Informatization management system and method for brewing process of Luzhou-flavor liquor

Abstract

The invention relates to the field of informatization management, and particularly discloses an informatization management system and method for the brewing process of Luzhou-flavor Baijiu, and the method comprises the steps: firstly obtaining Baijiu brewing raw material inventory data and Baijiu brewing production plans collected by a database, and then carrying out the feature extraction and correlation analysis of the Baijiu brewing raw material inventory data and Baijiu brewing production plans through a deep learning technology; and finally, an inventory optimization suggestion is generated through a generator, so that an enterprise is helped to purchase goods or adjust a production plan at a proper time, the situation of inventory overstock or raw material shortage is avoided, and the production efficiency and the quality control level are improved.

Description

Technical Field

[0001] The present application relates to the field of information management, and more specifically, to an information management system and method for the brewing process of Luzhou-flavor liquor. Background Art

[0002] Luzhou-aroma liquor is a unique traditional liquor made by high-temperature fermentation and distillation techniques. It is characterized by a full and thick body, rich and lasting aroma, a mellow and soft taste, a unique aroma and rich grain flavor. It is a major category of liquor.

[0003] However, compared with the highly automated and information-based production process of liquid liquors such as beer and wine, most liquor production companies are still in the initial stage of automation with single-unit equipment and assisted manual operation.

[0004] Therefore, an information management system and method for the brewing process of Luzhou-flavor liquor are desired. Summary of the invention

[0005] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides an information management system and method for the brewing process of Luzhou-flavor liquor, which first obtains the liquor brewing raw material inventory data and liquor brewing production plan collected by the database, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally generates inventory optimization suggestions through a generator, thereby helping the enterprise to purchase or adjust the production plan at the appropriate time, thereby avoiding inventory backlogs or raw material shortages, so as to improve production efficiency and quality control level.

[0006] According to one aspect of the present application, an information management system for a Luzhou-flavor liquor brewing process is provided, which includes:

[0007] The Luzhou-flavor liquor brewing inventory data acquisition module is used to obtain the liquor brewing raw material inventory data and liquor brewing production plan collected from the database;

[0008] A Luzhou-flavor liquor brewing inventory data extraction module is used to extract a liquor brewing raw material inventory semantic association feature vector and a liquor brewing production plan word granularity feature vector from the liquor brewing raw material inventory data collected from the database and the liquor brewing production plan;

[0009] The inventory optimization suggestion generation module is used to generate inventory optimization suggestions based on the semantic association feature vector of the liquor brewing raw material inventory and the liquor brewing production planning word granularity feature vector.

[0010] According to another aspect of the present application, there is provided an information management method for a Luzhou-flavor liquor brewing process, comprising:

[0011] Obtaining liquor brewing raw material inventory data and liquor brewing production planning collected from the database;

[0012] Extracting a liquor brewing raw material inventory semantic association feature vector and a liquor brewing production plan word granularity feature vector from the liquor brewing raw material inventory data collected from the database and the liquor brewing production plan;

[0013] Based on the semantic association feature vector of the liquor brewing raw material inventory and the liquor brewing production planning word granularity feature vector, an inventory optimization suggestion is generated.

[0014] Compared with the prior art, the present application provides an information management system and method for the brewing process of strong-flavor liquor, which first obtains the liquor brewing raw material inventory data and the liquor brewing production plan collected by the database, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally generates inventory optimization suggestions through a generator, thereby helping the company to purchase or adjust the production plan at the appropriate time, thereby avoiding inventory backlogs or raw material shortages, so as to improve production efficiency and quality control level. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 It is a block diagram of the information management system of the Luzhou-flavor liquor brewing process according to an embodiment of the present application.

[0017] Figure 2 It is a block diagram of a Luzhou-flavor liquor brewing inventory data extraction module in the information management system of the Luzhou-flavor liquor brewing process according to an embodiment of the present application.

[0018] Figure 3 It is a block diagram of the liquor brewing raw material inventory feature extraction unit in the information management system of the Luzhou-flavor liquor brewing process according to the embodiment of the present application.

[0019] Figure 4 The present invention is a block diagram of an inventory optimization suggestion generation module in an information management system of a Luzhou-flavor liquor brewing process according to an embodiment of the present application.

[0020] Figure 5 It is a flow chart of the information management method of the Luzhou-flavor liquor brewing process according to the embodiment of the present application.

[0021] Figure 6is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0023] Figure 1 Schematic diagram of the information management system for the brewing process of the Luzhou-flavor liquor of the present application embodiment. Figure 1 As shown, according to the information management system 100 of the strong-flavor liquor brewing process of the embodiment of the present application, it includes: a strong-flavor liquor brewing inventory data acquisition module 110, which is used to acquire the liquor brewing raw material inventory data and the liquor brewing production plan collected by the database; a strong-flavor liquor brewing inventory data extraction module 120, which is used to extract the liquor brewing raw material inventory semantic association feature vector and the liquor brewing production plan word granularity feature vector from the liquor brewing raw material inventory data collected by the database and the liquor brewing production plan; an inventory optimization suggestion generation module 130, which is used to generate inventory optimization suggestions based on the liquor brewing raw material inventory semantic association feature vector and the liquor brewing production plan word granularity feature vector.

[0024] In the information management system 100 of the above-mentioned Luzhou-flavor liquor brewing process, the Luzhou-flavor liquor brewing inventory data acquisition module 110 is used to obtain the inventory data of liquor brewing raw materials and liquor brewing production planning collected by the database. It should be understood that Luzhou-flavor liquor is a unique traditional wine, made by high-temperature fermentation and exquisite distillation technology. It has a full and thick body, a strong and lasting aroma, a mellow and soft taste, and exudes a unique wine aroma and rich grain flavor. It is a type of special wine in liquor. Compared with the highly automated and information-based brewing process of liquid liquors such as beer and wine, most liquor production companies are still in the initial stage of automation assisted by manual operation of single equipment. Therefore, in the technical solution of the present application, by obtaining the inventory data of liquor brewing raw materials and liquor brewing production planning collected by the database, and combining deep learning technology, generating inventory optimization suggestions helps enterprises to purchase or adjust production plans at the right time to prevent the problem of excess inventory or lack of raw materials, thereby improving production efficiency and improving quality control level.

[0025] Specifically, through the inventory data recorded and collected in the database, we can accurately grasp the quantity and type of brewing raw materials currently available, so as to effectively plan production scheduling and resource allocation. At the same time, obtaining production planning information helps to understand the production needs and plans for a period of time in the future, so as to adjust inventory strategies and production plans in time to cope with market changes and demand fluctuations. The acquisition of these data not only simplifies the production management process, but also improves production efficiency and product quality through analysis and optimization, ensuring that enterprises can maintain competitiveness and sustainable development in the fiercely competitive market.

[0026] In the information management system 100 of the brewing process of the above-mentioned Luzhou-flavor liquor, the Luzhou-flavor liquor brewing inventory data extraction module 120 is used to extract the semantic association feature vector of the liquor brewing raw material inventory and the word granularity feature vector of the liquor brewing production plan from the liquor brewing raw material inventory data collected by the database and the liquor brewing production plan. It should be understood that the purpose of extracting the semantic association feature vector and the word granularity feature vector from the liquor brewing raw material inventory data and the liquor brewing production plan collected from the database is to support more accurate and intelligent management decisions through in-depth analysis and understanding of the semantics and relevance behind the data. The extraction of the semantic association feature vector of the liquor brewing raw material inventory can capture the correlation between different raw materials and the dynamic changes in the supply chain by semantically encoding and extracting features from the inventory data, thereby optimizing the inventory management strategy, reducing the phenomenon of excess and out-of-stock, and improving the efficiency of resource utilization. At the same time, the extraction of the word granularity feature vector of the liquor brewing production plan helps to identify and analyze the key words and operation details in the production plan, thereby optimizing the production scheduling, ensuring on-time delivery and maximizing production efficiency. These feature vectors not only provide the system with a data-driven decision-making basis, but also lay the foundation for further data mining and predictive analysis, thereby driving the overall production process towards intelligence and efficiency.

[0027] Figure 2 FIG. 1 is a block diagram of a Luzhou-flavor liquor brewing inventory data extraction module in an information management system for a Luzhou-flavor liquor brewing process according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the strong-flavor liquor brewing inventory data extraction module 120 includes: a liquor brewing raw material inventory feature extraction unit 121, used to perform feature extraction on the liquor brewing raw material inventory data collected from the database to obtain the liquor brewing raw material inventory semantic association feature vector; a liquor brewing production plan feature extraction unit 122, used to perform feature extraction on the liquor brewing production plan to obtain the liquor brewing production plan word granularity feature vector.

[0028] It should be understood that the feature extraction process of the liquor brewing raw material inventory data collected by the database includes the use of advanced data analysis technologies, such as machine learning and deep learning models, to convert the raw material inventory data into numerical features or semantic features that can reflect the correlation and change trends between different raw materials. Through this process, the system can more accurately understand and analyze inventory data, such as identifying which raw materials have common usage patterns or supply chain connections, and which raw materials may have correlation changes or demand fluctuations in a specific period. This deep feature extraction not only helps to reduce blindness and excessive inventory in inventory management, but also optimizes production scheduling to ensure timely supply and maximize production efficiency. By extracting features from the liquor brewing raw material inventory data, the system can achieve more intelligent and data-driven management decisions, providing enterprises with reliable support to cope with market changes and production challenges.

[0029] Furthermore, the purpose of extracting features from the liquor brewing production plan to obtain word-granular feature vectors is to extract key information and operational features from the production plan to support more effective production management and resource optimization. The feature extraction process involves converting the production plan text into numerical or semantic features, which can accurately reflect the key words, operation steps and time nodes in the production process, helping the system to understand and analyze the complexity and variability of the production plan. Through this process, the system can identify and quantify the dependencies and priorities between different production tasks, such as determining which production links are critical paths and which operations may have a significant impact on overall production efficiency and product quality. This deep feature extraction not only helps to optimize production scheduling and resource allocation, but also identifies potential production risks and bottlenecks in advance, so as to adjust production strategies and plans in time to ensure on-time delivery and maximize production efficiency. By extracting features from the liquor brewing production plan, the system can achieve more intelligent and refined production management, providing enterprises with reliable decision support to cope with changes in market demand and production challenges.

[0030] Figure 3 FIG. 1 is a block diagram of a liquor brewing raw material inventory feature extraction unit in an information management system for a Luzhou-flavor liquor brewing process according to an embodiment of the present application. Figure 3 As shown, in a specific embodiment of the present application, the liquor brewing raw material inventory feature extraction unit 121 includes: a raw material inventory data semantic encoding subunit 1211, used to semantically encode the liquor brewing raw material inventory data collected by the database to obtain multiple liquor brewing raw material inventory semantic association feature vectors; a liquor brewing raw material inventory feature encoding subunit 1212, used to feature encode the multiple liquor brewing raw material inventory semantic association feature vectors to obtain the liquor brewing raw material inventory semantic association feature vector.

[0031] It should be understood that semantic coding of liquor brewing raw material inventory data collected by the database can provide support for refined inventory management and intelligent production decisions by deeply analyzing and understanding the semantic relationships in the data. The process of semantic coding includes using natural language processing technology and machine learning models to convert raw material inventory data into a set of feature vectors with semantic information. These feature vectors can capture the correlation, co-occurrence and change patterns between different raw materials, helping the system to accurately identify and predict demand fluctuations and resource shortages that may occur in the supply chain. Through this deep semantic coding, the system can optimize inventory management strategies, such as avoiding excessive inventory and out-of-stock phenomena through real-time monitoring and prediction, and improving raw material utilization efficiency and supply chain response speed. In addition, the extraction of semantically associated feature vectors can also support enterprises in making data-driven decisions in a complex market environment, thereby improving overall production efficiency and competitiveness. In summary, by semantically encoding liquor brewing raw material inventory data to obtain multiple semantically associated feature vectors, the system can achieve more accurate, intelligent and sustainable inventory management and production planning, bringing long-term strategic advantages and economic benefits to enterprises.

[0032] Furthermore, the purpose of feature encoding for multiple semantic association feature vectors of liquor brewing raw material inventory is to further process and handle these vectors to enhance the capabilities of data analysis and decision support systems. The feature encoding process involves converting semantic association feature vectors into numerical features or higher-level abstract representations, so that the system can understand and utilize these data more effectively. By applying specific encoding techniques, such as vectorization, embedding representation, or feature transformation based on machine learning models, semantic association feature vectors can be mapped to a higher-dimensional space, thereby enhancing their expressive power and analytical depth. This feature encoding not only helps to extract and strengthen the potential connections and patterns between different raw materials, but also optimizes the operability and applicability of data, such as playing an important role in predicting demand, optimizing supply chain strategies, and conducting intelligent inventory management. By encoding the feature vectors of semantic association feature vectors of liquor brewing raw material inventory, the system can achieve more accurate, reliable, and real-time data analysis, provide intelligent decision support for enterprises, and effectively respond to market changes and production challenges, and enhance the competitiveness and sustainable development capabilities of enterprises. In summary, feature encoding is a key step in converting semantic association feature vectors into higher-level and more useful representations, providing a solid foundation and innovative impetus for data-driven decision-making and operations.

[0033] In a specific embodiment of the present application, the raw material inventory data semantic encoding subunit 1211 includes: passing the liquor brewing raw material inventory data collected from the database through a liquor brewing raw material inventory text understanding model to obtain a liquor brewing raw material inventory word vector sequence; passing the liquor brewing raw material inventory word vector sequence through a liquor brewing raw material inventory long short-term memory network based on a deep learning model to obtain the multiple liquor brewing raw material inventory semantic association feature vectors.

[0034] It should be understood that the text understanding model of liquor brewing raw material inventory uses natural language processing techniques, such as word embedding, word frequency statistics, topic modeling and other methods to map the text information in the raw material inventory data to a high-dimensional vector space. These word vector sequences can capture key information such as raw material name, supplier information, quantity, storage conditions, etc., while retaining semantic and contextual information, which helps the system to accurately understand and interpret the content and change trends of inventory data. Through this text understanding model, the system can achieve efficient processing and analysis of large-scale inventory data, such as identifying and quantifying the similarities and differences between different raw materials, and predicting possible supply chain problems or demand fluctuations. This deep text understanding and word vector serialization not only helps to optimize inventory management and resource allocation, but also improves the production efficiency and market responsiveness of enterprises. In summary, generating word vector sequences through the text understanding model of liquor brewing raw material inventory can provide enterprises with smarter, more accurate and more real-time inventory data analysis and decision support, helping them maintain their leading position and sustainable development in a highly competitive market environment. Specifically, word segmentation processing is performed on the liquor brewing raw material inventory data collected from the database to obtain a raw material inventory word sequence; and the embedding layer of the liquor brewing raw material inventory text understanding model is used to map each raw material inventory word sequence into a raw material inventory word embedding vector to obtain a liquor brewing raw material inventory word vector sequence.

[0035] Furthermore, the word vector sequence of liquor brewing raw material inventory is processed by the long short-term memory network (LSTM) based on the deep learning model, aiming to use its powerful sequence modeling ability to capture the time series relationship and semantic features in the inventory data, and then generate multiple semantic association feature vectors of liquor brewing raw material inventory. As a deep learning model suitable for sequence data, LSTM can effectively process the time series information in the inventory data while retaining the semantic context in the word vector sequence. By training the LSTM model, the system can learn and extract the association features between different raw materials, such as the trend of inventory changes, seasonal changes, and potential fluctuations in the supply chain. These feature vectors can more accurately reflect the complex relationships and influencing factors between raw materials, helping enterprises to achieve more intelligent and predictable inventory management strategies. In addition, the processing based on the deep learning model can also automatically discover and analyze patterns and anomalies in inventory data, such as supply shortages or surpluses, so as to adjust production plans and supply chain strategies in a timely manner to cope with changes in market demand and maximize resource utilization efficiency. In summary, by using the LSTM network based on the deep learning model to process the word vector sequence of liquor brewing raw material inventory, the data expression ability and analysis depth can be improved, providing enterprises with more accurate, real-time and intelligent inventory management and decision-making support, thereby enhancing their competitiveness and sustainable development capabilities.

[0036] In a specific embodiment of the present application, the liquor brewing raw material inventory feature encoding subunit 1212 includes: arranging the multiple liquor brewing raw material inventory semantic association feature vectors to obtain a liquor brewing raw material inventory semantic association input matrix; passing the liquor brewing raw material inventory semantic association input matrix through a liquor brewing raw material inventory semantic association feature encoder to obtain the liquor brewing raw material inventory semantic association feature vector.

[0037] It should be understood that by arranging them in a matrix, the relationships and characteristics between different raw materials can be effectively represented, providing a structured data input format for subsequent algorithms and deep learning models. The arrangement of semantically related feature vectors forms a two-dimensional matrix, where each row represents a specific raw material and each column corresponds to a dimension or attribute of the feature vector. This arrangement enables the system to simultaneously consider and compare the performance and changing trends of each raw material in different characteristics, such as inventory, procurement costs, seasonal demand, etc. By performing statistical analysis, data mining, or applying machine learning algorithms to the matrix, the patterns and laws hidden behind the inventory data can be revealed, thereby providing a basis for enterprises to formulate more accurate and effective supply chain management strategies.

[0038] Furthermore, the semantic association input matrix of liquor brewing raw material inventory is processed by the feature encoder of liquor brewing raw material inventory semantic association, aiming to further optimize and improve the information expression ability in the matrix, so as to generate more efficient and meaningful feature vectors of liquor brewing raw material inventory semantic association. The role of the feature encoder is to convert complex matrix data into a more information-dense and analyzable vector representation, and extract and strengthen key features by adopting various machine learning techniques and encoding algorithms, such as deep neural networks, autoencoders or attention mechanisms, while reducing the redundancy and noise of data. The feature encoder of liquor brewing raw material inventory semantic association can capture the correlation, trend and change between different raw materials, such as seasonal fluctuations in inventory, supply and demand balance between different raw materials, and the correlation between procurement cost and market value. By learning and optimizing the feature encoder, the system can automatically discover and utilize potential patterns in the data, thereby improving prediction accuracy and the credibility of decision-making. These feature vectors not only help to more accurately describe and quantify inventory data, but also support enterprises to make agile and intelligent operational decisions in real-time environments, such as optimizing supply chain management, adjusting production plans or controlling costs. Specifically, each layer of the white wine brewing raw material inventory semantic association feature encoder is used to perform convolution processing, mean pooling processing based on the local feature matrix and nonlinear activation processing on the input data in the forward pass of the layer, so that the last layer of the white wine brewing raw material inventory semantic association feature encoder outputs the white wine brewing raw material inventory semantic association feature vector, wherein the input of the white wine brewing raw material inventory semantic association feature encoder is the white wine brewing raw material inventory semantic association input matrix.

[0039] In a specific embodiment of the present application, the liquor brewing production planning feature extraction unit 122 includes: passing the liquor brewing production planning through a converter-based liquor brewing production planning semantic encoder to obtain a plurality of liquor brewing production planning feature vectors; passing the plurality of liquor brewing production planning feature vectors through a multi-layer perceptron-based liquor brewing production planning feature encoder to obtain the liquor brewing production planning word granularity feature vector.

[0040] It should be understood that the processing of liquor brewing production planning through a transformer-based semantic encoder aims to convert complex production planning information into multiple feature vectors with high information density and operability. Transformer models, such as Transformer and its derivative models, have a strong ability to process sequence data, can effectively capture and encode semantic and contextual information in text, and are suitable for processing complex production plan texts in liquor brewing. The generation of feature vectors for liquor brewing production planning is achieved through a semantic encoder, the core of which is to use self-attention mechanism and multi-head attention mechanism to carefully associate and encode various elements in production planning. These feature vectors can reflect key information in production plans, such as production batches, raw material requirements, production line configuration, time schedule, etc., while retaining semantic and logical consistency, which helps the system understand and interpret the complexity of production planning. Through the application of the transformer model, the system can realize efficient processing and analysis of a large amount of production planning data, helping enterprises optimize production processes, improve resource utilization efficiency, and respond to dynamic changes in market demand. This advanced semantic encoding capability not only supports real-time monitoring and adjustment of planning, but also provides decision makers with data-based decision support, such as adjusting production plans to respond to emergencies or market opportunities. Specifically, the liquor brewing production plan is segmented to obtain a production planning word sequence; the embedding layer of the converter-based liquor brewing production planning semantic encoder is used to map each production planning word in the production planning word sequence into a production planning word embedding vector to obtain a sequence of production planning word embedding vectors; the converter-based Bert model of the converter-based liquor brewing production planning semantic encoder is used to perform global contextual semantic encoding on the sequence of production planning word embedding vectors to obtain multiple liquor brewing production planning feature vectors.

[0041] Furthermore, multiple liquor brewing production planning feature vectors are processed by a feature encoder based on a multi-layer perceptron (MLP), aiming to further optimize and refine the expression of feature vectors, thereby generating more fine-grained and specific word-granular feature vectors for liquor brewing production planning. As a classic deep learning model, MLP is suitable for nonlinear data modeling and feature learning. It can effectively extract and combine the information of input features through multi-layer neural networks to express complex patterns and associations in data in a more efficient way. In the process of generating word-granular feature vectors for liquor brewing production planning, the MLP feature encoder uses the ability of multi-layer neural networks to map the original feature vectors into a more abstract and representative feature space. This process not only includes nonlinear transformation of feature vectors and deep representation learning, but also combines various detailed elements and variables in production planning, such as specific production process steps, product formula parameters, process requirements, etc., to capture richer and more accurate production planning information. Through the application of MLP, the system can analyze and understand the liquor brewing production planning in more detail, and improve the expression ability and analysis depth of planning data. These word-granularity feature vectors not only help to more accurately describe and quantify the details and changes in the production plan, but also support the intelligent decision-making system to evaluate and optimize the planning scheme. For example, in the real-time adjustment and optimization process of the production plan, these feature vectors can help enterprises quickly identify and respond to changes, thereby improving production efficiency and resource utilization. Specifically, each fully connected layer of the liquor brewing production planning feature encoder based on the multi-layer perceptron is used to fully connect the multiple liquor brewing production planning feature vectors in the forward pass of the layer to extract the high-dimensional implicit features of the feature values ​​of each position in the multiple liquor brewing production planning feature vectors.

[0042] In the above-mentioned information management system 100 of the Luzhou-flavor liquor brewing process, the inventory optimization suggestion generation module 130 is used to generate inventory optimization suggestions based on the semantic association feature vector of the liquor brewing raw material inventory and the liquor brewing production planning word granularity feature vector. It should be understood that by analyzing the semantic association feature vector of the raw material inventory, the system can deeply understand the relationship and characteristics between different raw materials, such as supply, seasonal demand, procurement cost, etc. This information helps to identify which raw materials may be in excess or shortage, and how to adjust the supply chain to optimize inventory efficiency. Using the word granularity feature vector of the liquor brewing production planning, the system can accurately capture various details and requirements in the production plan, such as production batches, process flow, product formula, etc. These feature vectors provide detailed guidance on the implementation of the production plan, which helps to predict which raw materials will increase or decrease in demand in future production cycles, and how to adjust the production plan to maximize resource utilization. According to the results of the analysis of the semantic association feature vector of the raw material inventory, the system can identify which raw materials need to increase or decrease inventory reserves. By adjusting the inventory level, enterprises can reduce inventory costs and avoid excessive reserves. Generating inventory optimization suggestions based on the semantic association feature vector of liquor brewing raw material inventory and the feature vector of liquor brewing production planning word granularity can not only improve the company's inventory management efficiency and cost control capabilities, but also support the company to maintain agility and flexibility in the highly competitive market and achieve sustainable growth and development.

[0043] Figure 4 FIG. 1 is a block diagram of an inventory optimization suggestion generation module in an information management system for a Luzhou-flavor liquor brewing process according to an embodiment of the present application. Figure 4 As shown, in a specific embodiment of the present application, the inventory optimization suggestion generation module 130 includes: a liquor brewing inventory feature fusion unit 131, which is used to fuse the liquor brewing raw material inventory semantic association feature vector and the liquor brewing production planning word granularity feature vector to obtain an inventory management suggestion generation feature vector; a liquor brewing inventory feature optimization unit 132, which is used to perform feature compatibility evaluation on the inventory management suggestion generation feature vector based on learning parameters to obtain an optimized inventory management suggestion generation feature vector; a liquor brewing inventory optimization suggestion generation unit 133, which is used to pass the optimized inventory management suggestion generation feature vector through a generator to generate an inventory optimization suggestion.

[0044] It should be understood that the fusion of the semantic association feature vector of liquor brewing raw material inventory and the feature vector of liquor brewing production planning word granularity to generate the feature vector of inventory management suggestion is intended to use the comprehensive information of the two to provide enterprises with more accurate and comprehensive inventory management suggestions. Combining the analysis of the semantic association feature vector of raw material inventory and the feature vector of production planning word granularity, the system can recommend accurate inventory level adjustment strategies. For example, in response to seasonal demand changes and market forecasts, inventory management is optimized to ensure timely supply while reducing inventory costs.

[0045] In particular, in the technical solution of the present application, it is considered that the inventory data and production planning information of liquor brewing raw materials may be very complex, involving multiple variables and parameters, such as the supply of different raw materials, production cycle, market demand, etc. Through feature mining, key features can be extracted from these data to help the system understand and model inventory management needs more accurately. Secondly, the use of deep learning models such as long short-term memory networks (LSTM) for feature extraction can effectively capture the time series relationship and semantic association between different inventory data. This method can not only handle the long-term dependency of data, but also learn the potential characteristics of data during model training, thereby generating feature vectors with more information and predictive power. Further, using the prior information of the neural network model to adjust the feature manifold of the feature vector can better adapt to specific inventory management needs and business scenarios. Neural networks have strong representation learning capabilities when processing complex data, and can discover and utilize hidden patterns and structures in the data, thereby adjusting the generated feature vectors to reflect the complexity and dynamic changes of the real world. Finally, the concept of special instance perturbation is introduced to further improve the robustness and generalization ability of the feature vector generated by inventory management suggestions. By introducing a certain amount of randomness or noise in the generation process, the model can better cope with unknown environmental changes and data noise, thereby generating more reliable and practical inventory optimization recommendations.

[0046] Among them, the feature compatibility evaluation based on the learning parameters is performed on the inventory management suggestion generation feature vector to obtain the optimized inventory management suggestion generation feature vector, including: calculating the mean and variance of the inventory management suggestion generation feature vector; calculating the difference between the eigenvalue of the predetermined position of the inventory management suggestion generation feature vector and the mean of the inventory management suggestion generation feature vector to obtain a first intermediate calculation value; multiplying the network parameter matrix by the inventory management suggestion generation feature vector and then subtracting the network parameter matrix from the inventory management suggestion generation feature vector by position to obtain a first intermediate vector, and calculating the Frobenius norm of the first intermediate vector to obtain a first norm value; adding the first norm value to the inventory management suggestion generation feature vector; After generating the variance of the eigenvector, take the square root to obtain a second intermediate calculated value; divide the first intermediate calculated value by the second intermediate calculated value to obtain a third intermediate calculated value; calculate the square of each position in the inventory management suggestion generation eigenvector and then add and sum them to obtain a first sum value; take the square root of the first sum value and divide it by the scale of the network parameter matrix and then calculate the natural exponential function value to obtain a fourth intermediate calculated value, wherein the scale of the network parameter matrix is ​​equal to the product between the width value and the height value of the network parameter matrix; multiply the fourth intermediate calculated value by the weighted hyperparameter and then add the third intermediate calculated value to obtain the eigenvalue of the predetermined position of the optimized inventory management suggestion generation eigenvector.

[0047] Among them, the feature compatibility evaluation based on the learning parameters is performed on the inventory management suggestion generation feature vector to obtain the optimized inventory management suggestion generation feature vector, including:

[0048]

[0049] Where V is the feature vector for generating inventory management recommendations, M is the network parameter matrix, β represents the weighted hyperparameter, and v i The eigenvalue of the i-th feature vector in the inventory management suggestion generation is generated, N is the scale of the network parameter matrix, the scale of the network parameter matrix is ​​equal to the product between the width value and the height value of the network parameter matrix, L is the length of the feature vector generated by the inventory management suggestion, ||·|| F represents the Frobenius norm of the feature vector generated by inventory management suggestions, represents matrix multiplication, μ represents the mean of the feature set consisting of all eigenvalues ​​of the feature vector generated by the inventory management suggestion, σ 2 represents the variance of the feature set consisting of all eigenvalues ​​of the feature vector generated by the inventory management suggestion, v' i The i-th feature value in the feature vector is generated for the optimized inventory management suggestion.

[0050] In the technical solution of the present application, considering that when the neural network model performs feature mining on the input data based on the network parameter matrix, if the inherent implicit association between the network output feature and the network parameter matrix can be further considered, the prior information of the neural network model can be further used to modulate the feature manifold of the feature vector of the inventory management suggestion generation to reduce the possibility of the occurrence of special instance disturbances of its feature value. Based on this, in a preferred embodiment, the feature compatibility evaluation based on the learning parameter is performed on the feature vector of the inventory management suggestion generation, which first maps the feature vector of the inventory management suggestion generation to the high-dimensional feature space where the network parameter matrix is ​​located, and uses the compatibility perception factor between the mapping result and the original input as the inherent implicit association descriptor between the network output feature and the network parameter matrix. Then, the statistical characteristics of the feature group of the feature vector of the inventory management suggestion generation are used as the inherent modal information of the feature vector of the inventory management suggestion generation, and the inherent implicit association descriptor is used as the instance disturbance differential adjustment variable, and the attribute values ​​of each position of the feature vector of the inventory management suggestion generation are dynamically normalized. At the same time, the class-normalized modulus data of the feature vector generated by the inventory management suggestion is used as a bias item to constrain the acceptable range of the population distribution of the feature vector generated by the inventory management suggestion. In this way, the attribute influence of the interference distribution of abnormal samples can be suppressed as much as possible, thereby improving the expression accuracy of the feature vector generated by the inventory management suggestion.

[0051] Furthermore, the feature vector generated by the optimized inventory management suggestion is passed through the generator to generate inventory optimization suggestions in order to further improve the practicality and pertinence of the suggestions. The generator model is usually based on a deep learning architecture, such as a generative adversarial network (GAN) or a variational autoencoder (VAE), and its design is to generate specific suggestions that meet the actual application needs from a high-dimensional feature vector. By processing the optimized inventory management suggestion feature vector through the generator model, the generator model can generate personalized inventory optimization suggestions based on the input inventory management optimization suggestion feature vector. These suggestions can be adjusted according to the specific needs and market environment of the enterprise to ensure the practicality and adaptability of the suggestions. The generator model is not limited to simple data conversion, but can also combine complex contextual information and real-time data to generate specific suggestions that match the current market conditions and enterprise operating conditions. For example, in the event of emergencies or seasonal demand changes that may occur in the supply chain, the generator can provide corresponding countermeasures.

[0052] In summary, the embodiment of the present application first obtains the inventory data of liquor brewing raw materials and the liquor brewing production plan collected by the database, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally generates inventory optimization suggestions through a generator, thereby helping the company to purchase or adjust production plans at the appropriate time, thereby avoiding inventory backlogs or raw material shortages, so as to improve production efficiency and quality control levels.

[0053] As described above, the information management system 100 of the brewing process of the Luzhou-flavor liquor according to the embodiment of the present application can be implemented in various terminal devices. In one example, the information management system 100 of the brewing process of the Luzhou-flavor liquor can be integrated into the terminal device as a software module and / or a hardware module. For example, the information management system 100 of the brewing process of the Luzhou-flavor liquor can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the information management system 100 of the brewing process of the Luzhou-flavor liquor can also be one of the many hardware modules of the terminal device.

[0054] Alternatively, in another example, the information management system 100 of the strong-flavor liquor brewing process and the terminal device may also be separate devices, and the information management system 100 of the strong-flavor liquor brewing process may be connected to the terminal device via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0055] Figure 5 Flow chart of the information management method of the brewing process of Luzhou-flavor liquor according to the embodiment of the present application. Figure 5 As shown, according to the information management method of the strong-flavor liquor brewing process of the embodiment of the present application, it includes: S110, acquiring the liquor brewing raw material inventory data and the liquor brewing production plan collected from the database; S120, extracting the liquor brewing raw material inventory semantic association feature vector and the liquor brewing production plan word granularity feature vector from the liquor brewing raw material inventory data collected from the database and the liquor brewing production plan; S130, generating inventory optimization suggestions based on the liquor brewing raw material inventory semantic association feature vector and the liquor brewing production plan word granularity feature vector.

[0056] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned information management method for the brewing process of Luzhou-flavor liquor have been described in detail above. Figures 1 to 4 The information management system of the Luzhou-flavor liquor brewing process has been introduced in detail, and therefore, its repeated description will be omitted.

[0057] Below, reference Figure 6 To describe an electronic device according to an embodiment of the present application.

[0058] Figure 6 It is a structural block diagram of an electronic device according to an embodiment of the present invention.

[0059] like Figure 6 As shown, the electronic device 10 includes: a processor 11 and a memory 13. The processor 11 and the memory 13 are connected, such as through a bus 12. Optionally, the electronic device 10 may also include a transceiver 14. It should be noted that in actual applications, the transceiver 14 is not limited to one, and the structure of the electronic device 10 does not constitute a limitation on the embodiments of the present invention.

[0060] The processor 11 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present invention. The processor 11 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0061] The bus 12 may include a path for transmitting information between the above components. The bus 12 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0062] The memory 13 is used to store a computer program corresponding to the information management method for the brewing process of the Luzhou-flavor liquor of the above embodiment of the present invention, and the computer program is controlled and executed by the processor 11. The processor 11 is used to execute the computer program stored in the memory 13 to implement the contents shown in the above method embodiment.

[0063] Among them, the electronic device 10 includes but is not limited to: mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The electronic device 10 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0064] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0065] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0066] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0067] In the description of the present invention, it is to be understood that the terms “center”, “longitudinal”, “lateral”, “length”, “width”, “thickness”, “up”, “down”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, “clockwise”, “counterclockwise”, “axial”, “radial”, “circumferential”, etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0068] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0069] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0070] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0071] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. An information management system for the brewing process of Luzhou-flavor liquor, characterized in that: include: The Luzhou-flavor liquor brewing inventory data acquisition module is used to obtain the liquor brewing raw material inventory data and liquor brewing production plan collected from the database; A Luzhou-flavor liquor brewing inventory data extraction module is used to extract a liquor brewing raw material inventory semantic association feature vector and a liquor brewing production plan word granularity feature vector from the liquor brewing raw material inventory data collected from the database and the liquor brewing production plan; The inventory optimization suggestion generation module is used to generate inventory optimization suggestions based on the semantic association feature vector of the liquor brewing raw material inventory and the liquor brewing production planning word granularity feature vector.

2. The information management system for the brewing process of Luzhou-flavor liquor according to claim 1, characterized in that: The Luzhou-flavor liquor brewing inventory data extraction module comprises: A liquor brewing raw material inventory feature extraction unit, used for performing feature extraction on the liquor brewing raw material inventory data collected from the database to obtain a semantic association feature vector of the liquor brewing raw material inventory; The liquor brewing production planning feature extraction unit is used to extract features from the liquor brewing production planning to obtain a word granularity feature vector of the liquor brewing production planning.

3. The information management system for the brewing process of Luzhou-flavor liquor according to claim 2, characterized in that: The liquor brewing raw material inventory feature extraction unit comprises: A raw material inventory data semantic encoding subunit is used to semantically encode the liquor brewing raw material inventory data collected from the database to obtain a plurality of liquor brewing raw material inventory semantic association feature vectors; The liquor brewing raw material inventory feature encoding subunit is used to perform feature encoding on the multiple liquor brewing raw material inventory semantic association feature vectors to obtain the liquor brewing raw material inventory semantic association feature vectors.

4. The information management system for the brewing process of Luzhou-flavor liquor according to claim 3 is characterized in that: The raw material inventory data semantic encoding subunit includes: The liquor brewing raw material inventory data collected from the database is passed through the liquor brewing raw material inventory text understanding model to obtain a liquor brewing raw material inventory word vector sequence; The word vector sequence of the liquor brewing raw material inventory is passed through the liquor brewing raw material inventory long short-term memory network based on a deep learning model to obtain the multiple liquor brewing raw material inventory semantic association feature vectors.

5. The information management system for the brewing process of Luzhou-flavor liquor according to claim 4 is characterized in that: The liquor brewing raw material inventory feature coding subunit includes: Arranging the plurality of liquor brewing raw material inventory semantic association feature vectors to obtain a liquor brewing raw material inventory semantic association input matrix; The white wine brewing raw material inventory semantic association input matrix is ​​passed through a white wine brewing raw material inventory semantic association feature encoder to obtain the white wine brewing raw material inventory semantic association feature vector.

6. The information management system for the brewing process of Luzhou-flavor liquor according to claim 5, characterized in that: The liquor brewing production planning feature extraction unit comprises: Passing the liquor brewing production plan through a liquor brewing production plan semantic encoder based on a converter to obtain a plurality of liquor brewing production plan feature vectors; The multiple liquor brewing production planning feature vectors are passed through a liquor brewing production planning feature encoder based on a multi-layer perceptron to obtain the liquor brewing production planning word granularity feature vector.

7. The information management system for the brewing process of Luzhou-flavor liquor according to claim 6 is characterized in that: The inventory optimization suggestion generation module includes: A liquor brewing inventory feature fusion unit, used to fuse the liquor brewing raw material inventory semantic association feature vector and the liquor brewing production planning word granularity feature vector to obtain an inventory management suggestion generation feature vector; A liquor brewing inventory feature optimization unit is used to perform feature compatibility evaluation on the inventory management suggestion generation feature vector based on learning parameters to obtain an optimized inventory management suggestion generation feature vector; The liquor brewing inventory optimization suggestion generating unit is used to generate the feature vector of the optimization inventory management suggestion through a generator to generate an inventory optimization suggestion.

8. The information management system for the brewing process of Luzhou-flavor liquor according to claim 7, characterized in that: The liquor brewing inventory feature optimization unit comprises: Calculating the mean and variance of the feature vector generated by the inventory management suggestion; Calculating a difference between a feature value at a predetermined position of the inventory management suggestion generation feature vector and a mean value of the inventory management suggestion generation feature vector to obtain a first intermediate calculation value; Multiplying the network parameter matrix by the feature vector generated by the inventory management suggestion, and then subtracting the feature vector from the feature vector generated by the inventory management suggestion by position to obtain a first intermediate vector, and calculating the Frobenius norm of the first intermediate vector to obtain a first norm value; Add the first norm value to the variance of the feature vector generated by the inventory management suggestion and then take the square root to obtain a second intermediate calculated value; dividing the first intermediate calculated value by the second intermediate calculated value to obtain a third intermediate calculated value; Calculating the square of each position in the inventory management suggestion generation feature vector and then accumulating and summing them to obtain a first sum value; Calculate a natural exponential function value after taking the square root of the first sum value and dividing it by the scale of the network parameter matrix to obtain a fourth intermediate calculation value, wherein the scale of the network parameter matrix is ​​equal to the product of the width value of the network parameter matrix and the height value thereof; The fourth intermediate calculated value is multiplied by the weighted hyperparameter and then added to the third intermediate calculated value to obtain a feature value at a predetermined position of the feature vector for generating the optimized inventory management suggestion.

9. An information management method for the brewing process of Luzhou-flavor liquor, characterized in that: include: Obtaining liquor brewing raw material inventory data and liquor brewing production planning collected from the database; Extracting a liquor brewing raw material inventory semantic association feature vector and a liquor brewing production plan word granularity feature vector from the liquor brewing raw material inventory data collected from the database and the liquor brewing production plan; Based on the semantic association feature vector of the liquor brewing raw material inventory and the liquor brewing production planning word granularity feature vector, an inventory optimization suggestion is generated.

10. The information management method for the brewing process of Luzhou-flavor liquor according to claim 9, characterized in that: Extracting a liquor brewing raw material inventory semantic association feature vector and a liquor brewing production plan word granularity feature vector from the liquor brewing raw material inventory data collected from the database and the liquor brewing production plan, including: Performing feature extraction on the liquor brewing raw material inventory data collected from the database to obtain a semantic association feature vector of the liquor brewing raw material inventory; Feature extraction is performed on the liquor brewing production plan to obtain a word granularity feature vector of the liquor brewing production plan.

Images