Fermentation process prediction method and system based on generative algorithm, and terminal

Abstract

The invention provides a fermentation process prediction method and system based on a generative algorithm and a terminal, and the method comprises the steps: firstly, comprehensively collecting various fermentation parameters in the whole life cycle of biological fermentation, and carrying out the fine preprocessing of the parameters, thereby constructing a sample data set; then, aiming at the constructed sample data set, deeply mining domain features closely related to the fermentation process, and establishing a model to construct a data set according to the domain features; on the basis, one or more generative algorithms are used, and a staged generative model is constructed according to the model construction data set. Finally, the fermentation process is accurately predicted according to the input fermentation parameters by means of the constructed staged generation model, and then a reliable prediction result is obtained. According to the method, on the basis that the effectiveness of data set division and the stability of model training are not affected, domain features are effectively fused in the modeling process, and the prediction precision and generalization ability of the model are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of biological fermentation process prediction, and in particular to a fermentation process prediction method, system and terminal based on a generative algorithm. Background Art

[0002] Traditional fermentation process modeling methods mainly use statistical methods or machine learning models to analyze and model data. However, these methods usually ignore the field characteristics that play a key role in the fermentation process, such as differences in microbial species, exquisite formulations of culture media, and unique design structures of reactors. Looking at the existing technical paths, most of them focus on modeling based on the data itself, but lack the efficient integration and application of domain expertise. This lack directly leads to the lack of prediction accuracy of the constructed model. When faced with complex and changeable actual scenarios, the generalization ability is also poor, making it difficult to adapt to diverse application needs.

[0003] With the rapid development of big data and machine learning technology, combining domain features with data features for modeling has become the key to improving the prediction accuracy and optimization effect of the fermentation process. However, how to efficiently integrate domain knowledge into the model without sacrificing the effectiveness of data set division and model training stability in this process is still a technical challenge in this field. Summary of the invention

[0004] In view of the shortcomings of the prior art mentioned above, the purpose of the present invention is to provide a fermentation process prediction method, system and terminal based on a generative algorithm, which is used to solve the technical problems that the prior art ignores the integration of domain features in the fermentation process, resulting in insufficient model prediction accuracy and generalization ability.

[0005] To achieve the above-mentioned purpose and other related purposes, the present invention provides a fermentation process prediction method based on a generative algorithm, the method comprising: performing fermentation parameter collection and preprocessing operations to construct a sample data set, which comprises: collecting fermentation parameters throughout the life cycle of biological fermentation, and then preprocessing these fermentation parameters to construct a sample data set; wherein these fermentation parameters can be subdivided into pre-fermentation parameters, fermentation process parameters and fermentation result parameters; performing feature extraction of the sample data set and model data set establishment operations, which comprises: extracting domain features related to the fermentation process from the constructed sample data set, and then establishing a model construction data set; wherein each domain feature comprises fermentation static features, fermentation dynamic features and fermentation stage features; using one or more generative algorithms to construct a staged generation model according to the model construction data set; based on the constructed staged generation model, predicting the subsequent fermentation process according to the input fermentation parameters, and obtaining corresponding prediction results.

[0006] In one embodiment of the present invention, the extracting of domain features related to the fermentation process from the constructed sample data set and then establishing a model construction data set includes: extracting fermentation static features, fermentation dynamic features and fermentation stage features respectively according to the pre-fermentation parameters, fermentation process parameters and fermentation result parameters in the sample data set; performing corresponding feature processing operations on the extracted fermentation static features, fermentation dynamic features and fermentation stage features with the help of the fermentation domain feature processing mechanism; fusing the features after the feature processing operation to obtain a model construction data set.

[0007] In one embodiment of the present invention, the fermentation field feature processing mechanism includes: for non-one-dimensional features, a specific encoding method is used to convert them into low-dimensional embedded vectors; for unquantized features, domain knowledge is used to perform quantization operations; for time series data features, special time series feature processing and encoding operations are performed.

[0008] In one embodiment of the present invention, the features after the feature processing operation are fused to obtain the model building data set, including: using feature engineering to perform feature dimensionality upgrade operations on the processed fermentation static features, fermentation dynamic features and fermentation stage features to enhance the correlation between features in various fields; wherein the feature dimensionality upgrade operation includes: for continuous variables, dimensionality upgrade by polynomial expansion and logarithmic transformation; for discrete variables, dimensionality upgrade by unique hot encoding and combined features; after completing the feature dimensionality upgrade operation, the upgraded features are fused to obtain a unified feature representation, and finally the model building data set is obtained.

[0009] In one embodiment of the present invention, the use of one or more generative algorithms to construct a phased generative model based on the model construction data set includes: dividing the model construction data set according to different stages of the fermentation process into data sets for the fermentation start-up stage, the main fermentation stage, and the fermentation end stage; wherein the data set for each stage is divided into a training set, a test set, and a validation set; based on the data sets of each stage, using one or more generative algorithms to construct a generative model for the fermentation start-up stage, the main fermentation stage, and the fermentation end stage.

[0010] In one embodiment of the present invention, the generative model of each stage is composed of a plurality of sub-stage model structures using the same or different generative algorithms connected in sequence; for the first sub-stage model structure, it is used to take the features within a time period as input, and output the prediction results for the next time period after the time period; for the non-first sub-stage model structure, it is used to take the features input by the connected previous sub-stage model structure and the prediction results output by the previous sub-stage model structure as input, and output the prediction results for the next time period after the time period predicted by the previous sub-stage model.

[0011] To achieve the above-mentioned purpose and other related purposes, the present invention provides a fermentation process prediction system based on a generative algorithm, the system comprising: a parameter collection module, used to perform fermentation parameter collection and preprocessing operations to construct a sample data set, which comprises: collecting fermentation parameters throughout the life cycle of biological fermentation, and then preprocessing these fermentation parameters to construct a sample data set; wherein these fermentation parameters can be subdivided into pre-fermentation parameters, fermentation process parameters and fermentation result parameters; a domain feature extraction module, connected to the parameter collection module, used to perform feature extraction of the sample data set and model data set establishment operations, which comprises: extracting domain features related to the fermentation process from the constructed sample data set, and then establishing a model construction data set; wherein each domain feature comprises a fermentation static feature, a fermentation dynamic feature and a fermentation stage feature; a model construction module, connected to the domain feature extraction module, used to use one or more generative algorithms to construct a staged generation model according to the model construction data set; a fermentation process prediction module, connected to the model construction module, used to predict the subsequent fermentation process based on the constructed staged generation model and the input fermentation parameters, and obtain corresponding prediction results.

[0012] To achieve the above-mentioned purpose and other related purposes, the present invention provides an electronic terminal, comprising: one or more memories and one or more processors; the one or more memories are used to store computer programs; the one or more processors are connected to the memories and are used to run the computer program to execute the fermentation process prediction method based on the generative algorithm.

[0013] As described above, the present invention is a fermentation process prediction method, system and terminal based on a generative algorithm, which has the following beneficial effects: the present invention first comprehensively collects various fermentation parameters in the entire life cycle of biological fermentation, and performs fine preprocessing on them, so as to construct a sample data set. Subsequently, for the constructed sample data set, the domain features closely related to the fermentation process are deeply excavated, and a model construction data set is established accordingly. On this basis, one or more generative algorithms are used to construct a phased generation model based on the model construction data set. Finally, with the help of the constructed phased generation model, the fermentation process is accurately predicted according to the input fermentation parameters, and then a reliable prediction result is obtained. On the basis of ensuring that the effectiveness of the data set division and the stability of the model training are not affected, the present invention effectively incorporates domain features into the modeling process, significantly improving the prediction accuracy and generalization ability of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a schematic flow chart of a fermentation process prediction method based on a generative algorithm in one embodiment of the present invention.

[0015] Figure 2 Shown is a schematic diagram of domain feature extraction and processing in one embodiment of the present invention.

[0016] Figure 3 Shown is a schematic diagram of the structure of a phased generation model in an embodiment of the present invention.

[0017] Figure 4 It is a schematic flow chart of a fermentation process prediction method based on a generative algorithm in one embodiment of the present invention.

[0018] Figure 5 It is a schematic flow chart of a fermentation process prediction method based on a generative algorithm in one embodiment of the present invention.

[0019] Figure 6 Shown is a schematic diagram of the structure of a fermentation process prediction system based on a generative algorithm in one embodiment of the present invention.

[0020] Figure 7 Shown is a schematic structural diagram of an electronic terminal in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0022] It should be noted that in the following description, reference is made to the accompanying drawings, which describe several embodiments of the present invention. It should be understood that other embodiments may also be used, and that mechanical composition, structure, electrical and operational changes may be made without departing from the spirit and scope of the present invention. The following detailed description should not be considered restrictive, and the scope of the embodiments of the present invention is limited only by the claims of the published patents. The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. Spatially related terms, such as "upper", "lower", "left", "right", "below", "below", "lower", "above", "upper", etc., may be used in the text to facilitate the description of the relationship between an element or feature shown in the figure and another element or feature.

[0023] Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case of "direct connection" but also the case of "indirect connection" by placing other elements therebetween. In addition, when a part is said to "include" a certain constituent element, unless otherwise stated, it does not exclude other constituent elements, but means that other constituent elements may be included.

[0024] The terms first, second and third mentioned herein are used to describe various parts, components, regions, layers and / or segments, but are not limited thereto. These terms are only used to distinguish a certain part, component, region, layer or segment from other parts, components, regions, layers or segments. Therefore, the first part, component, region, layer or segment described below may refer to the second part, component, region, layer or segment within the scope of the present invention.

[0025] Furthermore, as used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless there is an indication to the contrary in the context. It should be further understood that the terms "comprise", "include" indicate the presence of the described features, operations, elements, components, items, kinds, and / or groups, but do not exclude the presence, occurrence or addition of one or more other features, operations, elements, components, items, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Therefore, "A, B or C" or "A, B and / or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B and C". Exceptions to this definition will only occur when the combination of elements, functions or operations is inherently mutually exclusive in some way.

[0026] The present invention provides a fermentation process prediction method based on a generative algorithm. First, various fermentation parameters in the entire life cycle of biological fermentation are comprehensively collected, and fine preprocessing is performed on them to construct a sample data set. Subsequently, for the constructed sample data set, the domain features closely related to the fermentation process are deeply excavated, and a model construction data set is established accordingly. On this basis, one or more generative algorithms are used to construct a phased generation model based on the model construction data set. Finally, with the help of the constructed phased generation model, the fermentation process is accurately predicted according to the input fermentation parameters, and then a reliable prediction result is obtained. On the basis of ensuring that the effectiveness of data set division and the stability of model training are not affected, the present invention effectively incorporates domain features into the modeling process, significantly improving the prediction accuracy and generalization ability of the model.

[0027] The following is a detailed description of the embodiments of the present invention with reference to the accompanying drawings so that those skilled in the art can easily implement the present invention. The present invention can be embodied in many different forms and is not limited to the embodiments described herein.

[0028] like Figure 1 A schematic flow chart showing a fermentation process prediction method based on a generative algorithm in an embodiment of the present invention.

[0029] The method comprises: Step S1: performing fermentation parameter collection and preprocessing operations to construct a sample data set.

[0030] In detail, the fermentation parameter collection and preprocessing operations include: collecting fermentation parameters throughout the entire life cycle of biological fermentation, and then preprocessing these fermentation parameters to construct a sample data set; all fermentation parameters throughout the entire life cycle of biological fermentation, including pre-fermentation parameters, fermentation process parameters and fermentation result parameters, which run through the entire biological fermentation period before, during and after fermentation.

[0031] In one embodiment, the pre-fermentation parameters mainly focus on the preparation and initial conditions before fermentation. These parameters include: 1. Raw material characteristic parameters, including: Component analysis includes the content of carbohydrates (such as glucose, sucrose, starch, etc.), proteins, fats, minerals, vitamins, etc. in the raw materials. For example, before fermentation, it is necessary to determine the starch content in the malt, which will affect the sugar substances that can be produced in subsequent fermentation, and thus affect the alcohol production.

[0032] Particle size: The size of raw material particles affects the contact area between microorganisms and substrates. For example, in the production of enzyme preparations by solid-state fermentation, smaller raw material particles can provide a larger surface area, which is conducive to the attachment of microorganisms and enzymatic reactions, but too small aeration may lead to poor aeration.

[0033] Moisture content: The appropriate moisture content of raw materials is crucial for the growth and metabolism of microorganisms. Before wine fermentation, the moisture content of grapes will affect the sugar concentration and the growth environment of microorganisms. Generally, the moisture content of fresh grapes is around 70% - 80%.

[0034] 2. Microbiological parameters, including: The type of strains. Different microbial strains determine the products and pathways of fermentation. For example, brewer's yeast is used for alcohol fermentation, and lactic acid bacteria is used for lactic acid fermentation. Before fermentation, it is necessary to ensure that the selected strains are pure microorganisms that meet the production goals to prevent contamination by foreign bacteria from affecting the fermentation results.

[0035] Bacteria concentration, the initial concentration of microorganisms connected to the fermentation system has an important impact on the start-up speed and fermentation cycle of fermentation. If the initial bacteria concentration is too low, the fermentation lag period may be extended; while too high a bacteria concentration may cause the microorganisms to face too fierce competition for nutrients at the beginning of fermentation.

[0036] 3. Environmental parameters, including: Temperature: Before fermentation, the appropriate ambient temperature needs to be set, which will affect the growth of microorganisms and the fermentation reaction rate. For example, before yogurt fermentation, milk is usually preheated to around 40-45°C, which is conducive to the growth and fermentation of lactic acid bacteria.

[0037] pH value: The appropriate pH value can provide a suitable growth environment for microorganisms and affect the activity of enzymes. Before citric acid fermentation, the pH of the fermentation medium is generally adjusted to about 3-7, which is the suitable pH range for Aspergillus niger growth and citric acid production.

[0038] The fermentation process parameters are key indicators that reflect the growth and metabolism of microorganisms during the fermentation process. These parameters include: 1. Physical parameters, including: Temperature. During the fermentation process, the temperature needs to be strictly controlled. Different microorganisms have their optimal growth temperature and fermentation temperature. For example, during the beer fermentation process, the temperature of the main fermentation stage of yeast is generally controlled at 10-12℃, and the temperature of the post-fermentation stage will be reduced to about 0-4℃, which can control the fermentation speed and the formation of beer flavor substances.

[0039] Pressure: In some anaerobic fermentation or fermentation processes with gas production, pressure control is very important. For example, in the process of biogas fermentation, appropriate pressure helps to maintain the anaerobic environment of the fermentation system and promote the growth and gas production of methanogens.

[0040] Agitation speed (if applicable). For liquid fermentation, agitation allows microorganisms to fully contact nutrients and increase dissolved oxygen (if it is aerobic fermentation). For example, in the process of antibiotic fermentation, agitation is used to evenly mix the mycelium with the nutrients in the culture medium, which is also conducive to the transfer of oxygen. The agitation speed is generally set at around 100 - 300 rpm according to the size of the fermentation tank and the needs of the microorganisms.

[0041] 2. Chemical parameters, including: pH value changes. Microorganisms produce acidic or alkaline metabolites during the fermentation process, which causes pH value changes. For example, during lactic acid fermentation, lactic acid bacteria produce lactic acid, which reduces the pH value of the fermentation liquid. Buffers or acid-base regulators are needed to control the pH value within an appropriate range to ensure the normal growth and fermentation of microorganisms.

[0042] Dissolved oxygen (if it is aerobic fermentation). For aerobic microbial fermentation, the concentration of dissolved oxygen directly affects the respiration and metabolic rate of microorganisms. For example, in the process of amino acid fermentation, the appropriate dissolved oxygen level is maintained by controlling the ventilation volume and stirring speed. Generally, the dissolved oxygen concentration should be kept at around 20% - 30% saturation.

[0043] Substrate concentration changes. As fermentation progresses, substrates (such as sugars, proteins, etc.) are used by microorganisms and their concentrations gradually decrease. During alcohol fermentation, monitoring changes in glucose concentration can help you understand the progress of fermentation. When the glucose concentration drops to a certain level, fermentation may gradually stop or enter the post-fermentation stage.

[0044] Changes in metabolite concentration: The concentration of metabolites (such as alcohol, organic acids, antibiotics, etc.) produced during the fermentation process will gradually increase. For example, during the penicillin fermentation process, it is necessary to regularly test the penicillin content and adjust the fermentation conditions according to the concentration changes to optimize the fermentation process.

[0045] 3. Microbiological parameters The growth state of microorganisms can be determined by observing the changes in the shape, size and quantity of microorganisms under a microscope. For example, during yeast fermentation, observe whether the yeast cells have budding and reproduction, as well as the activity state of the cells.

[0046] Microbial metabolic activity is measured by testing indicators such as the respiration intensity and enzyme activity of microorganisms. For example, during the fermentation process, the activity of enzymes such as amylase and protease is tested. These enzymes are key enzymes for microorganisms to decompose substrates. Changes in enzyme activity reflect the level of microbial metabolic activity.

[0047] Fermentation result parameters mainly focus on the quality and yield of the product after fermentation, as well as the growth and metabolism of microorganisms during the fermentation process. These parameters include: 1. Product quality parameters, including: Purity is an important indicator for fermentation products (such as enzymes, organic acids, antibiotics, etc.). For example, in the production of industrial enzyme preparations, high-purity enzymes can improve their application effects and reduce the adverse effects of impurities. Through various separation and purification technologies, the purity of enzymes can reach more than 90%.

[0048] Activity: For fermentation products with biological activity (such as enzymes, bioactive peptides, etc.), the level of activity determines their quality. For example, after protease fermentation, the activity of the protease needs to be tested, which is usually measured by the amount of product produced by decomposing the substrate per minute. The higher the activity, the better the product quality.

[0049] Physical and chemical properties, including the product's color, smell, solubility, density, etc. For example, after the citric acid fermentation is completed, the citric acid product should be a colorless or white crystalline powder, without odor, and with good water solubility.

[0050] 2. Production parameters, including: Product yield refers to the ratio of the actual amount of target product obtained to the theoretical amount of product calculated based on substrate consumption. For example, in alcohol fermentation, the theoretical alcohol yield is calculated based on the amount of glucose input. The ratio of the actual amount of alcohol obtained to the theoretical yield is the alcohol yield, which reflects the efficiency of the fermentation process.

[0051] Volumetric productivity, for liquid fermentation, is the amount of product produced per unit volume of fermentation liquid per unit time. For example, in the production of biodiesel in a fermenter, the volumetric productivity is measured by calculating the amount of biodiesel produced per liter of fermentation liquid per day.

[0052] 3. Microbiological parameters, including: The number of surviving microorganisms: After fermentation, it is necessary to understand the survival of microorganisms. For example, in the production process of active dry yeast, the number of surviving yeasts must be tested after fermentation. Usually, the number of yeast cells per gram of product is required to reach a certain standard to ensure the fermentation activity of the product in subsequent use.

[0053] Changes in the physiological state of microorganisms: observe the physiological state of microorganisms after the fermentation, such as whether they enter a dormant state, whether there is cell damage, etc. For example, after the high-temperature fermentation, some microorganisms may experience physiological changes such as cell membrane damage due to high temperature stress, which will affect the preservation and subsequent use of the microorganisms.

[0054] In one embodiment, the method of preprocessing the fermentation parameters includes: performing data cleaning and normalization processing on all fermentation parameters respectively to obtain a sample data set that meets the data processing requirements.

[0055] Specifically, in the data cleaning stage, the processing of missing values ​​and outliers is covered. Missing values ​​refer to the situation where some observations in the data set are missing or not recorded, which may have a negative impact on the accuracy and reliability of data analysis in fermentation parameter data. To this end, we give priority to the method of one-time linear fitting filling. This method performs linear fitting based on the data before and after the missing values ​​to make a reasonable estimate of the missing values. Outliers are values ​​that are significantly different from most data points, and the reasons for their occurrence may include measurement errors, data entry errors, etc. These outliers will interfere with the results of data analysis. Therefore, we give priority to using the quartile method to record outliers so that they can be replaced in subsequent preprocessing.

[0056] After data cleaning, the normalization stage begins. The purpose of normalization is to scale the data range to between 0 and 1, which not only helps speed up the convergence of the model, but also improves the overall performance of the model. First, each feature of the data set is normalized, and then a random float that conforms to the Gaussian distribution (normal distribution) is generated and added to the mean of each feature to simulate the random fluctuations of the data. Subsequently, the new data with the added Gaussian float is normalized again to ensure that the data range remains between 0 and 1. Finally, through the shearing operation, all normalized values ​​are forced to be limited to the range of 0 to 1 to ensure data consistency and obtain a sample data set that meets the requirements.

[0057] Step S2: Execute feature extraction of the sample data set and establishment of the model data set.

[0058] In detail, the feature extraction of the sample data set and the model data set establishment operation include: extracting domain features related to the fermentation process from the constructed sample data set, and then establishing a model construction data set; these fermentation parameters can be subdivided into fermentation static features, fermentation dynamic features and fermentation stage features. Specifically, the fermentation static features mainly originate from the preparatory stage before fermentation. These static features provide the tone for the fermentation process and have a profound impact on the growth environment and subsequent metabolic activities of microorganisms. By carefully analyzing these features, we can more comprehensively understand the initial state of the fermentation system and provide a solid foundation for model construction. The fermentation dynamic features focus on the real-time changes in the fermentation process. By continuously monitoring and recording these dynamic parameters, we can capture the evolution trend of the fermentation system over time and reveal the inherent laws of microbial metabolic activities. These dynamic features are crucial for accurately predicting the fermentation process, optimizing fermentation conditions and improving product purity. The fermentation stage characteristics often show obvious stage characteristics, and the metabolic activities, product generation rates and resource utilization efficiency of microorganisms in each stage are different. By identifying and distinguishing these stage characteristics, we can more accurately grasp the key nodes of the fermentation process and provide a strong basis for formulating targeted regulation strategies.

[0059] In one embodiment, if Figure 2 , step S2 comprises: According to the pre-fermentation parameters, fermentation process parameters and fermentation result parameters in the sample data set, the fermentation static characteristics, fermentation dynamic characteristics and fermentation stage characteristics are extracted respectively; specifically, the fermentation static characteristics are extracted according to the pre-fermentation parameters in the sample data set. These pre-fermentation parameters often contain relatively fixed data such as the composition of the original fermentation materials, the input amount, the initial environmental conditions (temperature, humidity and other basic setting values), etc., from which the static characteristics that can reflect the characteristics of the initial state of fermentation are extracted. The dynamic characteristics that reflect the changes in the fermentation process are extracted according to the fermentation process parameters of the sample data set. The fermentation process parameters are data that change in real time during the actual fermentation process, such as the temperature change curve during the fermentation process, the fluctuation of pH value, the growth rate of microorganisms and other dynamic information. The fermentation result parameters involve various indicators of the final fermentation product (such as yield, quality indicators, component proportion, etc.) and the state parameters at the end of the entire fermentation. By analyzing the fermentation result parameters, the stage characteristics that can characterize the characteristics of different stages of fermentation are summarized.

[0060] After extracting various features, based on the feature processing mechanism in the fermentation field, corresponding feature processing operations are performed on each fermentation static feature, dynamic feature, and stage feature to optimize the feature expression or improve the feature quality. This step aims to optimize the feature expression or improve the feature quality, which can lay a solid foundation for subsequent feature fusion and model construction.

[0061] The processed fermentation static features, fermentation dynamic features, and fermentation stage features are fused to generate a model building data set that can provide effective support for subsequent model building. This data set integrates multiple aspects of the fermentation process and can fully reflect the complexity and dynamics of the fermentation system. By using this data set for model training and verification, we can build a more accurate and reliable fermentation process prediction and optimization model, providing strong support for the intelligent transformation of the biomanufacturing industry.

[0062] In one embodiment, the fermentation static characteristics, fermentation dynamic characteristics and fermentation stage characteristics can be extracted in different ways.

[0063] Specifically, given that the parameters of the fermentation process show the characteristics of time series changes, a layer of LSTM (Long Short-Term Memory Network) can be used to extract time series features for these dynamically changing parameters. LSTM can effectively capture the long-term dependencies in the time series. For these data that evolve over time during the fermentation process, it can well learn the patterns and laws of their dynamic changes, thereby extracting valuable fermentation dynamic features.

[0064] When extracting fermentation stage features, a layer of Transformer module can be used to extract fermentation stage features from each fermentation result parameter within each stage. Transformer has a powerful self-attention mechanism, which can globally model the data within the stage and capture the relationship between data elements, thereby effectively extracting stage features that can reflect the unique properties of each fermentation stage.

[0065] As for the static characteristics of fermentation, since the pre-fermentation parameters are determined at the initial stage of fermentation and are relatively stable during the subsequent fermentation process, these pre-fermentation parameters can be directly used as the static characteristics of fermentation.

[0066] In one embodiment, in the fermentation field, in order to more effectively utilize data for model building and analysis, it is necessary to perform targeted processing on various features. The processing mechanism mainly involves performing one or more of the following operations on each feature: Processing of non-one-dimensional features: In the field of fermentation, there are many complex non-one-dimensional features, such as microbial metabolic pathways, which involve the mutual transformation of multiple substances and a series of biochemical reactions, and are a multi-dimensional complex system; there is also reactor design, including the shape, size, internal structure and other multi-dimensional parameter descriptions of the reactor. For these non-one-dimensional features, a specific encoding method is used to convert them into low-dimensional embedding vectors, which can be processed using specific encoding methods such as embedding layers and word vectors. Because non-one-dimensional features may have higher dimensions, they will increase the complexity of calculations and the burden of the model. By converting them into low-dimensional embedding vectors, the dimension can be reduced while retaining the important information of the features, which is convenient for subsequent processing and calculation. In the field of fermentation, we can regard the various elements in the microbial metabolic pathway or reactor design as "words" and convert these non-one-dimensional features into low-dimensional embedding vectors through appropriate training. Doing so can not only reduce the dimension of the data and reduce the amount of calculation, but also better extract the key information in the features, which is convenient for subsequent data analysis and model construction.

[0067] Non-quantitative feature processing: In the fermentation field, some features may only be qualitative descriptions and lack quantitative numerical representation. These features can be converted into quantifiable forms through domain knowledge so as to better participate in subsequent calculations and model training.

[0068] For example, we can use metabolic flux analysis (MFA) or growth models to quantitatively describe it. Metabolic flux analysis can reveal the intensity and direction of metabolic activity of microorganisms under different conditions by quantitatively analyzing the material flow in the process of microbial metabolism, thereby providing a quantitative basis for the dynamic characteristics of microbial growth. The growth model can quantitatively simulate the growth rate, biomass accumulation, etc. of microorganisms based on the growth laws of microorganisms and environmental factors (such as nutrient concentration, temperature, etc.). Through these methods based on domain knowledge, the originally non-quantified features are converted into specific values, making the data more standardized and easier to process, which helps to improve the accuracy of data analysis and the reliability of the model.

[0069] Time series data feature processing: In the fermentation process, these time series data features contain rich information about the fermentation process, and these features will undergo special time series feature processing and encoding operations. Since time series data has a temporal order and inherent correlation, it is necessary to use specific processing and encoding methods to extract and utilize the time information contained therein so that the model can better understand and learn the features of these data. First, in terms of feature processing, we may perform data smoothing to remove noise interference in the data and make the trend of the data more obvious; we may also calculate some statistical features of the data, such as mean, variance, slope, etc., to reflect the changes in the data over a period of time. In terms of encoding, recurrent neural networks (RNNs) and their variants, such as long short-term memory networks (LSTMs) and gated recurrent units (GRUs), can be used. These models can well capture the time series information and long-term dependencies in time series data, and extract key features by encoding the time series data into the model according to the time step, providing strong support for the analysis and prediction of the fermentation process.

[0070] In one embodiment, the step of fusing the processed fermentation static features, fermentation dynamic features, and fermentation stage features to obtain a model building data set includes: First, feature engineering is used to perform feature dimensionality upgrade on the processed three types of domain features. The purpose of feature dimensionality upgrade is to enhance the correlation between features and to dig out more valuable information hidden in the data, so that the subsequent model can better capture the various internal laws and relationships in the fermentation process.

[0071] The specific dimension-upgrading operations include: For continuous variables, polynomial expansion and logarithmic transformation are used to achieve feature dimensionality increase. For example, in the fermentation process, continuous variables such as the temperature in the fermentation tank and the substrate concentration can be expanded by polynomials to generate new variables of higher powers such as quadratic terms and cubic terms based on the original variables. The advantage of this is that it can capture the nonlinear relationship that may exist between variables, because in the actual fermentation process, the influence of many factors on the fermentation results is not a simple linear relationship. Logarithmic transformation is equally important. For example, for some concentration data, after logarithmic transformation, its numerical range can be reasonably adjusted to make its distribution more in line with the assumptions of the model, and it also helps to alleviate the heteroscedasticity problem that may exist in the data. Polynomial expansion can generate higher-order features through polynomial combination of original features and explore nonlinear relationships between variables; while logarithmic transformation can stretch or compress data, change its distribution, and make the data more in line with the assumptions of the model.

[0072] For discrete variables, one-hot encoding and combined features are used to increase the dimension. One-hot encoding is a method of converting categorical variables into numerical forms that can be understood by machine learning models. The basic idea is to create a new binary column (i.e., feature) for each category. If the sample belongs to this category, the value of the corresponding column is 1, otherwise it is 0. Combined features are to mine the correlation between different discrete variables. For example, considering the two discrete variables of microbial species and fermentation stage, by combining them, new features such as "a certain microorganism in a specific fermentation stage" can be generated, thereby more comprehensively reflecting the interaction between different discrete factors in the fermentation process. One-hot encoding converts each category of the discrete variable into a binary vector, avoiding the deviation caused by the size relationship between categories; combined features can mine the combination relationship between different discrete variables and increase the expressive power of the features.

[0073] After completing the dimension upgrade of the features of various variables, the upgraded fermentation static features, fermentation dynamic features, and fermentation stage features are integrated. The specific method is to add all the features in the dimension to obtain a unified feature representation, which constitutes the model construction data set and provides comprehensive data support for subsequent model construction. Such a data set integrates all-round information of the fermentation process from the initial state, dynamic changes to the characteristics of each stage, providing a rich data foundation for the training and analysis of subsequent models.

[0074] In order to scientifically and effectively evaluate the performance and generalization ability of the model, the processed data set needs to be divided into training set, validation set and test set based on the data volume. For example, the training set: validation set: test set can be divided in the ratio of 8:1:1.

[0075] Step S3: Using one or more generative algorithms, construct a phased generative model based on the model construction data set.

[0076] In one embodiment, based on one or more generative algorithms, constructing a phased generative model according to the model construction dataset includes: First, the constructed model building data set is divided according to the different stages of the fermentation process, specifically into the start-up stage, the main fermentation stage, and the end stage data set. The reason for this division is that the fermentation process has its own unique characteristics, change rules, and influencing factors at different stages. By processing the data in stages, it is possible to more accurately capture the internal mechanism of each stage, and then build a model that fits the actual fermentation situation.

[0077] The data set for each stage includes: training set, test set and validation set; the purpose of this is to be able to conduct targeted model training, testing and validation for the fermentation process at different stages to ensure the performance and effect of the model at different stages.

[0078] Subsequently, based on one or more generative algorithms, generative models of the startup phase, main fermentation phase, and end phase are constructed for their respective data sets. The main function of these generative models is to predict the characteristics of each fermentation phase, thereby achieving a refined prediction of the entire fermentation process. This phased model construction method can better capture the unique characteristics and changing laws of the fermentation process at each stage, thereby improving the accuracy and reliability of the prediction.

[0079] It should be noted that the types of generative algorithms used include naive Bayes model and its derivative models, Gaussian mixture model and its derivative models, hidden Markov model and its derivative models, generative adversarial network and its derivative models, variational autoencoder and its derivative models, deep Boltzmann machine and its derivative models, Transformer model and its derivative models, Mamba model and its derivative models, LSTM model and its derivative models, NBEATS model and its derivative models, etc.

[0080] In one embodiment, if Figure 3 The generative model of each stage is not a single structure, but is composed of multiple sub-stage model structures that use the same or different generative algorithms connected in sequence. This hierarchical and chain-like structural design aims to make full use of the advantages of different algorithms and the information transfer between sub-stage models, so as to more accurately predict the fermentation process of multiple consecutive time periods in the corresponding stage. In addition, the number of stage model structures of the generative model of each stage can be the same or different.

[0081] The first sub-stage model structure is Figure 3The sub-stage 1 model structure takes the features within a certain time period as the features of the initial time as input, and then outputs the prediction results of the next time period after the time period based on its internal algorithm and parameter settings. For example, in the start-up stage of fermentation, if the first sub-stage model takes the various fermentation features (such as temperature, pH value, microbial concentration, etc.) collected within the first hour as input, it can output the prediction values ​​of the relevant fermentation features in the next hour after this hour, which provides a basis for further prediction of the subsequent sub-stage model structure. During training, the features within a certain time period in the data set are taken as the features of the initial time as input, and the prediction results of the next time period after the initial time period are output. At the same time, the actual data corresponding to the predicted time period is extracted from the data set. By carefully comparing the prediction results of the model structure with the actual data, the parameters of the model structure are adjusted, and the first sub-stage model structure is constructed; this method can realize the gradual prediction of the fermentation process in the time series and provide initial prediction information for the subsequent fermentation process.

[0082] The non-first sub-stage model structure, that is Figure 3 The input of the sub-stage 2 model structure to the sub-stage N model structure not only includes the input features of the connected previous sub-stage model structure, but also covers the prediction results of the previous sub-stage model structure output. By integrating this information, it can output the prediction results of the next time period after the time period predicted by the previous sub-stage model structure. In other words, the subsequent sub-stage models will continuously and progressively predict the subsequent time periods based on the prediction results of the previous sub-stage and the original input features. This design utilizes the information of the previous sub-stages to form a coherent time series prediction chain, making the prediction of the entire fermentation process more coherent and accurate, and better reflecting the dynamic changes and mutual influences between the various stages in the fermentation process.

[0083] When training a non-first sub-stage model structure, the input features and output prediction results of the previous sub-stage model structure are used as input, and the prediction results of the next time period after this time period are output. At the same time, the actual data corresponding to the prediction time period is extracted from the data set. By carefully comparing the prediction results of the model structure with the actual data, the parameters of the model structure are adjusted, and the sub-stage model structure is constructed.

[0084] Step S4: Based on the constructed phased generation model, the subsequent fermentation process is predicted according to the input fermentation parameters, and the corresponding prediction results are obtained.

[0085] In one embodiment, step S4 includes: The input fermentation parameters to be predicted are preprocessed. This preprocessing process is similar to the preprocessing of the model building data set, which may include data cleaning (such as processing missing values, outliers, etc.) and normalization, etc., to ensure that the input data quality is reliable and the format is standardized, meeting the requirements of subsequent analysis and model operation.

[0086] After the preprocessing is completed, the domain feature extraction related to the fermentation process is carried out. The fermentation static features, fermentation dynamic features and fermentation stage features are extracted from the input fermentation parameters; after the domain features are extracted, these features are processed accordingly according to the fermentation domain feature processing mechanism.

[0087] The processed domain features are input into the constructed phased generation model to predict the fermentation process and the corresponding prediction results are output; It should be noted here that the prediction ranges that can be made vary depending on the type of fermentation parameter to be predicted. When the pre-fermentation parameters are input, since they reflect the initial state of fermentation, based on the staged generation model, not only can the characteristic parameters of the subsequent time period during the startup phase be predicted, such as the change of parameters such as temperature and microbial count in the fermentation tank in the next few hours during the startup phase, but also the characteristic parameters of the main fermentation phase and the end phase can be predicted, and the parameters of the entire fermentation process can be judged from the beginning of fermentation.

[0088] If the input is the fermentation process parameters, because these parameters have already reflected some dynamic changes in the fermentation process, the model can be used to predict the characteristic parameters of subsequent time periods in the main fermentation stage. For example, in the main fermentation stage, based on the parameters such as the substrate consumption rate in the current period, it is possible to predict at what rate the substrate will be consumed and how the product generation will change in the next period of time. At the same time, it is also possible to predict the characteristic parameters of the end stage and know in advance the various situations that may occur in the end stage so that corresponding regulatory measures can be taken in time.

[0089] When the input is the fermentation result parameters, it is mainly based on the existing fermentation result information to predict the characteristic parameters of the subsequent time period within the end stage. For example, it is judged that after the fermentation is completed, the product quality may change in the subsequent storage, transportation and other links, or it is predicted whether some further post-processing measures are needed to ensure the stability of the product.

[0090] Through such a complete and detailed process, based on different types of fermentation parameter inputs, the characteristic parameters of different stages and time periods of the fermentation process can be predicted with the help of a phased generation model.

[0091] In one embodiment, if Figure 4The method further includes: step S5 executed after step S4: performing fermentation parameter sensitivity analysis on the prediction results output by the constructed phased generation model, and optimizing the corresponding characteristic parameters in the prediction results based on the fermentation parameter sensitivity analysis results to obtain a prediction optimization result.

[0092] Specifically, the prediction results output by the staged generation model are used to carry out fermentation parameter sensitivity analysis, and the sensitivity of each characteristic parameter is ranked, so as to accurately identify the key parameters that have the greatest impact on the fermentation path. This process is one of the important innovations of the present invention, which provides a key basis for the subsequent optimization of the prediction results. Based on the above-mentioned fermentation parameter sensitivity ranking results, the corresponding characteristic parameters in the prediction results are targeted for optimization. For example, for those parameters that are identified as critical and have a significant impact on the fermentation results, the prediction results are made more in line with the actual fermentation needs by adjusting their value ranges, optimizing their control strategies, etc. After this optimization process, the prediction optimization results are finally obtained, which significantly improves the accuracy and practicality of the fermentation process prediction, and provides strong support for the improvement of the fermentation process and the improvement of production efficiency.

[0093] This is another key invention point of the present invention.

[0094] In order to better describe the fermentation process prediction method based on the generative algorithm, it is now explained in conjunction with the following specific embodiments.

[0095] Example 1: A fermentation process prediction method based on a generative algorithm. Figure 5 Schematic diagram of the fermentation process prediction method based on the generative algorithm in this embodiment.

[0096] The method comprises: Step 1: Collect all parameters in the entire life cycle of biological fermentation, including but not limited to raw material parameters, process parameters and result parameters.

[0097] Among them, the raw material parameters represent the various index parameters of the raw materials used in biological fermentation. The raw material parameters used in this example include more than 1,000 items such as carbon content, moisture content, wheat content, pH buffer, etc.

[0098] Process parameters represent various index parameters in the biological fermentation process. The process parameters used in this example include temperature, pH value, dissolved oxygen concentration, stirring rate, ventilation, fermentation time, reactor type and volume, waste gas emission, pressure, etc., more than 100 items. Process parameters also include control parameters, which represent index parameters that can be considered as controllable in the biological fermentation process. The control parameters used in this example include pH value control, temperature control, dissolved oxygen control, ventilation control, substrate dosage control, pressure control, etc., more than 20 items.

[0099] The result parameters represent the quality, yield and purity of the final product in the biofermentation process, including but not limited to product concentration, product yield, product yield, metabolites, product purity, cell concentration, byproduct inhibition, pH change, gas production, etc.

[0100] Step 2: Perform data cleaning and normalization on all parameters.

[0101] First, all parameters are cleaned. The data cleaning methods include but are not limited to fitting and filling missing values, and correcting or deleting outliers and putting them into the sample data set. The missing value fitting and filling operation adopts a linear fitting filling method; outliers are recorded by quartiles to facilitate their replacement in subsequent preprocessing.

[0102] Next, the cleaned data is normalized. Then, a random float that conforms to the Gaussian distribution is generated and the Gaussian float is added to the mean. Then the new fill value is normalized to the range of 0 to 1, ensuring that the filled value does not exceed 0 and 1, and a shear operation is used to obtain a sample data set.

[0103] Step 3: Perform feature extraction based on all processed parameters.

[0104] Based on all the processed parameters, the fermentation static characteristics, fermentation dynamic characteristics, fermentation stage characteristics and curve characteristics are extracted respectively; For dynamic features, a layer of LSTM is used to extract temporal features.

[0105] For stage-by-stage features, only one layer of Transformer module is used within each stage to extract stage-by-stage features.

[0106] Draw curve graphs for all features and extract curve features. The curve features used in this example are: frequency domain, tangent slope, normal slope, tangent vector, normal vector, simple moving average, weighted moving average, exponential moving average, single exponential smoothing, double exponential smoothing, triple exponential smoothing, volatility, volatility, absolute rate of change, relative rate of change, autocorrelation, sliding correlation, skewness, kurtosis, hurst index, coefficient of variation, cumulative distribution function. The extracted curve features are also filled with missing values, outlier processing, and normalization.

[0107] Step 4: Perform feature dimensionality upgrade on all extracted features, and fuse the upgraded features to obtain a unified feature representation and obtain the model construction data set.

[0108] Use feature engineering to increase the dimension of all features, including: (1) Perform polynomial expansion and logarithmic transformation on continuous variables; (2) Perform one-hot encoding and combined features on discrete variables.

[0109] All the features after dimensionality increase are fused, and all the features are added in the dimension to obtain a unified feature representation to form a data set.

[0110] The processed data set is divided into training set, validation set, and test set, and the division ratio corresponds to 8:1:1 for training set: validation set: test set.

[0111] Step 5: Design a phased generative model and use one or more generative algorithms to predict and generate fermentation parameters.

[0112] Use the data set prepared in step 4 to design a phased generative model, and use one or more generative algorithms, including but not limited to the Transformer model and its derivative models, the Mamba model and its derivative models, the LSTM model and its derivative models, the NBEATS model and its derivative models, etc. to train, verify, and test the model.

[0113] This embodiment uses the Mamba sequence model as a generative algorithm, and divides the training data set, the verification data set, and the test data set into three stages: the startup stage, the main fermentation stage, and the end stage. The data of each stage is input into the Mamba sequence model of each stage, and the model is trained, verified, and tested to obtain three generative models.

[0114] Each generated model also includes: sub-stage 1 model, sub-stage 2 model and sub-stage 3 model. The training process of these three models includes: The input of the sub-stage 1 model is the data X for the time period 0-x 1 , output the prediction result Y for the time period x-2x 1 '; Where x is in hours; The input of the sub-stage 2 model is the input X of the stage 1 model 1 Add the prediction result Y of the stage 1 model 1 , X 2 = [X 1 ,Y 1 '], the total input length is 2x. Output data Y for the 2x-4x time period 2 '; The input of the sub-stage 3 model is the input X of the stage 2 model 2 Add the predictions Y from the stage 2 model 2 , X 3 = [X 2 ,Y 2 '], the total input length is 2x. Output data Y for the 4x-6x time period 3'; Specifically, the loss function formula is as follows: ; (1) Where: N represents the total number of sub-stages, L total Represents the total loss, Represents the loss function of the i-th stage, measuring the true value With the predicted value Here we use cosine similarity to calculate the error between . The final model accuracy reaches 99.6%.

[0115] It should be noted that the NBEATS model can also be used as a generative algorithm in this example. The training data set, validation data set, and test data set are first divided into three stages: the startup stage, the main fermentation stage, and the end stage. The data of each stage is input into the NBEATS model of each stage, and the model is trained, validated, and tested to obtain three generative models. Specifically, the model training method is consistent with the Mamba sequence model training method, and the loss function is changed from cosine similarity to MSE. The model accuracy finally reached 95.7%-97.5%.

[0116] Step 6: Based on the constructed phased generation model, the fermentation process is predicted according to the input fermentation parameters to obtain the corresponding prediction results.

[0117] Step 7: Use the prediction results of the model output generated in step 6 to perform parameter sensitivity analysis to identify the key parameters that have the greatest impact on the fermentation path. Based on the results of the parameter sensitivity analysis, optimize the corresponding parameters in the prediction results to obtain the prediction optimization results.

[0118] Similar to the principle of the above embodiment, the present invention provides a fermentation process prediction system based on a generative algorithm.

[0119] The following provides specific embodiments in conjunction with the accompanying drawings: like Figure 6 A schematic diagram of the structure of a fermentation process prediction system based on a generative algorithm in an embodiment of the present invention is shown.

[0120] The system comprises: The parameter collection module 1 is used to perform fermentation parameter collection and preprocessing operations to construct a sample data set, which includes: collecting fermentation parameters throughout the life cycle of biological fermentation, and then preprocessing these fermentation parameters to construct a sample data set; wherein these fermentation parameters can be subdivided into pre-fermentation parameters, fermentation process parameters, and fermentation result parameters; The domain feature extraction module 2 is connected to the parameter collection module 1 and is used to perform feature extraction of the sample data set and model data set establishment operations, which includes: extracting domain features related to the fermentation process from the constructed sample data set, and then establishing a model construction data set; wherein each domain feature includes a fermentation static feature, a fermentation dynamic feature, and a fermentation stage feature; A model building module 3, connected to the domain feature extraction module 2, for building a phased generation model based on the model building data set using one or more generative algorithms; The fermentation process prediction module 4 is connected to the model building module 3 and is used to predict the subsequent fermentation process based on the constructed phased generation model and the input fermentation parameters, and obtain corresponding prediction results.

[0121] Since the implementation principle of the fermentation process prediction method based on the generative algorithm has been described in the above embodiments, it will not be repeated here.

[0122] The fermentation process prediction method based on the generative algorithm provided in the embodiment of the present invention can be implemented on the terminal side or the server side. As for the hardware structure of the electronic terminal, please refer to Figure 7 , is an optional hardware structure diagram of an electronic terminal 1000 provided in an embodiment of the present invention. The terminal 1000 may be a mobile phone, a computer device, a tablet device, etc. The terminal 1000 includes: at least one processor 1001, a memory 1002, at least one network interface 10010 and a user interface 1009. The various components in the device are coupled together through a bus system 1005. It can be understood that the bus system 1005 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 1005 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, Figure 7 In the specification, various buses are labeled as bus systems.

[0123] The user interface 1009 may include a display, a keyboard, a mouse, keys, buttons, a touch panel or a touch screen.

[0124] It is understood that the memory 1002 can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), which is used as an external cache. By way of exemplary but not limiting explanation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM). The memory described in the embodiments of the present invention is intended to include but is not limited to these and any other suitable categories of memory.

[0125] The memory 1002 in the embodiment of the present invention is used to store various categories of data to support the operation of the terminal 1000. Examples of these data include: any executable program for operating on the terminal 1000, such as an operating system 10021 and an application 10022; the operating system 10021 includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks. The application 10022 may include various applications, such as a media player (MediaPlayer), a browser (Browser), etc., for implementing various application services. The fermentation process prediction method based on a generative algorithm provided in the embodiment of the present invention may be included in the application 10022.

[0126] The method disclosed in the above embodiment of the present invention can be applied to the processor 1001, or implemented by the processor 1001. The processor 1001 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 1001 or the instruction in the form of software. The above processor 1001 may be a general processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor 1001 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiment of the present invention. The general processor 1001 can be a microprocessor or any conventional processor, etc. In combination with the steps of the accessory optimization method provided in the embodiment of the present invention, it can be directly embodied as a hardware decoding processor to execute, or it can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium, which is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0127] In an exemplary embodiment, the terminal 1000 may be implemented by one or more application specific integrated circuits (ASIC), DSP, programmable logic device (PLD), complex programmable logic device (CPLD) to execute the aforementioned method.

[0128] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.

[0129] In the embodiments provided in the present application, the computer readable and writable storage medium may include a read-only memory, a random access memory, an EEPROM, a CD-ROM or other optical disk storage device, a disk storage device or other magnetic storage device, a flash memory, a USB flash drive, a mobile hard disk, or any other medium that can be used to store the desired program code in the form of an instruction or data structure and can be accessed by a computer. In addition, any connection can be appropriately referred to as a computer-readable medium. For example, if the instruction is sent from a website, a server or other remote source using a coaxial cable, an optical fiber cable, a twisted pair, a digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwaves, the coaxial cable, optical fiber cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwaves are included in the definition of the medium. However, it should be understood that computer readable and writable storage media and data storage media do not include connections, carriers, signals, or other temporary media, but are intended to be non-temporary, tangible storage media. Disk and disc, as used in this application, includes compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers.

[0130] In summary, the fermentation process prediction method, system and terminal based on the generative algorithm of the present invention first comprehensively collect various fermentation parameters in the entire life cycle of biological fermentation, and perform fine preprocessing on them, so as to construct a sample data set. Subsequently, for the constructed sample data set, the field features closely related to the fermentation process are deeply excavated, and a model construction data set is established accordingly. On this basis, one or more generative algorithms are used to construct a staged generation model based on the model construction data set. Finally, with the help of the constructed staged generation model, the fermentation process is accurately predicted according to the input fermentation parameters, and then a reliable prediction result is obtained. On the basis of ensuring that the effectiveness of the data set division and the stability of the model training are not affected, the present invention effectively incorporates the field features in the modeling process, and significantly improves the prediction accuracy and generalization ability of the model. Therefore, the present invention effectively overcomes the various shortcomings in the prior art and has a high industrial utilization value.

[0131] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A fermentation process prediction method based on a generative algorithm, characterized in that: The method comprises: Performing fermentation parameter collection and preprocessing operations to construct a sample data set, which includes: collecting fermentation parameters throughout the entire life cycle of biological fermentation, and then preprocessing the fermentation parameters to construct a sample data set; wherein the fermentation parameters are subdivided into pre-fermentation parameters, fermentation process parameters, and fermentation result parameters; Performing feature extraction of the sample data set and model data set establishment operations, which includes: extracting domain features related to the fermentation process from the constructed sample data set, and then establishing a model construction data set; wherein each domain feature includes a fermentation static feature, a fermentation dynamic feature, and a fermentation stage feature; Using one or more generative algorithms, construct a staged generative model based on the model construction data set; Based on the constructed phased generation model, the subsequent fermentation process is predicted according to the input fermentation parameters, and the corresponding prediction results are obtained.

2. The fermentation process prediction method based on a generative algorithm according to claim 1, characterized in that: The extracting of domain features related to the fermentation process from the constructed sample data set and then establishing a model construction data set includes: According to the pre-fermentation parameters, fermentation process parameters and fermentation result parameters in the sample data set, the fermentation static characteristics, fermentation dynamic characteristics and fermentation stage characteristics are extracted respectively; With the help of the feature processing mechanism in the fermentation field, corresponding feature processing operations are performed on the extracted fermentation static features, fermentation dynamic features and fermentation stage features; The features after feature processing operations are fused to obtain the model building data set.

3. The fermentation process prediction method based on the generative algorithm according to claim 2, characterized in that: The fermentation field feature processing mechanism includes: For non-one-dimensional features, a specific encoding method is used to convert them into low-dimensional embedding vectors; For unquantified features, use domain knowledge to perform quantification operations; For time series data features, special time series feature processing and encoding operations will be performed.

4. The fermentation process prediction method based on a generative algorithm according to claim 2, characterized in that: The fusing of the features after the feature processing operation to obtain the model building data set includes: The feature engineering is used to perform feature dimension upgrading operation on the processed fermentation static features, fermentation dynamic features and fermentation stage features to enhance the correlation between the features in various fields; wherein the feature dimension upgrading operation includes: for continuous variables, the dimension upgrading is performed by polynomial expansion and logarithmic transformation; for discrete variables, the dimension upgrading is performed by unique hot encoding and combined features; After completing the feature dimensionality upgrade operation, the upgraded features are fused to obtain a unified feature representation, and finally the model construction data set is obtained.

5. The fermentation process prediction method based on the generative algorithm according to claim 4, characterized in that: The step of using one or more generative algorithms to construct a staged generative model based on the model building data set includes: The model construction data set is divided into data sets of a fermentation start-up stage, a main fermentation stage, and a fermentation end stage according to different stages of the fermentation process; wherein the data set of each stage is divided into a training set, a test set, and a validation set; According to the data sets of each stage, one or more generative algorithms are used to construct generative models of the fermentation startup stage, the main fermentation stage and the fermentation end stage.

6. The fermentation process prediction method based on generative algorithm according to claim 5, characterized in that: The generative model of each stage is composed of multiple sub-stage model structures that use the same or different generative algorithms connected in sequence; The first sub-stage model structure is used to take the features within a time period as input and output the prediction results for the next time period after the time period; For non-first sub-stage model structures, it is used to take the input features of the connected previous sub-stage model structure and the prediction results output by the previous sub-stage model structure as input, and output the prediction results for the next time period after the time period predicted by the previous sub-stage model.

7. A fermentation process prediction system based on a generative algorithm, characterized in that: The system comprises: The parameter collection module is used to perform fermentation parameter collection and preprocessing operations to construct a sample data set, which includes: collecting fermentation parameters throughout the life cycle of biological fermentation, and then preprocessing these fermentation parameters to construct a sample data set; wherein these fermentation parameters are subdivided into pre-fermentation parameters, fermentation process parameters, and fermentation result parameters; A domain feature extraction module is connected to the parameter collection module and is used to perform feature extraction of the sample data set and model data set establishment operations, which includes: extracting domain features related to the fermentation process from the constructed sample data set, and then establishing a model construction data set; wherein each domain feature includes a fermentation static feature, a fermentation dynamic feature, and a fermentation stage feature; A model building module, connected to the domain feature extraction module, for building a phased generation model based on the model building data set using one or more generative algorithms; The fermentation process prediction module is connected to the model building module and is used to predict the subsequent fermentation process based on the constructed phased generation model according to the input fermentation parameters and obtain corresponding prediction results.

8. An electronic terminal, characterized in that: include: one or more memories and one or more processors; The one or more memories are used to store computer programs; The one or more processors, connected to the memory, are configured to run the computer program to perform the method according to any one of claims 1 to 6.

Images