Beer fermentation management method and system based on yeast metabolic activity regulation

Abstract

The embodiment of the invention discloses a beer fermentation management method and system based on yeast metabolic activity regulation, and relates to the technical field of beer fermentation. The method comprises the following steps: obtaining fermentation parameters of an initial time node and a first time node; inputting the fermentation parameters into a pre-trained fermentation prediction model for fermentation result prediction to obtain predicted fermentation result information of a second time node; acquiring expected fermentation result information preset for a specified time node, wherein the specified time node comprises a second time node; in order to reduce the error between the predicted fermentation result information and the expected fermentation result information of the corresponding time node, the fermentation parameter of the first time node is adjusted until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets a preset condition; obtaining target fermentation parameters; and regulating and controlling the yeast metabolic activity of the beer fermentation system based on the target fermentation parameters.

Description

Technical Field

[0001] The present application relates to the technical field of beer fermentation, and in particular to a beer fermentation management method and system based on the regulation of yeast metabolic activity. Background Art

[0002] The fermentation process is one of the important links in beer brewing. Since a series of complex biochemical reactions in the fermentation process are caused by the nutritional metabolism of yeast, the metabolic activity of yeast has an important influence on the production and fermentation quality of beer, as well as the physical and chemical indicators of beer and its typical flavor.

[0003] Based on this, in order to better control the beer production process, it is necessary to study a beer fermentation management method and system based on the regulation of yeast metabolic activity. Summary of the invention

[0004] To achieve the above objectives, one aspect of the embodiments of this specification provides a beer fermentation management method based on yeast metabolic activity regulation, the method comprising: Obtaining fermentation parameters of the beer fermentation system at an initial time node and a first time node, wherein the fermentation parameters are used to reflect the fermentation state of the beer fermentation system at the corresponding time node, and the fermentation parameters include dissolved oxygen concentration, pH value, temperature, pressure, sugar content, fermentation degree and yeast cell concentration; Inputting the fermentation parameters corresponding to the initial time node and the first time node into a pre-trained fermentation prediction model to predict the fermentation result, and obtaining the predicted fermentation result information of the beer fermentation system at a second time node, wherein the second time node is located after the first time node; Acquire expected fermentation result information preset for a specified time node, wherein the specified time node includes the second time node; For the purpose of reducing the error between the predicted fermentation result information and the expected fermentation result information at the corresponding time node, the fermentation parameter at the first time node is adjusted until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets a preset condition, thereby obtaining a target fermentation parameter; The yeast metabolic activity of the beer fermentation system is regulated based on the target fermentation parameter.

[0005] In some embodiments, the fermentation prediction model includes a plurality of fermentation prediction sub-models, and the plurality of fermentation prediction sub-models are used to predict a plurality of indicators in the fermentation result of the beer fermentation system based on the fermentation parameters; The step of inputting the fermentation parameters corresponding to the initial time node and the first time node into a pre-trained fermentation prediction model to predict the fermentation result, and obtaining the predicted fermentation result information of the beer fermentation system at the second time node, comprises: Obtaining a first feature vector based on the fermentation parameters of the initial time node and the first time node; Inputting the first feature vector into the fermentation prediction model, and predicting corresponding fermentation result indicators through the multiple fermentation prediction sub-models in the fermentation prediction model, to obtain multiple fermentation result indicator prediction data of the beer fermentation system at the second time node; The plurality of fermentation result indicator prediction data are integrated to obtain the predicted fermentation result information of the beer fermentation system at the second time node.

[0006] In some embodiments, the multiple fermentation prediction sub-models respectively correspond to different learning parameters, and the learning parameters are obtained by training the fermentation prediction model through sample fermentation parameters corresponding to several sample time nodes and sample fermentation result information; the training process includes: Acquire a plurality of historical fermentation data of the beer fermentation system, wherein each historical fermentation data includes sample fermentation parameters and sample fermentation result information corresponding to each sample time node from an initial time node to a final time node; For each historical fermentation data, the initial time node is combined with the sample fermentation parameter corresponding to any first sample time node, and any second sample time node to be predicted, to obtain sample input data, wherein the second sample time node is located after the first sample time node; Using the fermentation result information corresponding to the second sample time node as sample output data, wherein the sample output data includes a plurality of sample fermentation result indicators; The sample input data and the sample output data are used as a training sample, and the sample input data is input into the multiple fermentation prediction sub-models for prediction, so as to obtain a predicted fermentation result index corresponding to each fermentation prediction sub-model; Obtaining first loss data according to the error between the predicted fermentation result index corresponding to all training samples and the sample fermentation result index; The learning parameters of the multiple fermentation prediction sub-models are adjusted until the first loss data meets a preset error threshold, and the learning parameters corresponding to each fermentation prediction sub-model are fixed to obtain the trained fermentation prediction model.

[0007] In some embodiments, the fermentation parameter at the first time node is adjusted for the purpose of reducing the error between the predicted fermentation result information and the expected fermentation result information at the corresponding time node until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets a preset condition, thereby obtaining a target fermentation parameter, including: determining second loss data based on errors between each expected fermentation result indicator in the expected fermentation result information and each predicted fermentation result indicator in the predicted fermentation result information; Optimizing and calculating the fermentation parameters at the first time point according to the second loss data to obtain optimized feedback data corresponding to the second loss data; According to the optimization feedback data, obtaining preliminary optimization parameters corresponding to the fermentation parameters at the first time node; Testing whether the preliminary optimization parameters meet the preset optimization conditions, and taking the preliminary optimization parameters as target fermentation parameters when the preliminary optimization parameters meet the preset optimization conditions.

[0008] In some embodiments, the fermentation parameters at the first time node are optimized and calculated based on the second loss data to obtain optimized feedback data corresponding to the second loss data, including: calculating the gradient of the second loss data, wherein the gradient is used to reflect the contribution of each parameter of the fermentation parameters at the first time node to the second loss data.

[0009] In some embodiments, obtaining preliminary optimization parameters corresponding to the fermentation parameters at the first time node according to the optimization feedback data includes: According to the optimization feedback data and the preset parameter adjustment amplitude corresponding to each parameter, the preliminary optimization parameters corresponding to the fermentation parameters at the first time node are obtained.

[0010] In some embodiments, the testing whether the preliminary optimization parameters meet the preset optimization conditions includes: Inputting the preliminary optimization parameters into the fermentation prediction model, and using the fermentation prediction model to predict the fermentation results based on the preliminary optimization parameters, to obtain the predicted optimized fermentation result information of the beer fermentation system at the second time point; According to the error between the predicted optimized fermentation result information and the expected fermentation result information corresponding to the second time node, it is judged whether the preliminary optimization parameter meets the preset optimization condition.

[0011] In some embodiments, the method further includes: adjusting a preset parameter adjustment range corresponding to each parameter according to an error between the predicted optimized fermentation result information and the expected fermentation result information corresponding to the second time node.

[0012] In some embodiments, the fermentation parameters include intervenable fermentation parameters and non-intervenable fermentation parameters, and the adjusting of the fermentation parameters at the first time node includes: The intervenable fermentation parameters among the fermentation parameters corresponding to the first time node are adjusted until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets a preset condition.

[0013] Another aspect of the embodiments of this specification also provides a beer fermentation management system based on yeast metabolic activity regulation, the system comprising: A first acquisition module is used to acquire fermentation parameters of the beer fermentation system at an initial time node and a first time node, wherein the fermentation parameters are used to reflect the fermentation state of the beer fermentation system at the corresponding time node, and the fermentation parameters include dissolved oxygen concentration, pH value, temperature, pressure, sugar content, fermentation degree and yeast cell concentration; A fermentation result prediction module, used for inputting the fermentation parameters corresponding to the initial time node and the first time node into a pre-trained fermentation prediction model to predict the fermentation result, and obtaining the predicted fermentation result information of the beer fermentation system at a second time node, wherein the second time node is located after the first time node; A second acquisition module, used to acquire expected fermentation result information preset for a specified time node, wherein the specified time node includes the second time node; a fermentation parameter adjustment module, for adjusting the fermentation parameter at the first time node for the purpose of reducing the error between the predicted fermentation result information and the expected fermentation result information at the corresponding time node, until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets a preset condition, thereby obtaining a target fermentation parameter; The yeast metabolic activity control module is used to control the yeast metabolic activity of the beer fermentation system based on the target fermentation parameter.

[0014] The beer fermentation management method and system based on yeast metabolic activity regulation provided in the embodiments of this specification may bring about at least the following beneficial effects: (1) In the beer fermentation management method and system based on yeast metabolic activity regulation provided in some embodiments of the present specification, the fermentation result is predicted based on the fermentation parameters of the beer fermentation system at the initial time node and the first time node, and then the fermentation parameters at the first time node are adjusted according to the error between the predicted fermentation result and the expected fermentation result to obtain the target fermentation parameter. Finally, the yeast metabolic activity of the beer fermentation system is regulated according to the target fermentation parameter, so that the beer fermentation process can be accurately managed, abnormal fluctuations in the fermentation process can be effectively avoided, and the fermentation quality of the beer as well as the final beer quality and flavor can be ensured; (2) In the beer fermentation management method and system based on yeast metabolic activity regulation provided in some embodiments of this specification, by adopting a fermentation prediction model including multiple sub-models to realize the prediction of different fermentation result indicators, each sub-model can learn different learning parameters during the training process, thereby improving the accuracy and reliability of the model's prediction results for different indicators; (3) Through the beer fermentation management method and system based on yeast metabolic activity regulation provided in some embodiments of this specification, the expected fermentation results and fermentation parameters can be flexibly adjusted according to different beer types and production requirements, thereby realizing various types of fermentation control.

[0015] Additional features will be described in part in the following description. For those skilled in the art, it will become apparent by reviewing the following and accompanying drawings, or it may be understood by the production or operation of the examples. The features of this specification may be implemented and obtained by practicing or using various aspects of the methods, tools, and combinations described in the following detailed examples. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein: Figure 1 is an exemplary flow chart of a beer fermentation management method based on yeast metabolic activity regulation according to some embodiments of this specification; Figure 2 is an exemplary structural schematic diagram of a fermentation prediction model according to some embodiments of this specification; Figure 3 is a flow chart of an exemplary training method of a fermentation prediction model according to some embodiments of this specification; Figure 4 It is an exemplary module diagram of a beer fermentation management system based on the regulation of yeast metabolic activity according to some embodiments of this specification. DETAILED DESCRIPTION

[0017] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0018] It should be understood that the "system", "device", "unit" and / or "module" used in this specification is a method for distinguishing different components, elements, parts, parts or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0019] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0020] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0021] The beer fermentation management method and system based on yeast metabolic activity regulation provided in the embodiments of this specification are described in detail below in conjunction with the accompanying drawings.

[0022] Figure 1 is an exemplary flow chart of a beer fermentation management method based on yeast metabolic activity regulation according to some embodiments of this specification. In some embodiments, the beer fermentation management method based on yeast metabolic activity regulation can be executed by processing logic, which can include hardware (e.g., circuits, dedicated logic, programmable logic, microcode, etc.), software (instructions running on a processing device to perform hardware simulation), etc. or any combination thereof. In some embodiments, Figure 1One or more operations in the flowchart of the beer fermentation management method based on yeast metabolic activity regulation can be implemented by a processing device and / or a terminal device. For example, the beer fermentation management method based on yeast metabolic activity regulation can be stored in a storage device in the form of a computer program or instruction, and called and / or executed by a processing device and / or a terminal device.

[0023] Reference Figure 1 The beer fermentation management method based on yeast metabolic activity regulation provided in the embodiment of the present application may include: Step S110, obtaining fermentation parameters of the beer fermentation system at the initial time node and the first time node. In some embodiments, step S110 may be executed by the first obtaining module 210 mentioned below.

[0024] In the embodiment of the present application, the initial time node can be understood as the starting moment of the beer fermentation process, and the first time node is an arbitrary time point in the fermentation process. Since the beer fermentation process takes a certain amount of time, in order to ensure the accuracy of the subsequent prediction of the fermentation result information, there should be a certain time interval between the first time node and the initial time node. For example, the first time node can be any time node after 24 hours from the start of fermentation and before the end of fermentation.

[0025] In an embodiment of the present application, the fermentation parameters of the beer fermentation system at the initial time node and the first time node can be obtained, and then the fermentation parameters of the initial time node and the first time node are processed through subsequent steps to obtain predicted fermentation result information. In an embodiment of the present application, the fermentation parameters can be used to reflect the fermentation state of the beer fermentation system at the corresponding time node. For example, in some embodiments of the present application, the fermentation parameters may include dissolved oxygen concentration, pH value, temperature, pressure, sugar content, fermentation degree and yeast cell concentration.

[0026] It should be pointed out that during the beer fermentation process, dissolved oxygen concentration is one of the key factors affecting the metabolic activity of yeast. At the initial time node, the measurement of dissolved oxygen concentration can reflect the oxygen supply status at the initial stage of fermentation, which is crucial for the growth and metabolism of yeast. At the first time node, the change in dissolved oxygen concentration can indicate the oxygen consumption during the fermentation process, thereby providing a basis for adjusting the fermentation conditions. Monitoring of pH value is equally important because it affects the metabolic pathways of yeast and the generation of fermentation byproducts. The control of temperature and pressure is crucial to maintaining the stability and efficiency of the fermentation process, while the determination of sugar content and fermentation degree is directly related to the final flavor and quality of beer. Monitoring of yeast cell concentration helps to evaluate the growth status and fermentation activity of yeast, and provides data support for the optimization of the fermentation process. In the embodiment of the present application, by combining these fermentation parameters, the fermentation process of beer can be more accurately predicted and controlled, thereby improving the quality and production efficiency of beer.

[0027] Specifically, in the embodiment of the present application, the dissolved oxygen concentration can be monitored in real time by a dissolved oxygen sensor installed in a beer fermentation tank. The measurement of pH value can be achieved by a pH meter, which can provide continuous pH readings to facilitate timely adjustment of the pH of the fermentation liquid. The control of temperature and pressure can be achieved by the temperature and pressure control system of the fermentation tank itself, which can maintain constant conditions during the fermentation process. The determination of sugar content can be achieved by a refractometer, while the determination of fermentation degree can be completed by specific chemical analysis methods (such as alcohol determination method, reducing sugar consumption method, gas release determination method, etc.). The yeast cell concentration can be obtained by taking a certain amount (generally about 1 ml) of fermentation liquid from the beer fermentation tank and using a microscope to count or using modern biotechnology means such as flow cytometer to detect the fermentation liquid. The specific acquisition methods of the above fermentation parameters can be regarded as prior art, and will not be discussed in detail in this specification.

[0028] In the embodiment of the present application, the above fermentation parameters can be obtained once at the initial time node and the first time node. In some embodiments, the above fermentation parameters can be stored in a storage device, and the first acquisition module 210 can obtain the fermentation parameters of the beer fermentation system at the initial time node and the first time node from the storage device.

[0029] Step S120, input the fermentation parameters corresponding to the initial time node and the first time node into a pre-trained fermentation prediction model to predict the fermentation result, and obtain the predicted fermentation result information of the beer fermentation system at the second time node. In some embodiments, step S120 can be performed by the fermentation result prediction module 220 mentioned below.

[0030] In an embodiment of the present application, the fermentation parameters corresponding to the initial time node and the first time node obtained in the above step can be input into a pre-trained fermentation prediction model to predict the fermentation result, and obtain the predicted fermentation result information of the beer fermentation system at the second time node, wherein the second time node can refer to any time node after the first time node. Exemplarily, in some embodiments, the second time node can refer to the time node corresponding to the end of fermentation; in some embodiments, the second time node can be a time node between the end of fermentation and the first time node.

[0031] Figure 2 is an exemplary structural diagram of a fermentation prediction model according to some embodiments of this specification. Figure 2 In this embodiment of the present application, the fermentation prediction model may include multiple fermentation prediction sub-models (such as Figure 2 ), the multiple fermentation prediction sub-models can be used to predict the multiple indicators in the fermentation result of the beer fermentation system based on the above-mentioned fermentation parameters, so as to obtain the predicted fermentation result information corresponding to the second time node. In other words, in an embodiment of the present application, the predicted fermentation result information may include multiple predicted fermentation result indicators, wherein each predicted fermentation result indicator can be obtained by a corresponding fermentation prediction sub-model. Exemplarily, in some embodiments of the present application, the fermentation result indicators may include alcohol content, carbon dioxide content, fermentation residual sugar content, foam stability, total acid and volatile acid content, etc.

[0032] Specifically, in an embodiment of the present application, the fermentation parameters corresponding to the initial time node and the first time node can be input as input data into the fermentation prediction model, and then the input data is subjected to feature extraction by the feature extraction unit to obtain a first feature vector. In some embodiments of the present application, the feature extraction unit may include one or more neural network layers for processing and converting the input data to extract features that have an important impact on predicting the fermentation results. Exemplarily, in some embodiments, the one or more neural network layers may include convolutional layers, pooling layers, and the like. In an embodiment of the present application, the feature extraction unit is capable of learning complex patterns and correlations in the input data, thereby improving the accuracy and reliability of the prediction. In some embodiments, the feature extraction unit may also include dimensionality reduction techniques, such as principal component analysis (PCA) or autoencoders, to reduce the dimensionality of the data and remove noise.

[0033] Furthermore, the first feature vector extracted by the feature extraction unit can be input into each fermentation prediction sub-model to predict the corresponding fermentation result index, and obtain multiple fermentation result index prediction data of the beer fermentation system at the second time node. Further, the multiple fermentation result index prediction data are integrated by the fermentation result index integration unit to obtain the predicted fermentation result information of the beer fermentation system at the second time node.

[0034] In an embodiment of the present application, the multiple fermentation prediction sub-models may have the same or similar structure, for example, they may include a fully connected layer, an activation function layer, and an output layer. In an embodiment of the present application, the multiple fermentation prediction sub-models may correspond to different learning parameters, respectively, and the learning parameters may be obtained by training the fermentation prediction model through sample fermentation parameters corresponding to several sample time nodes and sample fermentation result information.

[0035] Figure 3 is a flow chart of an exemplary training method of a fermentation prediction model according to some embodiments of this specification. Figure 3 , the training process of the fermentation prediction model may include: Step S101 : acquiring a plurality of historical fermentation data of a beer fermentation system, wherein each historical fermentation data includes sample fermentation parameters and sample fermentation result information corresponding to each sample time node from an initial time node to a final time node.

[0036] In the embodiment of the present application, the data in a complete fermentation process can be regarded as a historical fermentation data. In each complete fermentation process, fermentation parameter collection and fermentation result information collection can be performed every certain period of time (for example, 4 hours). The specific implementation method of fermentation parameter collection can be referred to above and will not be repeated here.

[0037] As for the fermentation result information, in the embodiment of the present application, it may include indicators such as alcohol content, carbon dioxide content, fermentation residual sugar content, foam stability, total acid and volatile acid content. Among them, alcohol content is one of the basic characteristics of beer, which indicates the content of ethanol in the beer and directly affects the taste, flavor and consumer choice of beer. In the embodiment of the present application, the ethanol content in the fermentation liquid can be determined by density method or gas chromatography, and then expressed in the form of volume fraction (% vol) to obtain the alcohol content of the beer fermentation liquid at each time point. Carbon dioxide is one of the by-products of beer fermentation, which can give beer a unique bubble and refreshing feeling. Therefore, the carbon dioxide content can affect the taste and flavor performance of beer. In the embodiment of the present application, the carbon dioxide content of the fermentation liquid at each time point can be determined by pressure method or carbonation content meter. The fermentation residual sugar content refers to the total amount of sugar in the beer that is not completely converted by yeast after the fermentation process is completed. It has an important influence on the sweetness, taste and final flavor of the beer. In the embodiment of the present application, the determination of the fermentation residual sugar content can be carried out by chemical analysis, such as using enzyme method to determine the content of glucose and maltose, and then calculate the total residual sugar content. Foam is one of the typical characteristics of beer, and its stability affects the sensory quality of beer and the perception of consumers (good foam can enhance the visual appeal and taste of beer). In the embodiments of the present application, the foam retention time and height can be measured by shaking detection or foam stability measuring instrument to evaluate the foam stability and persistence of the beer fermentation liquid at various time points. The total acid and volatile acid content are key factors affecting the flavor balance of beer, and their determination can be achieved by acid-base titration, in which the total acid represents the sum of all acidic substances in beer, and volatile acid refers specifically to those volatile acidic substances produced during the fermentation process. In some embodiments, the volatile acid content of the beer fermentation liquid at various time points can be determined by distillation and titration.

[0038] In an embodiment of the present application, the sample fermentation parameter and sample fermentation result information can be collected for each sample time node from the initial time node to the final time node in the above manner to obtain multiple historical fermentation data.

[0039] Exemplarily, in some embodiments of the present application, each historical fermentation data can be expressed as: ; Among them, T0 represents the initial time node, T1~Ts represents the 1st~sth time nodes; Indicates the 1st to nth fermentation parameters corresponding to the initial time node, Indicates the 1st to mth fermentation result indicators corresponding to the initial time node. And so on, Indicates the 1st to nth fermentation parameters corresponding to the sth time node, Indicates the 1st to mth fermentation result indicators corresponding to the sth time node. In the embodiment of the present application, the values ​​of m and n can be the same or different.

[0040] Step S102 , for each historical fermentation data, the initial time node and the sample fermentation parameter corresponding to any first sample time node, and any second sample time node to be predicted are combined to obtain sample input data.

[0041] In the embodiment of the present application, the second sample time node should be located after the first sample time node. Sample fermentation parameters corresponding to the 5th sample time node , and the 10th sample time node (i.e., the second sample time node to be predicted) T10 are combined to obtain a sample input data. The sample input data can be expressed as .

[0042] It can be understood that by combining the initial time node in each historical fermentation data with the sample fermentation parameters corresponding to any first sample time node and any second sample time node to be predicted in a similar manner, more sample input data can be obtained, which will not be listed one by one in this specification.

[0043] Step S103: taking the fermentation result information corresponding to the second sample time node as sample output data, wherein the sample output data includes a plurality of sample fermentation result indicators.

[0044] In the embodiment of the present application, the fermentation result information corresponding to the second sample time node can be used as the sample output data corresponding to the sample input data. , you can As the corresponding sample output data, the sample output data includes There are m sample fermentation result indicators in total, and each sample fermentation result indicator can be used as a reference output of a fermentation prediction sub-model.

[0045] Step S104, taking the sample input data and the sample output data as a training sample, and inputting the sample input data into the multiple fermentation prediction sub-models for prediction, to obtain a predicted fermentation result index corresponding to each fermentation prediction sub-model.

[0046] After obtaining the sample input data and sample output data through the above steps, the matching sample input data and sample output data can be used as a training sample, and the sample input data can be input into multiple fermentation prediction sub-models in the fermentation prediction model for prediction to obtain the predicted fermentation result index corresponding to each fermentation prediction sub-model.

[0047] Step S105, obtaining first loss data according to the errors between the predicted fermentation result indicators corresponding to all training samples and the sample fermentation result indicators.

[0048] In the embodiment of the present application, the error between the predicted fermentation result index and the sample fermentation result index may refer to the difference between the corresponding values. Specifically, in the embodiment of the present application, each fermentation result index may be mapped to a corresponding value through a preset mapping method, and then the first loss data may be obtained by calculating the difference between the values.

[0049] Exemplarily, in some embodiments of the present application, the first loss data may be expressed as follows: ; Wherein, L1 represents the first loss data; w represents the total number of all training samples; m represents the number of fermentation result indicators in the fermentation result information; represents the j-th sample fermentation result indicator in the fermentation result information corresponding to the second sample time node t2 in the i-th training sample, Represents the j-th predicted fermentation result indicator in the predicted fermentation result information corresponding to the second sample time node t2 in the i-th training sample.

[0050] Step S106, adjusting the learning parameters of the multiple fermentation prediction sub-models until the first loss data meets a preset error threshold, fixing the learning parameters corresponding to each fermentation prediction sub-model, and obtaining a trained fermentation prediction model.

[0051] In an embodiment of the present application, the learning parameters of the multiple fermentation prediction sub-models can be adjusted based on the first loss data. For example, in some embodiments, the learning parameters (such as weights and biases) of the multiple fermentation prediction sub-models can be adjusted by a gradient descent algorithm. Specifically, the gradient of the first loss data with respect to the learning parameters of each fermentation prediction sub-model can be calculated, and the learning parameters can be updated according to the reverse direction of the gradient. The update rule can be expressed as: ; in, represents the updated learning parameters, represents the learning parameters before updating; Represents the learning rate, which determines the magnitude of parameter adjustment at each update; Represents the first loss data L1 about the learning parameter The gradient (i.e., partial derivative) of the above update rule is iteratively executed until the first loss data L1 is less than or equal to the preset error threshold (the error threshold can be set according to actual needs and is not specifically limited in this specification), and accurate training of each fermentation prediction sub-model can be achieved. After the training is completed, each fermentation prediction sub-model will have the ability to predict specific fermentation result indicators, thereby providing strong support for the optimization management of the beer fermentation process.

[0052] In the embodiment of the present application, when the first loss data meets the preset error threshold, the learning parameters corresponding to each fermentation prediction sub-model can be fixed to obtain a trained fermentation prediction model. More details about model training can be regarded as prior art and will not be repeated in this specification.

[0053] Step S130, obtaining the expected fermentation result information preset for the designated time node, wherein the designated time node includes the second time node. In some embodiments, step S130 can be performed by the second obtaining module 230 mentioned below.

[0054] In the embodiment of the present application, the expected fermentation result information can be set for the specified time node, and the expected fermentation result information can reflect multiple expected fermentation result indicators corresponding to each specified time node. It should be pointed out that in the embodiment of the present application, the specified time node should include the above-mentioned second time node, which is the target time node for the fermentation result prediction.

[0055] In some embodiments of the present application, the expected fermentation result information can be obtained based on historical fermentation data statistics. In some embodiments, the setting of the expected fermentation result information can be customized based on factors such as the type of beer, quality requirements, and market demand. In other words, in the embodiments of the present application, the expected fermentation result can be flexibly adjusted according to different beer types and production requirements. For example, for beer that pursues a rich taste, a higher fermentation degree and a specific flavor compound content (such as carbon dioxide content) may be set as indicators of the expected fermentation result.

[0056] Step S140, in order to reduce the error between the predicted fermentation result information and the expected fermentation result information at the corresponding time node, the fermentation parameter at the first time node is adjusted until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets the preset condition, thereby obtaining the target fermentation parameter. In some embodiments, step S140 can be performed by the fermentation parameter adjustment module 240 mentioned later.

[0057] Specifically, in the embodiment of the present application, the second loss data can be determined based on the error between each expected fermentation result index in the expected fermentation result information and each predicted fermentation result index in the predicted fermentation result information. In some embodiments, the second loss data can be expressed as follows: .

[0058] Wherein, L2 represents the second loss data; m represents the number of fermentation result indicators in the fermentation result information; represents the jth expected fermentation result indicator in the expected fermentation result information corresponding to the second time node t2, Represents the j-th predicted fermentation result indicator in the predicted fermentation result information corresponding to the second time node t2.

[0059] After obtaining the above-mentioned second loss data L2, the fermentation parameters of the first time node can be optimized and calculated according to the second loss data to obtain the optimization feedback data corresponding to the second loss data; then, the preliminary optimization parameters corresponding to the fermentation parameters of the first time node can be obtained according to the optimization feedback data; finally, it can be tested whether the preliminary optimization parameters meet the preset optimization conditions, and when the preliminary optimization parameters meet the preset optimization conditions, the preliminary optimization parameters can be used as the target fermentation parameters.

[0060] It should be noted that, in the embodiment of the present application, the above-mentioned optimization feedback data may refer to the gradient of the second loss data L2 relative to the fermentation parameter, and the gradient may be used to reflect the contribution of each parameter in the fermentation parameter at the first time node to the second loss data. Specifically, in the embodiment of the present application, the gradient of the second loss data may be calculated, and the gradient may be used as the optimization feedback data corresponding to the second loss data.

[0061] In some embodiments of the present application, the process of obtaining preliminary optimization parameters corresponding to the fermentation parameters at the first time node based on the optimization feedback data may include: obtaining preliminary optimization parameters corresponding to the fermentation parameters at the first time node based on the optimization feedback data and the preset parameter adjustment amplitude corresponding to each parameter.

[0062] Among them, the preset parameter adjustment range can be understood as the adjustment step corresponding to a single adjustment, which is used to indicate the adjustment value of each fermentation parameter in a single adjustment process. In practical applications, the preset parameter adjustment range can be set based on historical fermentation data and experience to ensure the stability and efficiency of the fermentation parameter adjustment process. In the embodiment of the present application, different fermentation parameters can correspond to the same or different preset parameter adjustment ranges.

[0063] In some embodiments, the determination process of the above preliminary optimization parameters can be expressed by the following formula: ; in, represents the adjusted fermentation parameters, represents the fermentation parameters before adjustment; Indicates fermentation parameters The corresponding preset parameter adjustment range, which determines the parameter adjustment range at each adjustment; The second loss data L2 represents the fermentation parameters The gradient (i.e. partial derivative) of .

[0064] It can be understood that in the embodiments of the present application, each parameter in the fermentation parameter can be adjusted separately by referring to the above method to obtain the preliminary optimized parameters.

[0065] Further, after obtaining the preliminary optimization parameters, the preliminary optimization parameters can be input into the trained fermentation prediction model for processing, and the fermentation result prediction is performed based on the preliminary optimization parameters by the fermentation prediction model to obtain the predicted optimization fermentation result information of the beer fermentation system at the second time node. Then, according to the error between the predicted optimization fermentation result information and the expected fermentation result information corresponding to the second time node, it is judged whether the preliminary optimization parameters meet the preset optimization conditions. For example, when the error is within an acceptable range, it can be considered that the preliminary optimization parameters determined by the above process are appropriate and can be directly applied to the actual fermentation process. On the contrary, if the error exceeds the preset range, it is necessary to further iteratively adjust the preliminary optimization parameters until the error between the predicted optimization fermentation result information corresponding to the preliminary optimization parameters and the expected fermentation result information corresponding to the second time node meets the preset optimization conditions. In the embodiment of the present application, the preset optimization conditions can be set according to actual needs, and are not specifically limited in this specification.

[0066] In some embodiments of the present application, during the fermentation parameter adjustment process, the preset parameter adjustment amplitude corresponding to each parameter can be adjusted according to the error between the predicted optimized fermentation result information and the expected fermentation result information corresponding to the second time node. For example, when the error between the predicted optimized fermentation result information and the expected fermentation result information corresponding to the second time node is large, the corresponding parameter adjustment amplitude can be increased to speed up the convergence speed; conversely, if the error is small, the parameter adjustment amplitude can be reduced to improve the adjustment accuracy of the fermentation parameter.

[0067] It should be noted that, in some embodiments of the present application, the above-mentioned fermentation parameters may include intervenable fermentation parameters and non-intervenable fermentation parameters, wherein the non-intervenable fermentation parameters may include fermentation parameters that cannot be regulated and fermentation parameters that are difficult to regulate (such as fermentation degree, pH value, sugar content). Based on this, in the embodiments of the present application, in order to facilitate the implementation of subsequent regulation operations, in the process of fermentation parameter optimization, only the intervenable fermentation parameters among the fermentation parameters corresponding to the first time node may be adjusted until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameters and the expected fermentation result information meets the preset conditions, thereby obtaining the target fermentation parameters.

[0068] It should be noted that, in the present application embodiment, although the above-mentioned non-interventional fermentation parameters are not regulated, by using them as reference factors in the fermentation process, the fermentation process can be more comprehensively monitored and analyzed. For example, although parameters such as fermentation degree, pH value, sugar content are not directly involved in regulation, their changing trends and final values ​​can provide important information for the optimization of the fermentation process. By analyzing the dynamic changes of these non-interventional fermentation parameters, early identification and early warning of potential problems in the fermentation process can be carried out, so as to take corresponding measures to ensure the stability of the fermentation process and the quality of the final product. In addition, in the present application embodiment, by using the monitoring data of these non-interventional fermentation parameters to establish a fermentation prediction model, the prediction accuracy and reliability of the fermentation prediction results can be improved to a certain extent.

[0069] Step S150, regulating the yeast metabolic activity of the beer fermentation system based on the target fermentation parameter. In some embodiments, step S150 may be performed by the yeast metabolic activity regulating module 250 mentioned below.

[0070] After determining the target fermentation parameters through the above steps, the yeast metabolic activity regulation module 250 can adjust the fermentation parameters of the beer fermentation system based on the target fermentation parameters, thereby achieving management of the beer fermentation process and optimizing the fermentation quality of the beer by regulating the yeast metabolic activity.

[0071] Specifically, in some embodiments of the present application, the oxygen supply can be adjusted through an oxygen pipeline connected to a beer fermentation tank, thereby adjusting the dissolved oxygen concentration in the fermentation tank to control the respiration of the yeast and thereby affect its metabolic activity. For pressure and temperature, the regulation can be achieved through the temperature and pressure control system of the fermentation tank. For the regulation of yeast cell concentration, the regulation of yeast cell concentration can be achieved by adjusting the supply of nutrients (including nitrogen sources (such as amino acids, urea, etc.), vitamins and minerals, etc.).

[0072] It should be noted that the yeast cell concentration is also related to factors such as dissolved oxygen concentration and temperature. In some embodiments, the control amount of the yeast cell concentration can be decomposed into fermentation parameters such as dissolved oxygen concentration and temperature, so as to achieve the regulation of the yeast cell concentration by regulating other fermentation parameters such as dissolved oxygen concentration and temperature. Regarding the implementation method of decomposing the control amount of the yeast cell concentration into fermentation parameters such as dissolved oxygen concentration and temperature, the mapping relationship can be determined through a large amount of experimental data (for example, determined by data fitting), and then the parameter decomposition can be achieved based on the mapping relationship. The mathematical method involved in this process can be regarded as a prior art and will not be described in detail in this specification.

[0073] In some embodiments, when the sugar content needs to be increased, the sugar content can also be regarded as an intervenable fermentation parameter. As for its specific regulation method, the sugar content during the fermentation process can be regulated by adding an appropriate amount of sugar substances, such as maltose, glucose, etc., to the beer fermentation tank.

[0074] It can be understood that in the embodiment of the present application, by predicting the fermentation result based on the fermentation parameters of the beer fermentation system at the initial time node and the first time node, and then adjusting the fermentation parameters of the first time node according to the error between the predicted fermentation result and the expected fermentation result, the target fermentation parameters are obtained, and finally the yeast metabolic activity of the beer fermentation system is regulated according to the target fermentation parameters, the beer fermentation process can be accurately managed, abnormal fluctuations in the fermentation process can be effectively avoided, and the fermentation quality of the beer and the final beer quality and flavor can be ensured. In addition, in the embodiment of the present application, the method can also flexibly adjust the expected fermentation results and fermentation parameters according to different beer types and production requirements to achieve the best fermentation effect.

[0075] Figure 4 Schematic diagram of a module of a beer fermentation management system based on yeast metabolic activity regulation according to some embodiments of this specification. Figure 4The beer fermentation management system 200 shown in the figure based on the regulation of yeast metabolic activity can be implemented in the form of software and / or hardware. For example, it can be configured to a processing device and / or a terminal device in the form of software and / or hardware to predict the fermentation result based on the fermentation parameters of the beer fermentation system at the initial time node and the first time node, and then adjust the fermentation parameters at the first time node based on the error between the predicted fermentation result and the expected fermentation result to obtain the target fermentation parameters, and finally regulate the yeast metabolic activity of the beer fermentation system based on the target fermentation parameters.

[0076] Reference Figure 4 In some embodiments, the beer fermentation management system 200 based on yeast metabolic activity regulation may include a first acquisition module 210, a fermentation result prediction module 220, a second acquisition module 230, a fermentation parameter adjustment module 240 and a yeast metabolic activity regulation module 250.

[0077] The first acquisition module 210 may be used to acquire fermentation parameters of the beer fermentation system at an initial time node and a first time node.

[0078] The fermentation result prediction module 220 can be used to input the fermentation parameters corresponding to the initial time node and the first time node into a pre-trained fermentation prediction model to predict the fermentation result, and obtain the predicted fermentation result information of the beer fermentation system at the second time node.

[0079] The second acquisition module 230 may be used to acquire expected fermentation result information preset for a specified time node, wherein the specified time node includes the second time node.

[0080] The fermentation parameter adjustment module 240 can be used to adjust the fermentation parameters of the first time node with the purpose of reducing the error between the predicted fermentation result information and the expected fermentation result information of the corresponding time node, until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets the preset conditions, thereby obtaining the target fermentation parameters.

[0081] The yeast metabolic activity regulation module 250 may be used to regulate the yeast metabolic activity of the beer fermentation system based on the target fermentation parameter.

[0082] For more details about the above modules, please refer to other places in this manual (for example Figure 1~Figure 3 part and its related description) will not be repeated here.

[0083] It should be understood that Figure 4The beer fermentation management system 200 and its modules based on yeast metabolic activity regulation shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art can understand that the above-mentioned method and system can be implemented using computer executable instructions and / or included in a processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of this specification can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software (for example, firmware).

[0084] It should be noted that the above description of the beer fermentation management system 200 based on yeast metabolic activity regulation is provided for illustrative purposes only and is not intended to limit the scope of this specification. It is understandable that for those skilled in the art, according to the description of this specification, without departing from this principle, the various modules can be arbitrarily combined, or a subsystem can be formed to connect with other modules. For example, Figure 4 The first acquisition module 210, the fermentation result prediction module 220, the second acquisition module 230, the fermentation parameter adjustment module 240 and the yeast metabolic activity control module 250 described in the system can be different modules in a system, or a module can realize the functions of two or more of the above modules. For another example, the beer fermentation management system 200 based on yeast metabolic activity regulation can also include a pretreatment module, which can be used to pretreat the aforementioned fermentation parameters. Such variations are all within the scope of protection of this specification. In some embodiments, the aforementioned modules can be part of a processing device and / or a terminal device.

[0085] In summary, the beneficial effects that may be brought about by the embodiments of the present specification include but are not limited to: (1) In the beer fermentation management method and system based on yeast metabolic activity regulation provided in some embodiments of the present specification, by predicting the fermentation result based on the fermentation parameters of the beer fermentation system at the initial time node and the first time node, and then adjusting the fermentation parameters at the first time node according to the error between the predicted fermentation result and the expected fermentation result, the target fermentation parameter is obtained, and finally the yeast metabolic activity of the beer fermentation system is regulated according to the target fermentation parameter, the beer fermentation process can be accurately managed, abnormal fluctuations in the fermentation process can be effectively avoided, and the fermentation quality of the beer as well as the final beer quality and flavor can be ensured; (2) In the beer fermentation management method and system based on yeast metabolic activity regulation provided in some embodiments of the present specification, by flexibly adjusting the expected fermentation results and fermentation parameters according to different beer types and production requirements, various types of fermentation control can be achieved; (3) In the beer fermentation management method and system based on yeast metabolic activity regulation provided in some embodiments of the present specification, by adopting a fermentation prediction model including multiple sub-models to achieve the prediction of different fermentation result indicators, each sub-model can learn different learning parameters during the training process, thereby improving the accuracy and reliability of the model's prediction results for different indicators.

[0086] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or may be any other possible beneficial effects.

[0087] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0088] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0089] In addition, it will be understood by those skilled in the art that various aspects of this specification may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of this specification may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of this specification may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0090] A computer storage medium may include a propagated data signal containing computer program code, for example, in baseband or as part of a carrier wave. The propagated signal may be in a variety of forms, including electromagnetic, optical, etc., or a suitable combination. A computer storage medium may be any computer-readable medium other than a computer-readable storage medium, which can be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit the program for use. The program code on the computer storage medium may be transmitted via any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0091] The computer program codes required for the operation of each part of this specification can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C language, Visual Basic, Fortran2003, Perl, COBOL2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby and Groovy, or other programming languages, etc. The program code can be run entirely on the user's computer, or run on the user's computer as an independent software package, or run partly on the user's computer and partly on a remote computer, or run entirely on a remote computer or processing device. In the latter case, the remote computer can be connected to the user's computer through any network form, such as a local area network (LAN) or a wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as software as a service (SaaS).

[0092] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing processing device or mobile device.

[0093] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0094] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining the digits. Although the numerical domains and parameters used to confirm the breadth of the range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0095] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, documents, etc., cited in this specification is hereby incorporated by reference in its entirety. Except for application history documents that are inconsistent with or conflicting with the contents of this specification, documents that limit the broadest scope of the claims of this specification (currently or later attached to this specification) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the materials attached to this specification are inconsistent or conflicting with the contents described in this specification, the descriptions, definitions, and / or use of terms in this specification shall prevail.

[0096] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A beer fermentation management method based on yeast metabolic activity regulation, characterized in that: include: Obtaining fermentation parameters of the beer fermentation system at an initial time node and a first time node, wherein the fermentation parameters are used to reflect the fermentation state of the beer fermentation system at the corresponding time node, and the fermentation parameters include dissolved oxygen concentration, pH value, temperature, pressure, sugar content, fermentation degree and yeast cell concentration; Inputting the fermentation parameters corresponding to the initial time node and the first time node into a pre-trained fermentation prediction model to predict the fermentation result, and obtaining the predicted fermentation result information of the beer fermentation system at a second time node, wherein the second time node is located after the first time node; Acquire expected fermentation result information preset for a specified time node, wherein the specified time node includes the second time node; For the purpose of reducing the error between the predicted fermentation result information and the expected fermentation result information at the corresponding time node, the fermentation parameter at the first time node is adjusted until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets a preset condition, thereby obtaining a target fermentation parameter; The yeast metabolic activity of the beer fermentation system is regulated based on the target fermentation parameter.

2. The method according to claim 1, characterized in that The fermentation prediction model includes a plurality of fermentation prediction sub-models, and the plurality of fermentation prediction sub-models are used to predict a plurality of indicators in the fermentation result of the beer fermentation system based on the fermentation parameters; The step of inputting the fermentation parameters corresponding to the initial time node and the first time node into a pre-trained fermentation prediction model to predict the fermentation result, and obtaining the predicted fermentation result information of the beer fermentation system at the second time node, comprises: Obtaining a first feature vector based on the fermentation parameters of the initial time node and the first time node; Inputting the first feature vector into the fermentation prediction model, and predicting corresponding fermentation result indicators through the multiple fermentation prediction sub-models in the fermentation prediction model, to obtain multiple fermentation result indicator prediction data of the beer fermentation system at the second time node; The plurality of fermentation result indicator prediction data are integrated to obtain the predicted fermentation result information of the beer fermentation system at the second time node.

3. The method according to claim 2, characterized in that The multiple fermentation prediction sub-models respectively correspond to different learning parameters, and the learning parameters are obtained by training the fermentation prediction model through sample fermentation parameters corresponding to several sample time nodes and sample fermentation result information; The training process includes: Acquire a plurality of historical fermentation data of the beer fermentation system, wherein each historical fermentation data includes sample fermentation parameters and sample fermentation result information corresponding to each sample time node from an initial time node to a final time node; For each historical fermentation data, the initial time node is combined with the sample fermentation parameter corresponding to any first sample time node, and any second sample time node to be predicted, to obtain sample input data, wherein the second sample time node is located after the first sample time node; Using the fermentation result information corresponding to the second sample time node as sample output data, wherein the sample output data includes a plurality of sample fermentation result indicators; The sample input data and the sample output data are used as a training sample, and the sample input data is input into the multiple fermentation prediction sub-models for prediction, so as to obtain a predicted fermentation result index corresponding to each fermentation prediction sub-model; Obtaining first loss data according to the error between the predicted fermentation result index corresponding to all training samples and the sample fermentation result index; The learning parameters of the multiple fermentation prediction sub-models are adjusted until the first loss data meets a preset error threshold, and the learning parameters corresponding to each fermentation prediction sub-model are fixed to obtain the trained fermentation prediction model.

4. The method according to claim 1, characterized in that The method of adjusting the fermentation parameter at the first time node for the purpose of reducing the error between the predicted fermentation result information and the expected fermentation result information at the corresponding time node until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets a preset condition, thereby obtaining a target fermentation parameter, comprises: determining second loss data based on errors between each expected fermentation result indicator in the expected fermentation result information and each predicted fermentation result indicator in the predicted fermentation result information; Optimizing and calculating the fermentation parameters at the first time point according to the second loss data to obtain optimized feedback data corresponding to the second loss data; According to the optimization feedback data, obtaining preliminary optimization parameters corresponding to the fermentation parameters at the first time node; Testing whether the preliminary optimization parameters meet the preset optimization conditions, and taking the preliminary optimization parameters as target fermentation parameters when the preliminary optimization parameters meet the preset optimization conditions.

5. The method according to claim 4, characterized in that The method of optimizing the fermentation parameters at the first time node based on the second loss data to obtain optimized feedback data corresponding to the second loss data includes: calculating the gradient of the second loss data, wherein the gradient is used to reflect the contribution of each parameter of the fermentation parameters at the first time node to the second loss data.

6. The method according to claim 5, characterized in that The step of obtaining preliminary optimization parameters corresponding to the fermentation parameters at the first time node according to the optimization feedback data includes: According to the optimization feedback data and the preset parameter adjustment amplitude corresponding to each parameter, the preliminary optimization parameters corresponding to the fermentation parameters at the first time node are obtained.

7. The method according to claim 6, characterized in that The testing of whether the preliminary optimization parameters meet the preset optimization conditions includes: Inputting the preliminary optimization parameters into the fermentation prediction model, and using the fermentation prediction model to predict the fermentation results based on the preliminary optimization parameters, to obtain the predicted optimized fermentation result information of the beer fermentation system at the second time point; According to the error between the predicted optimized fermentation result information and the expected fermentation result information corresponding to the second time node, it is judged whether the preliminary optimization parameter meets the preset optimization condition.

8. The method according to claim 7, characterized in that The method further includes: adjusting a preset parameter adjustment range corresponding to each parameter according to an error between the predicted optimized fermentation result information and the expected fermentation result information corresponding to the second time node.

9. The method according to claim 1, characterized in that The fermentation parameters include intervenable fermentation parameters and non-intervenable fermentation parameters, and the adjusting of the fermentation parameters at the first time node includes: The intervenable fermentation parameters among the fermentation parameters corresponding to the first time node are adjusted until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets a preset condition.

10. A beer fermentation management system based on yeast metabolic activity regulation, characterized in that: include: A first acquisition module is used to acquire fermentation parameters of the beer fermentation system at an initial time node and a first time node, wherein the fermentation parameters are used to reflect the fermentation state of the beer fermentation system at the corresponding time node, and the fermentation parameters include dissolved oxygen concentration, pH value, temperature, pressure, sugar content, fermentation degree and yeast cell concentration; A fermentation result prediction module, used for inputting the fermentation parameters corresponding to the initial time node and the first time node into a pre-trained fermentation prediction model to predict the fermentation result, and obtaining the predicted fermentation result information of the beer fermentation system at a second time node, wherein the second time node is located after the first time node; A second acquisition module, used to acquire expected fermentation result information preset for a specified time node, wherein the specified time node includes the second time node; a fermentation parameter adjustment module, for adjusting the fermentation parameter at the first time node for the purpose of reducing the error between the predicted fermentation result information and the expected fermentation result information at the corresponding time node, until the error between the predicted fermentation result information corresponding to the adjusted fermentation parameter and the expected fermentation result information meets a preset condition, thereby obtaining a target fermentation parameter; The yeast metabolic activity control module is used to control the yeast metabolic activity of the beer fermentation system based on the target fermentation parameter.

Images