Wine warehouse environment monitoring and management application system based on intelligent sensor network

Through the intelligent sensor network and dynamic environmental impact factor analysis model, combined with multi-scale time series data fusion algorithm and quality prediction error adaptive correction mechanism, the problem of the difficulty in accurately quantifying and predicting the impact of environmental factors in the wine warehouse on the quality of wine is realized, and the wine is realized The accuracy and intelligence of library environment management.

CN118734002BActive Publication Date: 2024-11-08贵州轻工职业大学
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411187692.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-11-08
Estimated Expiration
2044-08-28

AI Technical Summary

Technical Problem

It is difficult for the prior art to accurately quantify and predict the impact of environmental factors in wine warehouses on wine quality, and traditional monitoring systems lack in-depth analysis of the relationship between environmental changes and wine quality, making it difficult to provide accurate quality prediction and management suggestions.

Method used

Using a system based on intelligent sensor network, the wine library environmental parameters are collected through sensor network units, the data is processed using a multi-scale time series data fusion algorithm, and combined with a dynamic environmental impact factor analysis model, the wine quality changes trend is predicted, and the quality prediction error adaptation is adopted. The calibration mechanism optimizes model parameters and generates an optimization plan.

Benefits of technology

It has achieved accurate quantification and prediction of the impact of environmental factors in the wine warehouse on the quality of wine products, improved the accuracy and adaptability of environmental management, and realized the intelligence and automation of environmental management in the wine warehouse.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118734002B_ABST
    Figure CN118734002B_ABST
Patent Text Reader

Abstract

The present application discloses a wine warehouse environment monitoring and management application system based on an intelligent sensor network, the system comprising: a sensor network unit for deploying a sensor network to collect wine warehouse environmental parameters, and using a multi-scale time series data fusion algorithm to process data of different time scales; a dynamic environment analysis unit for calculating the degree of influence of the environment on the quality of wine products based on a dynamic environment impact factor analysis model, and predicting the quality change trend in combination with historical data and current status; an adaptive correction unit for utilizing a quality prediction error adaptive correction mechanism, updating the dynamic environment impact factor analysis model and multi-scale time series data fusion algorithm parameters according to the prediction error, and re-evaluating the environmental conditions based on the adjusted model and generating an optimization plan. Through the solution of the present application, the technical problem that the influence of wine warehouse environmental factors on wine quality is difficult to accurately quantify and predict can be solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of management, and in particular to a wine cellar environment monitoring and management application system based on an intelligent sensor network. Background Art

[0002] Wine cellar environmental monitoring and management has become a vital part of the wine industry, which is closely related to the quality of wine, storage life, and the taste and value of the final product. The stability of the wine cellar environment directly affects the aging process of the wine, which in turn determines the quality and market value of the wine. However, not all wine cellars can maintain ideal environmental conditions, which leads to the concepts of precise environmental control wine cellars and extensive management wine cellars. Precise environmental control wine cellars can strictly control key parameters such as temperature, humidity, and light, while extensive management wine cellars may have environmental fluctuations and instability. The distinction between precise environmental control wine cellars and extensive management wine cellars involves evaluating the differences in the environmental management level of wine cellars. Although the investment cost of precise environmental control wine cellars is high, they can provide the best aging conditions for wine and minimize the negative impact of environmental fluctuations; extensive management wine cellars may face problems such as unstable environment and fluctuations in wine quality. Therefore, precise monitoring and management of the wine cellar environment is particularly important for ensuring wine quality, improving product value, and reducing loss rates, which helps to enhance the competitiveness and economic benefits of winemaking enterprises. At present, wine cellar environmental monitoring and management methods are mainly based on real-time monitoring of parameters such as temperature, humidity, light, and carbon dioxide concentration, as well as regular manual inspections and sampling tests. Although these methods are effective to a certain extent, they have certain limitations, such as the difficulty in fully capturing the complex interactions between environmental factors and the long-term cumulative effects of environmental changes on wine quality. In addition, traditional monitoring systems are usually passive responses, making it difficult to predict potential environmental problems and take proactive preventive measures.

[0003] In recent years, with the advancement of the Internet of Things and artificial intelligence technologies, intelligent wine warehouse environmental monitoring and management systems have gradually attracted attention. These systems deploy a large number of sensors to collect multi-dimensional environmental data in real time, and use machine learning algorithms to analyze these data to identify potential environmental anomalies and predict future trends. However, there are still some problems with existing intelligent monitoring systems. First, most systems only focus on a single or a few environmental parameters, ignoring the complex interactions between multiple factors. Secondly, these systems usually use fixed thresholds or simple statistical models to judge environmental anomalies, which is difficult to adapt to the special needs of different types of wine and aging stages. Furthermore, existing systems often lack in-depth analysis of the relationship between environmental changes and wine quality, making it difficult to provide accurate quality predictions and management recommendations.

[0004] Therefore, there is an urgent need for a technical solution that can solve the technical problem that the impact of wine warehouse environmental factors on wine quality is difficult to accurately quantify and predict. Summary of the invention

[0005] In order to solve the deficiencies of the prior art, the embodiment of the present application provides a wine warehouse environment monitoring and management application system based on an intelligent sensor network. The present application solves the technical problems that the impact of temperature and humidity fluctuations in the wine warehouse environment on the quality of wine products is difficult to accurately quantify and predict, and the correlation analysis between sensor data and changes in wine quality lacks an accurate model.

[0006] The embodiment of the present application provides a wine warehouse environment monitoring and management application system based on an intelligent sensor network, including: a sensor network unit, a dynamic environment analysis unit and an adaptive correction unit; wherein the sensor network unit is used to deploy a sensor network to collect wine warehouse environmental parameters, and adopt a multi-scale time series data fusion algorithm to process data of different time scales, so as to input the fused data into a dynamic environment impact factor analysis model for processing; the dynamic environment analysis unit is used to calculate the degree of influence of the environment on the quality of the wine based on the dynamic environment impact factor analysis model, and predict the quality change trend in combination with historical data and current status, and compare the predicted results with the actual detection results to trigger the quality prediction error adaptive correction mechanism; which includes: based on the dynamic environment The influencing factor analysis model calculates the comprehensive impact of multiple environmental parameters on wine quality and obtains the wine quality index at the current moment; the time series prediction model is used to predict the change trend of environmental parameters within the target time, and a variety of environmental change paths are generated through Monte Carlo simulation; the simulated environmental change path is substituted into the dynamic environmental influencing factor analysis model, the corresponding quality index sequence is calculated, and statistical analysis is performed to obtain the expected trend and confidence interval of quality change; the predicted quality change trend is compared with the actual test results to trigger the quality prediction error adaptive correction mechanism; among them, based on the dynamic environmental influencing factor analysis model, the comprehensive impact of multiple environmental parameters on wine quality is calculated to obtain the wine quality index at the current moment, including: ,

[0007] in, express The wine quality index at the moment, represents the value of the i-th environmental factor at time t, represents the basic weight of the i-th environmental factor, represents the parameter describing the periodic impact of the i-th environmental factor, represents the impact of the change rate of the jth environmental factor on quality, represents the weight of the change rate of the jth environmental factor, represents the attenuation coefficient of the jth environmental factor, represents the cumulative effect of the kth environmental factor, represents the weight of the cumulative effect of the kth environmental factor, represents the error term; the adaptive correction unit is used to utilize the quality prediction error adaptive correction mechanism to update the dynamic environmental impact factor analysis model and the multi-scale time series data fusion algorithm parameters according to the prediction error, re-evaluate the environmental conditions based on the adjusted model and generate an optimization plan.

[0008] In a possible implementation, an intelligent sensor network is deployed to collect wine warehouse environmental parameters, a multi-scale time series data fusion algorithm is used to process data at different time scales, and the fused data is input into a dynamic environmental impact factor analysis model for processing, including: deploying a variety of sensors in the wine warehouse environment to collect environmental parameters including temperature, humidity, light, air pressure and / or air quality, and the sensors are networked through wireless communication protocols; the collected sensor data is divided according to multiple time scales, and the data is pre-processed, including data cleaning, standardization and outlier removal; the pre-processed multi-scale sensor data is input into a data fusion algorithm, and short-term data fluctuations, long-term data trends and periodic changes are fused through adaptive weight calculation.

[0009] In a possible implementation, a quality prediction error adaptive correction mechanism is utilized to update the dynamic environmental impact factor analysis model and the multi-scale time series data fusion algorithm parameters according to the prediction error, and the environmental conditions are re-evaluated based on the adjusted model to generate an optimization plan, including: calculating the quality prediction error by comparing the predicted quality with the actual detection quality, and feeding the error back to the dynamic environmental impact factor analysis model; adaptively adjusting the model parameters based on the quality prediction error by using the gradient descent method; regularly optimizing the adaptive hyperparameters by using a grid search or random search method; and updating the dynamic environmental impact factor analysis model and the multi-scale time series data fusion algorithm based on the adjusted parameters and the optimized hyperparameters to re-evaluate the environmental conditions and generate an optimization plan.

[0010] In one possible implementation, the pre-processed multi-scale sensor data is input into a data fusion algorithm, and short-term data fluctuations, long-term data trends and periodic changes are fused through adaptive weight calculation, including: ,

[0011] in, represents the fused data, Indicates the current time. represents the balance factor, represents a short-term data series, represents the weight of short-term data, represents the decay factor of short-term data, timestamp representing short-term data, represents a long-term data series, represents the weight of long-term data, represents the decay factor of long-term data, timestamp representing long-term data, represents the mid-term data series, represents the weight of mid-term data, represents the period factor, Indicates the timestamp of the medium-term data, and i, j, k represent the indexes of the short-term, long-term, and medium-term data respectively.

[0012] In a possible implementation, based on the quality prediction error, a gradient descent method is used to adaptively adjust model parameters, including: ,

[0013] in, represents the updated model parameters, represents the model parameters before updating, represents the learning rate, represents the gradient of the loss function, represents the adaptive factor, represents the prediction quality, Indicates actual quality.

[0014] In a wine cellar environment monitoring and management application system based on an intelligent sensor network as provided above, the embodiment of the present application can solve the technical problem that it is difficult to accurately quantify and predict the impact of wine cellar environmental factors on wine quality through a dynamic environmental influencing factor analysis model, a multi-scale time series data fusion algorithm, and a quality prediction error adaptive correction mechanism. Furthermore, in some embodiments, the dynamic environmental influencing factor analysis model is used to accurately quantify the impact of environmental factors on wine quality, which can improve the accuracy of wine cellar environmental management. Furthermore, in some embodiments, the multi-scale time series data fusion algorithm can be used to comprehensively capture the short-term and long-term impact of environmental changes on wine quality. Furthermore, in some embodiments, the quality prediction error adaptive correction mechanism can continuously improve the accuracy and adaptability of the model, and realize the intelligent and automated wine cellar environmental management. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1A schematic block diagram of a wine cellar environment monitoring and management application system based on an intelligent sensor network provided in an embodiment of the present application;

[0017] Figure 2 A flow chart of a wine cellar environment monitoring and management application method based on an intelligent sensor network provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present application.

[0019] Those skilled in the art will appreciate that the terms "first", "second" and the like in the embodiments of the present application are only used to distinguish different steps, devices or modules, etc., and do not represent any specific technical meaning, nor do they represent the necessary logical order between them. It should also be understood that in the embodiments of the present application, "multiple" may refer to two or more, and "at least one" may refer to one, two or more. It should also be understood that for any component, data or structure mentioned in the embodiments of the present application, in the absence of explicit limitation or contrary revelation given in the context, it can generally be understood as one or more. In addition, the term "and / or" in the present application is only a kind of association relationship describing the associated objects, indicating that there can be three kinds of relationships, for example, A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present application generally indicates that the associated objects before and after are an "or" relationship. It should also be understood that the description of each embodiment in the present application emphasizes the differences between the embodiments, and the same or similar parts can refer to each other. For the sake of brevity, they will not be repeated one by one.

[0020] At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present application and its application or use. The techniques, methods and devices known to ordinary technicians in the relevant fields may not be discussed in detail, but where appropriate, the techniques, methods and devices should be considered as part of the specification. It should be noted that similar numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] Figure 1 A schematic block diagram of a wine cellar environment monitoring and management system based on an intelligent sensor network provided in an embodiment of the present application. It should be understood that the system shown in the figure is exemplary and not restrictive. This means that the system architecture involved is not limited to a specific form or design, but is presented as an example. In other words, the architecture shown in the figure can be regarded as a way of expression to clearly describe related concepts and relationships, and does not exclude other forms of architecture. Therefore, when interpreting the architecture in the picture, it should be understood that the model is flexible and diverse, and its purpose is to provide an exemplary description rather than a restrictive provision on a specific form.

[0023] The present application discloses a wine warehouse environment monitoring and management system based on an intelligent sensor network, including: a sensor network unit 101, a dynamic environment analysis unit 102 and an adaptive correction unit 103; wherein, the sensor network unit 101 is used to deploy a sensor network to collect wine warehouse environmental parameters, and adopt a multi-scale time series data fusion algorithm to process data of different time scales, so as to input the fused data into a dynamic environment impact factor analysis model for processing; the dynamic environment analysis unit 102 is used to calculate the degree of influence of the environment on the quality of the wine based on the dynamic environment impact factor analysis model, and predict the quality change trend in combination with historical data and current status, and compare the predicted results with the actual detection results, and trigger the quality prediction error adaptive correction mechanism; the adaptive correction unit 103 is used to utilize the quality prediction error adaptive correction mechanism, update the dynamic environment impact factor analysis model and the multi-scale time series data fusion algorithm parameters according to the prediction error, and re-evaluate the environmental conditions according to the adjusted model and generate an optimization plan. Next, we will combine Figure 2 The specific process of this application plan is described in detail.

[0024] Figure 2 The present invention provides a flow chart of a wine cellar environment monitoring and management method based on an intelligent sensor network. Figure 2As shown, at step S201, a sensor network is deployed to collect wine warehouse environmental parameters, and a multi-scale time series data fusion algorithm is used to process data of different time scales, so as to input the fused data into the dynamic environmental impact factor analysis model for processing. This includes: deploying a variety of sensors in the wine warehouse environment to collect environmental parameters including temperature, humidity, light, air pressure and / or air quality, and the sensors are networked through wireless communication protocols; dividing the collected sensor data according to multiple time scales, and preprocessing the data, including data cleaning, standardization and outlier removal; inputting the preprocessed multi-scale sensor data into the data fusion algorithm, and fusing short-term data fluctuations, long-term data trends and periodic changes through adaptive weight calculation.

[0025] Specifically, a series of smart sensors are deployed in the wine warehouse environment to comprehensively monitor various environmental parameters. The temperature sensor can use the PT100 platinum resistance temperature sensor with a measurement range of -50°C to 150°C and an accuracy of ±0.1°C. One is installed every 50 square meters in the wine warehouse to ensure comprehensive temperature monitoring. The humidity sensor can use the Honeywell HIH-4000 series capacitive humidity sensor with a measurement range of 0-100%RH and an accuracy of ±2%RH. It is installed in conjunction with the temperature sensor. The light sensor can use the TSL2561 light sensor with a measurement range of 0.1-40,000 Lux and be installed at key locations such as the entrance of the wine warehouse and near the windows. The air pressure sensor can use the BMP280 air pressure sensor with a measurement range of 300-1100hPa and an accuracy of ±1hPa. It is installed at different heights in the wine warehouse to monitor air pressure changes. The air quality sensor can use the PMS5003 particulate matter sensor to monitor PM2.5 and PM10 concentrations and be installed at the vents and the center of the wine warehouse. These sensors are networked via Bluetooth Low Energy (BLE) or ZigBee protocols to transmit data to the corresponding computing units.

[0026] In order to effectively process sensor data of different time scales, the present application embodiment designs an innovative multi-scale time series data fusion algorithm. The algorithm can capture the short-term and long-term impact of environmental changes on wine quality. The core formula of the algorithm is as follows: ,in, is the fused data, Represents a short-term data series (such as hourly data), represents a long-term data series (such as monthly data), represents a mid-term data series (such as weekly data), , , are the weights of short-term, long-term and medium-term data respectively. is a balancing factor used to adjust the impact ratio of short-term and long-term data. , is the decay factor, controlling the speed of the impact of short-term and long-term data, is the periodic factor, used to capture periodic changes, , , The timestamps for short-term, long-term, and medium-term data, respectively.

[0027] The implementation steps of the algorithm are as follows: Data preprocessing, cleaning and standardizing the raw sensor data to remove outliers and noise. Time scale division, dividing the data into different time scales such as hours, days, weeks, and months. Weight calculation, using an adaptive weight algorithm to dynamically calculate , , , the initial value can be determined by historical data analysis. Parameter optimization, optimized by particle swarm optimization algorithm (PSO) , , Parameters such as t(t) are used to minimize the fusion error. Data fusion: The core formula is used to perform multi-scale data fusion to obtain the comprehensive environmental index S(t). Result verification: The fusion results are compared with the actual wine quality data to evaluate the fusion effect.

[0028] In one embodiment, there are the following data: short-term temperature data D1(t) = 25°C (1 hour ago), long-term temperature trend L1(t) = 22°C (average of the past month), medium-term temperature fluctuation M1(t) = 2°C (periodic change). Substituting into the formula (α=0.7, β1=0.6, γ1=0.4, δ1=0.5, λ1=0.1, μ1=0.01, ω1=π / 12), we get S(t) ≈ 23.8°C. This fusion result comprehensively considers short-term fluctuations, long-term trends, and periodic changes.

[0029] When computing resources cannot respond quickly to computing needs, it can be simplified as follows: ,in, represents the data at time t after fusion, represents a short-term data series, represents the first weight, represents a long-term data series, represents the second weight, Represents the balance factor.

[0030] At step S202, the influence of the environment on the quality of the wine is calculated based on the dynamic environmental impact factor analysis model, the quality change trend is predicted in combination with historical data and the current state, the prediction result is compared with the actual test result, and the quality prediction error adaptive correction mechanism is triggered. This includes: based on the dynamic environmental impact factor analysis model, the comprehensive influence of multiple environmental parameters on the quality of the wine is calculated to obtain the wine quality index at the current moment; using the time series prediction model to predict the change trend of environmental parameters within the target time, and generating multiple environmental change paths through Monte Carlo simulation; substituting the simulated environmental change path into the dynamic environmental impact factor analysis model, calculating the corresponding quality index sequence, and performing statistical analysis to obtain the expected trend and confidence interval of quality change; comparing the predicted quality change trend with the actual test result to trigger the quality prediction error adaptive correction mechanism.

[0031] Specifically, the embodiment of the present application constructs a dynamic environmental impact factor analysis model based on deep learning, which can analyze the comprehensive impact of multiple environmental factors on wine quality in real time. The core formula of the model is as follows: , where Q(t) represents the wine quality index at time t, represents the value of the i-th environmental factor at time t, is the corresponding basic weight, , , is the parameter describing the periodic effect, represents the impact of the change rate of the jth environmental factor on quality, is the weight of the change rate, To affect the attenuation coefficient, represents the cumulative effect of the kth environmental factor, is the weight of the cumulative effect, is the error term, which follows the normal distribution N(0, σ²).

[0032] The model implementation steps include: data collection and preprocessing, collecting historical wine quality data and corresponding environmental parameter data, data cleaning, standardization and missing value processing, and using the sliding window method to generate time series features. Model architecture design, the current value, change rate and cumulative value of each environmental factor in the input layer, the hidden layer uses LSTM (Long Short-Term Memory Network) to capture long-term dependencies, introduces a self-attention layer, dynamically adjusts the importance of different factors, and outputs the predicted wine quality index. Model training, using the Adam optimizer, the initial learning rate is set to 0.001, using a small batch gradient descent method with a batch size of 64, using the early stopping method to prevent overfitting, patience is set to 20 epochs, and the loss function is selected as the root mean square error (RMSE). Parameter optimization, using the Bayesian optimization method to adjust hyperparameters such as the number of LSTM units and the number of hidden layers, and determining the best parameters such as ai, bi, ci, dj through cross-validation. Model evaluation and deployment, using the test set to evaluate model performance, calculating indicators such as RMSE, MAE and R², and deploying the trained model to the edge computing device to achieve real-time prediction.

[0033] In one embodiment, when only focusing on the three main environmental parameter factors of temperature (F1), humidity (F2) and light (F3), the data at a certain moment is as follows:

[0034] F1(t) = 23°C, ΔF1(t) = 0.5°C / h, ∫F1(t)dt = 550°C·h;

[0035] F2(t) = 60%, ΔF2(t) = -1% / h, ∫F2(t)dt = 1440%·h;

[0036] F3(t) = 100lux, ΔF3(t) = 20lux / h, ∫F3(t)dt = 2400lux·h.

[0037] Substituting into the formula: Q(t) = [0.5 * 23 * (1 + 0.1 * sin(0.2t + 0.5)) + 0.3 * 60 *(1 + 0.05 * sin(0.1t + 0.3)) + 0.2 * 100 * (1 + 0.02 * sin(0.3t + 0.1))] +[0.4 * (0.5 * exp(-0.01t)) + 0.3 * (-1 * exp(-0.02t)) + 0.3 * (20 * exp(-0.03t))] + [0.1 * log(550) + 0.1 * log(1440) + 0.1 * log(2400)] + ε, we get Q(t) ≈ 85.6 (when ε ≈ 0), this quality index comprehensively considers the current value, rate of change and cumulative effect of each environmental factor. When computing resources cannot respond quickly to computing needs, it can be simplified as: ,in, represents the wine quality index at time t, represents the value of the i-th environmental parameter at time t, represents the corresponding weight, represents the impact of the change rate of the jth environmental factor on quality, represents the corresponding weight, represents the error term.

[0038] Based on the dynamic environmental impact factor analysis model, the trend of wine quality changes in the future can be predicted. The prediction methods are as follows: Time series prediction uses the ARIMA model to predict the changes in environmental factors in the short term (such as 24 hours) in the future. Monte Carlo simulation takes into account the uncertainty of environmental factors, performs multiple Monte Carlo simulations, and generates possible environmental change paths. Quality index calculation substitutes the simulated environmental change path into the dynamic environmental impact factor analysis model to calculate the corresponding quality index sequence. Trend analysis performs statistical analysis on multiple simulation results to obtain the expected trend and confidence interval of quality changes.

[0039] Example of prediction result visualization:

[0040] This prediction method can provide wine warehouse managers with the possible range of future quality changes, helping to adjust environmental parameters in a timely manner and prevent quality degradation.

[0041] At step S203, the quality prediction error adaptive correction mechanism is used to update the dynamic environmental impact factor analysis model and the multi-scale time series data fusion algorithm parameters according to the prediction error, and the environmental conditions are re-evaluated and an optimization plan is generated based on the adjusted model. This includes: by comparing the predicted quality with the actual detection quality, calculating the quality prediction error, and feeding the error back to the dynamic environmental impact factor analysis model; based on the quality prediction error, using the gradient descent method to adaptively adjust the model parameters; using the grid search or random search method to regularly optimize the adaptive hyperparameters; updating the dynamic environmental impact factor analysis model and the multi-scale time series data fusion algorithm according to the adjusted parameters and optimized hyperparameters to re-evaluate the environmental conditions and generate an optimization plan.

[0042] Specifically, the system regularly compares the predicted quality with the actual detection quality, calculates the error and feeds it back to the model. The error calculation formula is as follows: ,in, To predict quality, To test the actual quality.

[0043] The embodiment of the present application designs an adaptive correction mechanism, the core formula of which is as follows: ,in, represents the model parameter vector, is the basic learning rate, is the loss function, is the adaptive factor, and are the predicted quality and the actual quality, respectively. is the error penalty index, which is used to adjust the impact of errors on learning. is the time decay factor, making recent errors have a greater impact, is the time since the last update, is the regularization coefficient, is a regularization term to prevent overfitting.

[0044] Based on error feedback, the system uses the gradient descent method to adaptively adjust the model parameters, which can also be simplified into the following adjustment formula: ,in, represents the updated model parameters, represents the model parameters before updating, represents the learning rate, represents the gradient of the loss function, represents the adaptive factor, represents the prediction quality, Indicates actual quality.

[0045] In one implementation scenario, the learning rate is adjusted dynamically. ,in is the initial learning rate, is the attenuation coefficient. The loss function is selected as follows: (mean square error), or (Mean absolute error). Adaptive factor adjustment, ,in is the initial adaptive factor, is the adjustment coefficient. Regularization, (L2 regularization) or (L1 regularization). Hyperparameter optimization, using grid search or random search methods to regularly optimize hyperparameters such as p, γ, ρ, etc.

[0046] In one embodiment, a certain prediction quality = 85, actual mass = 87, current parameters =[0.5, 0.3, 0.2], substituting into the formula (η = 0.01, λ = 0.1, p = 2, γ = 0.01, t = 7 days, ρ = 0.001): = [0.49344, 0.298064, 0.195376]. In this way, the system can adaptively adjust the model parameters according to the prediction error and continuously improve the prediction accuracy.

[0047] Next, based on the adaptive correction mechanism, the system will regularly update the relevant parameters of the dynamic environmental impact factor analysis model and the multi-scale time series data fusion algorithm. The update frequency can be adjusted according to actual conditions, such as a comprehensive update once a week or a month. After the update, the system will re-evaluate the current environmental conditions and generate optimization suggestions. These suggestions may include: temperature and humidity control strategy adjustment, lighting management optimization, air quality improvement measures, wine storage location adjustment, detection frequency and sampling point optimization. In order to quantify the potential effect of the optimization suggestions, this embodiment of the application introduces the benefit-cost ratio (BCR) indicator: ,in, is the expected quality improvement, V is the value of unit quality improvement, and C is the implementation cost. The system will prioritize optimization measures with higher BCR to ensure effective use of resources.

[0048] Furthermore, this process includes: state assessment, using the updated model to recalculate the current wine quality index. Sensitivity analysis, by slightly changing each environmental factor, analyzing their impact on the quality index. Optimization goal setting: setting the optimization objective function according to the wine warehouse management goals (such as maximizing quality or minimizing energy consumption or BCR indicators, etc.). Optimization algorithm application, using optimization algorithms (such as genetic algorithms or particle swarm optimization) to find the optimal combination of environmental parameters. Recommendation generation, converting the optimization results into specific management recommendations, such as adjusting the temperature setting value, changing the ventilation strategy, etc. In this way, the system can not only continuously improve its own prediction accuracy, but also provide timely and effective decision-making support for wine warehouse managers.

[0049] In summary, this solution achieves accurate quantification of the impact of environmental factors on wine quality through a dynamic environmental influencing factor analysis model, which can improve the accuracy of wine warehouse environmental management. Furthermore, in some embodiments, through a multi-scale time series data fusion algorithm, the short-term and long-term impact of environmental changes on wine quality can be fully captured. Furthermore, in some embodiments, through a quality prediction error adaptive correction mechanism, the accuracy and adaptability of the model can be continuously improved, realizing intelligent and automated wine warehouse environmental management.

[0050] Furthermore, an embodiment of the present application also provides a wine warehouse environment monitoring and management device based on an intelligent sensor network, including: a processor, a memory, and a system bus; the processor and the memory are connected through the system bus; the memory is used to store one or more programs, and the one or more programs include instructions, and when the instructions are executed by the processor, the processor executes any of the above methods.

[0051] Furthermore, an embodiment of the present application also provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute any of the above methods.

[0052] It can be known from the description of the above implementation mode that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., including several instructions for enabling a computer device (which can be a personal computer, a server, or a network communication device such as a media gateway, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.

[0053] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0054] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0055] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A wine warehouse environment monitoring and management application system based on intelligent sensor network, characterized in that: include: A sensor network unit, a dynamic environment analysis unit and an adaptive correction unit; wherein, The sensor network unit is used to deploy a sensor network to collect wine warehouse environmental parameters, and use a multi-scale time series data fusion algorithm to process data of different time scales, so as to input the fused data into a dynamic environmental impact factor analysis model for processing; the method includes: deploying a variety of sensors in the wine warehouse environment to collect environmental parameters including temperature, humidity, light, air pressure and / or air quality, and the sensors are networked through a wireless communication protocol; dividing the collected sensor data according to multiple time scales, and preprocessing the data, including data cleaning, standardization and outlier removal; inputting the preprocessed multi-scale sensor data into a data fusion algorithm, and fusing short-term data fluctuations, long-term data trends and periodic changes through adaptive weight calculation; Among them, the pre-processed multi-scale sensor data is input into the data fusion algorithm, and the short-term data fluctuations, long-term data trends and periodic changes are fused through adaptive weight calculation, including: ,in, represents the fused data, Indicates the current time. represents the balance factor, represents a short-term data series, represents the weight of short-term data, represents the decay factor of short-term data, timestamp representing short-term data, represents a long-term data series, represents the weight of long-term data, represents the decay factor of long-term data, timestamp representing long-term data, represents the mid-term data series, represents the weight of mid-term data, represents the period factor, Indicates the timestamp of the medium-term data, i, j, k represent the indexes of the short-term, long-term and medium-term data respectively; The dynamic environment analysis unit is used to calculate the degree of influence of the environment on the quality of wine based on the dynamic environment influence factor analysis model, predict the quality change trend in combination with historical data and current status, compare the prediction result with the actual test result, and trigger the quality prediction error adaptive correction mechanism; which includes: based on the dynamic environment influence factor analysis model, calculate the comprehensive influence of multiple environmental parameters on the quality of wine, and obtain the wine quality index at the current moment; use the time series prediction model to predict the change trend of environmental parameters within the target time, and generate multiple environmental change paths through Monte Carlo simulation; substitute the simulated environmental change path into the dynamic environment influence factor analysis model, calculate the corresponding quality index sequence, and perform statistical analysis to obtain the expected trend and confidence interval of quality change; compare the predicted quality change trend with the actual test result to trigger the quality prediction error adaptive correction mechanism; Among them, based on the dynamic environmental impact factor analysis model, the comprehensive impact of multiple environmental parameters on wine quality is calculated to obtain the wine quality index at the current moment, including: , in, express The wine quality index at the moment, represents the value of the i-th environmental factor at time t, represents the basic weight of the i-th environmental factor, represents the parameter describing the periodic impact of the i-th environmental factor, represents the impact of the change rate of the jth environmental factor on quality, represents the weight of the change rate of the jth environmental factor, represents the attenuation coefficient of the jth environmental factor, represents the cumulative effect of the kth environmental factor, represents the weight of the cumulative effect of the kth environmental factor, represents the error term; The adaptive correction unit is used to utilize the quality prediction error adaptive correction mechanism to update the dynamic environmental impact factor analysis model and multi-scale time series data fusion algorithm parameters according to the prediction error, re-evaluate the environmental conditions based on the adjusted model and generate an optimization plan.

2. The system according to claim 1, characterized in that in, Using the adaptive correction mechanism of quality prediction error, the dynamic environmental impact factor analysis model and multi-scale time series data fusion algorithm parameters are updated according to the prediction error, and the environmental conditions are re-evaluated based on the adjusted model and an optimization plan is generated, including: By comparing the predicted quality with the actual detected quality, the quality prediction error is calculated, and the error is fed back to the dynamic environmental impact factor analysis model; Based on the quality prediction error, a gradient descent method is used to adaptively adjust model parameters; Regularly optimize adaptive hyperparameters using grid search or random search methods; According to the adjusted parameters and optimized hyperparameters, the dynamic environmental impact factor analysis model and the multi-scale time series data fusion algorithm are updated to re-evaluate the environmental conditions and generate optimization solutions.

3. The system according to claim 2, characterized in that in, Based on the quality prediction error, the model parameters are adaptively adjusted using the gradient descent method, including: , in, represents the updated model parameters, represents the model parameters before updating, represents the learning rate, represents the gradient of the loss function, represents the adaptive factor, represents the prediction quality, Indicates actual quality.