merged into a larger time domain, that is, the third time domain.
[0028] Furthermore, when merging adjacent time domains, the pH record value of the new time domain (third time domain) is set to the average of the pH record value of the first time domain and the pH record value of the second time domain, thereby achieving the purpose of smoothly reflecting the change trend of pH in the merged time domain. The above steps are executed repeatedly until there are no adjacent time domains that can be merged, that is, the pH record value deviation of any adjacent time domain is greater than the pH deviation threshold. At this time, the output time series information set is the pH reconstruction time series information set, which can reflect the stable change trend of pH over time after removing noise and outliers.
[0029] Furthermore, step P23-3 of the embodiment of the present application also includes: P23-31: traverse the pH reconstruction time series information set and extract the first time domain pH reconstruction value set; P23-32: Performing mode fitting on the first time-domain pH reconstruction value set to generate a first time-domain pH expected threshold value; P23-33: Add the first time domain expected pH threshold to the expected pH threshold timing information.
[0030] It should be understood that the specific process of performing mode fitting on the pH reconstruction time series information set can be, first, traversing the entire pH reconstruction time series information set, for each time domain, extracting all pH reconstruction values in the time domain, for example, for the first time domain, forming a first time domain pH reconstruction value set, this set contains all reconstructed pH values in the time domain. Further, the first time domain pH reconstruction value set is subjected to mode fitting, that is, the mode fitting method is used for analysis to determine the most likely value of pH in the first time domain, that is, the expected threshold value of pH, and generate the first time domain pH expected threshold value.
[0031] Furthermore, the first time domain expected pH threshold is added to the expected pH threshold time series information, and so on, and the mode fitting is repeated for each time domain until all time domains are processed to generate a complete pH expected threshold time series information. This information reflects the expected value range of pH in different time domains, which can be used for subsequent stirring control optimization to provide an ideal range of pH changes under similar fermentation conditions.
[0032] P30: Obtain the biological fermentation tank constraint conditions and the stirring paddle constraint conditions, wherein the biological fermentation tank constraint conditions include the fermentation tank diameter and the fermentation tank height, and the stirring paddle constraint conditions include the stirring paddle diameter and the marked stirring length.
[0033] Specifically, based on industry experience data, biological fermentation tank constraints and stirring paddle constraints are obtained. Among them, the biological fermentation tank constraints include the fermentation tank diameter and the fermentation tank height. The fermentation tank diameter can determine the capacity and flow characteristics of the fermentation liquid in the tank. In the process of stirring control optimization, it is necessary to consider the influence of the fermentation tank diameter on the stirring effect and fluid dynamics characteristics. The fermentation tank height, together with the diameter, determines the total volume and shape of the fermentation tank, which has an important influence on the mixing and flow pattern of the fluid during the stirring process.
[0034] The stirring paddle constraint conditions include the stirring paddle diameter and the marked stirring length. The stirring paddle diameter determines the stirring capacity of the stirring paddle and the contact area with the fermentation liquid, which will affect the stirring effect and power consumption of the stirring paddle. The marked stirring length refers to the effective length of the stirring paddle or the length of the stirring shaft, which determines the stirring range of the stirring paddle in the fermentation tank and has a direct impact on the uniformity and efficiency of the stirring.
[0035] P40: Traverse the set of stirring control factors and configure the set of constraint intervals for the stirring control factors.
[0036] P50: Generate a plurality of first stirring control factor value result sets according to the stirring control factor value constraint interval set.
[0037] Optionally, a set of stirring control factors is obtained, wherein the stirring control factor set includes stirring control factors such as ventilation volume, stirring speed, and temperature, which have a direct impact on the stirring effect of the fermentation process. The stirring control factor set is traversed, and a value constraint interval is configured for each stirring control factor to generate a set of stirring control factor value constraint intervals. For example, for numerical factors such as stirring speed and stirring time, the value constraint interval is formed by the minimum value and the maximum value. For discrete factors such as rotation direction, a given list is directly used as the constraint interval.
[0038] Furthermore, according to the set of stirring control factor value constraint intervals, the stirring control factor value is randomly selected within each stirring control factor value constraint interval, and different stirring control factor value combinations are generated, thereby generating several first stirring control factor value result sets for subsequent stirring control optimization experiments or simulations.
[0039] P60: Taking the fermentation agent type, the fermentation substrate type, the fermentation agent ratio, the fermentation substrate ratio, the fermentation tank diameter, the fermentation tank height, the agitator diameter and the identified stirring length as constraints, the fermentation state is predicted for the result sets of the values of the several first stirring control factors to obtain several fermentation state prediction results.
[0040] Further, such as Figure 3 As shown, step P60 of the embodiment of the present application also includes: P61: Obtaining the kth stirring control factor value result of the plurality of first stirring control factor value result sets; P62: Taking the fermentation agent type, the fermentation substrate type, the fermentation agent ratio, the fermentation substrate ratio, the fermentation tank diameter, the fermentation tank height, the stirring paddle diameter and the marked stirring length as constraints, the fermentation state sampling is performed on the k-th stirring control factor value result to obtain the k-th pH record time series information set and the k-th dissolved oxygen content record time series information set; P63: Perform a fusion analysis on the kth pH record time series information set and the kth dissolved oxygen content record time series information set to generate a kth fermentation state prediction result, and add it to the several fermentation state prediction results.
[0041] It should be understood that, by using multiple key parameters in the fermentation process, such as the fermentation agent type, the fermentation substrate type, the fermentation agent ratio, the fermentation substrate ratio, the geometric dimensions of the fermentation tank, the diameter of the stirring paddle and the identified stirring length as constraints, the fermentation state is predicted for several first stirring control factor value result sets, so as to evaluate the impact of different stirring control factor values on the fermentation process and predict the corresponding fermentation state.
[0042] Specifically, a specific value result is selected from the plurality of first stirring control factor value result sets, and recorded as the kth stirring control factor value result. This value result contains the specific numerical values of all stirring control factors. Furthermore, with the fermentation agent type, the fermentation substrate type, the fermentation agent ratio, the fermentation substrate ratio, the fermentation tank diameter, the fermentation tank height, the stirring paddle diameter and the identified stirring length as constraints, the fermentation process is simulated or experimented in combination with the kth stirring control factor value result, and the time series information of the pH value and the dissolved oxygen content in the fermentation process is recorded, forming the kth pH record time series information set and the kth dissolved oxygen content record time series information set respectively.
[0043] Further, referring to the above method, the k-th pH record time series information set and the k-th dissolved oxygen content record time series information set are fused and analyzed, including time domain hierarchical analysis and time domain mode fitting. Time domain hierarchical analysis is a data analysis method that takes into account the time sequence and hierarchy. It can time align the k-th pH record time series information set and the k-th dissolved oxygen content record time series information set, and identify the time sequence and hierarchical relationship between them. Simultaneous domain mode fitting is a method of fitting data in the same time domain. Here, the time series information of pH and dissolved oxygen content can be fitted into a unified model to predict the fermentation state. By using statistical or machine learning methods (such as regression analysis, neural networks, etc.), the k-th pH record time series information set and the k-th dissolved oxygen content record time series information set are fitted to establish a prediction model that can predict the future fermentation state based on the current pH and dissolved oxygen content.
[0044] After completing the time domain hierarchical analysis and the time domain mode fitting, the kth fermentation state prediction result is generated according to the fitted model. This prediction result includes the predicted values of pH and dissolved oxygen content in the future, as well as the overall evaluation of the fermentation state based on these predicted values. The kth fermentation state prediction result is added to the several fermentation state prediction results, which include the fermentation state prediction results corresponding to all the first stirring control factor value results, which can provide an important basis for the subsequent stirring control optimization.
[0045] P70: Based on the several fermentation state prediction results, combined with the expected pH threshold timing information and the expected dissolved oxygen content threshold timing information, perform optimization on the several first stirring control factor value result sets, generate stirring control optimization parameters, and send them to the user end.
[0046] Furthermore, step P70 of the embodiment of the present application further includes: P71: extracting a kth fermentation state prediction result according to the plurality of fermentation state prediction results, wherein the kth fermentation state prediction result has a kth pH time series prediction information and a kth dissolved oxygen content time series prediction information; P72: Calculate the kth pH deviation time series information between the kth pH time series prediction information and the pH expected threshold time series information, and count the proportion of moments in the kth pH deviation time series information that are less than or equal to the pH tolerance deviation, and set it as the first fitness parameter; P73: Calculate the kth dissolved oxygen content deviation time series information between the kth dissolved oxygen content time series prediction information and the dissolved oxygen content expected threshold time series information, and count the proportion of moments in the kth dissolved oxygen content deviation time series information that is less than or equal to the dissolved oxygen content tolerance deviation, and set it as the second fitness parameter; P74: Set the minimum value of the first fitness parameter and the second fitness parameter as the kth fitness evaluation value, and add the kth fitness to the fitness evaluation value set; P75: According to the fitness evaluation value set, performing optimization on the plurality of first stirring control factor value result sets to generate the stirring control optimization parameters; P76: Among them, the first optimization convergence condition is: the fitness evaluation value is greater than or equal to the fitness evaluation value threshold, and the second optimization convergence condition is: the preset number of optimization iterations. Convergence occurs when at least one of the first optimization convergence condition and the second optimization convergence condition is met.
[0047] In a possible embodiment of the present application, based on several fermentation state prediction results, combined with the expected threshold time series information of pH and the expected threshold time series information of dissolved oxygen content, several first stirring control factor value result sets are optimized to generate stirring control optimization parameters.
[0048] Specifically, from the several fermentation state prediction results, the kth fermentation state prediction result is randomly extracted for analysis, and the kth fermentation state prediction result has the kth pH time series prediction information and the kth dissolved oxygen content time series prediction information. First, the deviation between the kth pH time series prediction information and the expected pH threshold time series information is calculated to form the kth pH deviation time series information. Furthermore, the proportion of moments in the kth pH deviation time series information that is less than or equal to the set pH tolerance deviation is counted, and this proportion is set as a first fitness parameter, which can reflect the performance of the stirring control factor value in maintaining pH stability.
[0049] Similarly, the deviation between the k-th dissolved oxygen content time series prediction information and the expected dissolved oxygen content threshold time series information is calculated to form the k-th dissolved oxygen content deviation time series information. And the proportion of moments in this deviation time series information that is less than or equal to the set dissolved oxygen content tolerance deviation is counted, and this proportion is set as the second fitness parameter. The second fitness parameter can reflect the performance of the stirring control factor value in maintaining the stability of the dissolved oxygen content.
[0050] Further, the smaller value of the first fitness parameter and the second fitness parameter is extracted as the fitness evaluation value of the k-th stirring control factor value result, that is, the k-th fitness evaluation value, and the k-th fitness is added to the fitness evaluation value set. Further, using the fitness evaluation value set as a guide, a plurality of first stirring control factor value result sets are optimized. Exemplarily, a plurality of optimization algorithms, such as genetic algorithm, particle swarm optimization algorithm, etc., can be used for iteration and optimization, and finally a set of stirring control optimization parameters are generated.
[0051] In addition, in order to ensure the effectiveness and efficiency of the optimization process, two convergence conditions are set. The first convergence condition for optimization is that when the fitness evaluation value is greater than or equal to the set fitness evaluation value threshold, it is considered that satisfactory optimization parameters have been found and the optimization process converges. The second convergence condition for optimization is that when the number of optimization iterations reaches a preset number, regardless of whether the first condition is met, the optimization process is considered to have been completed and the iteration is stopped. When at least one of these two conditions is met, the optimization process will converge and output the optimization parameters. Finally, the stirring control optimization parameters are sent to the user end for user reference and application, so as to improve fermentation efficiency and product quality.
[0052] In summary, the embodiments of the present application have at least the following technical effects: The present application receives a stirring control optimization request, performs healthy fermentation state sampling, obtains the expected pH threshold timing information and the expected dissolved oxygen content threshold timing information, traverses the stirring control factor set, generates several first stirring control factor value result sets, performs fermentation state prediction and control parameter optimization, and generates stirring control optimization parameters.
[0053] The technical effect of adaptively optimizing stirring control parameters based on specific fermentation conditions and requirements, improving fermentation efficiency and product quality, and realizing industrial production was achieved. Embodiment 2
[0054] Based on the same inventive concept as the stirring control method for biological fermentation in the aforementioned embodiment, the present application provides a stirring mechanism for biological fermentation. The stirring mechanism and the stirring control method embodiment in the embodiment of the present application are based on the same inventive concept.
[0055] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0056] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.
[0057] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. A stirring control method for biological fermentation, characterized in that: include: Receiving, through a user terminal, a stirring control optimization request, wherein the stirring control optimization request includes a fermentation agent type, a fermentation substrate type, a fermentation agent ratio, and a fermentation substrate ratio; Perform healthy fermentation state sampling according to the fermentation agent type, the fermentation substrate type, the fermentation agent ratio and the fermentation substrate ratio to obtain expected threshold time series information of pH value and expected threshold time series information of dissolved oxygen content; Obtaining biological fermentation tank constraint conditions and stirring paddle constraint conditions, wherein the biological fermentation tank constraint conditions include fermentation tank diameter and fermentation tank height, and the stirring paddle constraint conditions include stirring paddle diameter and marked stirring length; Traverse the stirring control factor set and configure the stirring control factor value constraint interval set; According to the set of constraint intervals for the stirring control factor values, a plurality of first stirring control factor value result sets are generated; Taking the fermentation agent type, the fermentation substrate type, the fermentation agent ratio, the fermentation substrate ratio, the fermentation tank diameter, the fermentation tank height, the stirring paddle diameter and the marked stirring length as constraints, performing fermentation state prediction on the plurality of first stirring control factor value result sets to obtain a plurality of fermentation state prediction results; According to the several fermentation state prediction results, combined with the expected pH threshold time series information and the expected dissolved oxygen content threshold time series information, the several first stirring control factor value result sets are optimized to generate stirring control optimization parameters and send them to the user end.
2. The method according to claim 1, characterized in that The healthy fermentation state sampling is performed according to the fermentation agent type, the fermentation substrate type, the fermentation agent ratio and the fermentation substrate ratio to obtain the expected threshold time series information of the pH value and the expected threshold time series information of the dissolved oxygen content, including: Receive historical fermentation logs uploaded by the user; Sampling the healthy fermentation state of the historical fermentation log according to the fermentation agent type, the fermentation substrate type, the fermentation agent ratio and the fermentation substrate ratio to obtain a pH record time series information set and a dissolved oxygen content record time series information set; The pH recording time series information set and the dissolved oxygen content recording time series information set are fused and analyzed to generate the pH expected threshold time series information and the dissolved oxygen content expected threshold time series information.
3. The method according to claim 2, characterized in that The pH recording time series information set and the dissolved oxygen content recording time series information set are fused and analyzed to generate the pH expected threshold time series information and the dissolved oxygen content expected threshold time series information, including: Traversing the pH record time series information set to perform time domain hierarchical analysis to generate a pH reconstruction time series information set; Traversing the dissolved oxygen content record time series information set to perform time domain hierarchical analysis to generate a dissolved oxygen content reconstruction time series information set; Performing simultaneous domain mode fitting on the pH reconstruction time series information set to generate the pH expected threshold time series information; The dissolved oxygen content reconstructed time series information set is subjected to a mode fitting in the same time domain to generate the dissolved oxygen content expected threshold time series information.
4. The method according to claim 3, characterized in that Traversing the pH record time series information set to perform time domain hierarchical analysis to generate a pH reconstruction time series information set, including: Configure pH deviation threshold; Extracting a first time domain pH record value and a second time domain pH record value according to the pH record time series information set, wherein the first time domain and the second time domain are neighbors of each other; When the pH record value deviation between the pH record value in the first time domain and the pH record value in the second time domain is less than or equal to the pH deviation threshold, merging the first time domain and the second time domain into a third time domain; The third time-domain pH record value is equal to the average of the first time-domain pH record value and the second time-domain pH record value; Cyclic analysis is performed, and when the deviation of the pH recording values in any adjacent time domains is greater than the pH deviation threshold, the pH reconstruction time series information set is output.
5. The method according to claim 3, characterized in that Performing simultaneous domain mode fitting on the pH reconstruction time series information set to generate the pH expected threshold time series information includes: Traversing the pH reconstruction time series information set, extracting a first time domain pH reconstruction value set; Performing mode fitting on the first time-domain pH reconstruction value set to generate a first time-domain pH expected threshold value; The first time-domain expected pH threshold is added to the expected pH threshold timing information.
6. The method according to claim 1, characterized in that Taking the fermentation agent type, the fermentation substrate type, the fermentation agent ratio, the fermentation substrate ratio, the fermentation tank diameter, the fermentation tank height, the stirring paddle diameter and the identified stirring length as constraints, a fermentation state prediction is performed on the plurality of first stirring control factor value result sets to obtain a plurality of fermentation state prediction results, including: Obtaining a kth stirring control factor value result of the plurality of first stirring control factor value result sets; Taking the fermentation agent type, the fermentation substrate type, the fermentation agent ratio, the fermentation substrate ratio, the fermentation tank diameter, the fermentation tank height, the stirring paddle diameter and the marked stirring length as constraints, the fermentation state sampling is performed on the kth stirring control factor value result to obtain the kth pH record time series information set and the kth dissolved oxygen content record time series information set; The kth pH record time series information set and the kth dissolved oxygen content record time series information set are fused and analyzed to generate a kth fermentation state prediction result, which is added to the several fermentation state prediction results.
7. The method according to claim 1, characterized in that According to the plurality of fermentation state prediction results, combined with the pH expected threshold time series information and the dissolved oxygen content expected threshold time series information, the plurality of first stirring control factor value result sets are optimized to generate stirring control optimization parameters, including: Extracting a kth fermentation state prediction result according to the plurality of fermentation state prediction results, wherein the kth fermentation state prediction result has a kth pH time series prediction information and a kth dissolved oxygen content time series prediction information; Calculate the kth pH deviation time series information between the kth pH time series prediction information and the pH expected threshold time series information, and count the proportion of moments in the kth pH deviation time series information that are less than or equal to the pH tolerance deviation, and set it as a first fitness parameter; Calculate the kth dissolved oxygen content deviation time series information between the kth dissolved oxygen content time series prediction information and the dissolved oxygen content expected threshold time series information, and count the proportion of moments in the kth dissolved oxygen content deviation time series information that is less than or equal to the dissolved oxygen content tolerance deviation, and set it as a second fitness parameter; The minimum value of the first fitness parameter and the second fitness parameter is set as the kth fitness evaluation value, and the kth fitness is added to the fitness evaluation value set; According to the fitness evaluation value set, performing optimization on the plurality of first stirring control factor value result sets to generate the stirring control optimization parameters; Among them, the first optimization convergence condition is: the fitness evaluation value is greater than or equal to the fitness evaluation value threshold, and the second optimization convergence condition is: the preset number of optimization iterations. Convergence occurs when at least one of the first optimization convergence condition and the second optimization convergence condition is met.
8. A stirring mechanism, characterized in that: Used to implement the stirring control method for biological fermentation as described in any one of claims 1 to 7.
Citation Information
Patent Citations
Intelligent Food Liquid Fermentation Parameter Control Method
AU2020100379A4
Dissolved oxygen linearity control method of bioreactor
CN101392222A
Automatic fermentation process monitoring and adjusting system
CN118109287A
Automatically controlled fermentor device for process optimization and data amplification
CN1312368A
Method and Device for Controlling the Fermentation Process
US20110039297A1
Cited By
Adjustment control method and system of intelligent mulberry leaf feed fermentation device
CN119979321A
