Methods, devices and electronic equipment for synergistic optimization of tomato refrigeration quality and refrigeration emissions
By collecting and analyzing multi-source data on cold storage environment, tomato quality, and refrigeration equipment status, and utilizing an improved xPatch time-series prediction model and differential evolution algorithm, automated intelligent control of the cold storage was achieved. This solved the problems of quality deterioration and increased energy consumption caused by temperature and humidity fluctuations in the cold storage, and extended the shelf life of tomatoes while reducing energy consumption.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING RES CENT FOR INFORMATION TECH & AGRI
- Filing Date
- 2025-11-04
- Publication Date
- 2026-06-30
AI Technical Summary
Existing cold storage facilities experience frequent fluctuations in temperature and humidity during tomato refrigeration, leading to quality deterioration and increased energy consumption. The lack of a dynamic response mechanism results in chilling injury and increased carbon emissions.
By collecting multi-source data, performing preprocessing and correlation analysis, extracting key feature parameters, and using the improved xPatch time-series prediction model and differential evolution algorithm, an objective function is constructed for rolling solution to obtain the optimal control strategy and realize the automated intelligent control of cold storage.
It extends the shelf life of tomatoes, reduces refrigeration energy consumption and carbon emissions, improves the level of intelligence in cold storage operation, and reduces human intervention.
Smart Images
Figure CN121684249B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of cold chain storage and low-carbon control technology for agricultural products, and in particular to a method, apparatus and electronic equipment for synergistic optimization of tomato cold storage quality and refrigeration emissions. Background Technology
[0002] Tomatoes are physiologically active after harvesting and are extremely sensitive to fluctuations in the cold storage environment. They need to be stored under low temperature and high humidity conditions to delay ripening and reduce losses. However, in the pre-cooling process at the production site and in the distribution process, there are common problems such as aging cold storage equipment, insufficient control precision, and non-standard operation, which lead to frequent fluctuations in temperature and humidity inside the storage. Such fluctuations not only accelerate the deterioration of tomatoes such as water loss, softening, and rotting, resulting in high loss rates in the distribution process, but also cause increased energy consumption and carbon emissions due to frequent start-ups and over-operation of the refrigeration system.
[0003] Currently, most cold storage facilities still operate using fixed temperature settings, lacking dynamic response mechanisms based on changes in fruit condition and environmental load. There is no effective linkage between environmental parameters, tomato quality indicators, and the operating conditions of refrigeration equipment, and a lack of closed-loop control capabilities for synergistic optimization. This often results in a mismatch between temperature control and fruit ripening progress, potentially inducing chilling injury and easily leading to over-refrigeration, energy waste, and increased carbon emissions. This is particularly pronounced under conditions of significant external climate fluctuations or load changes, further exacerbating the contradiction between cold chain expansion and carbon emission control. Summary of the Invention
[0004] This application provides a method, apparatus, and electronic device for synergistic optimization of tomato cold storage quality and refrigeration emissions, aiming to extend the shelf life of tomatoes, reduce refrigeration energy consumption, and reduce emissions.
[0005] In a first aspect, this application provides a method for synergistic optimization of tomato refrigeration quality and refrigeration emissions, including:
[0006] Collect and preprocess multi-source data characterizing the cold storage environment, tomato quality, and refrigeration equipment operation status.
[0007] Correlation analysis was performed on the preprocessed multi-source data to extract key characteristic parameters related to tomato quality and the refrigeration process;
[0008] Based on the extracted key feature parameters and the improved xPatch time-series prediction model, tomato quality-related indicators and refrigeration emission-related indicators are predicted.
[0009] An objective function related to the tomato quality indicators and the refrigeration emission indicators is constructed, and the objective function is solved by a differential evolution algorithm to obtain the optimal control strategy for the refrigeration environment parameters.
[0010] The optimal control strategy is adjusted and sent to the cold storage control system for execution.
[0011] Secondly, this application also provides a device for synergistic optimization of tomato refrigeration quality and refrigeration emissions, comprising:
[0012] The data acquisition and preprocessing module is used to collect and preprocess multi-source data that characterizes the cold storage environment, tomato quality, and the operating status of refrigeration equipment.
[0013] The correlation analysis module is used to perform correlation analysis on the preprocessed multi-source data and extract key characteristic parameters related to tomato quality and the refrigeration process.
[0014] The model prediction module is used to predict tomato quality-related indicators and refrigeration emission-related indicators based on extracted key feature parameters and a time-series prediction model based on improved xPatch.
[0015] The optimization solution module is used to construct objective functions related to the tomato quality indicators and the refrigeration emission indicators, and to use the differential evolution algorithm to solve the objective functions in a rolling manner to obtain the optimal control strategy for the refrigeration environment parameters.
[0016] The strategy execution module is used to distribute the optimal control strategy to the cold storage control system for execution.
[0017] Thirdly, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the above-described methods for synergistic optimization of tomato refrigeration quality and refrigeration emissions.
[0018] Fourthly, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements any of the above-described methods for synergistic optimization of tomato refrigeration quality and refrigeration emissions.
[0019] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements any of the above-described methods for synergistic optimization of tomato refrigeration quality and refrigeration emissions.
[0020] The tomato cold storage quality and refrigeration emissions co-optimization method, apparatus, and electronic equipment provided in this application improve data quality by collecting and preprocessing multi-source data characterizing the cold storage environment, tomato quality, and refrigeration equipment operating status, laying the foundation for the accuracy of subsequent model predictions. Correlation analysis of the preprocessed multi-source data extracts key feature parameters related to tomato quality and the refrigeration process, reducing redundant information in the model input and improving model training efficiency and prediction accuracy. Based on the extracted key feature parameters and an improved xPatch-based time-series prediction model, tomato quality-related indicators and refrigeration emissions-related indicators are predicted. This system effectively captures trends and seasonal changes in time-series data, improving prediction accuracy. It constructs objective functions related to tomato quality and refrigeration emissions, and uses a differential evolution algorithm to solve these functions in a rolling manner to obtain the optimal control strategy for cold storage environment parameters. This achieves the goal of minimizing refrigeration energy consumption and emissions while ensuring tomato quality. The differential evolution algorithm has global search capabilities to find the optimal control strategy, and the rolling solution adapts to changes in the environment and tomato state during cold storage, ensuring the effectiveness of the control strategy. By distributing the optimal control strategy to the cold storage control system, automated intelligent control of the cold storage is achieved, reducing manual intervention. Overall, this system extends the shelf life of tomatoes, reduces refrigeration energy consumption and emissions under complex cold storage conditions, and improves the intelligence level of cold storage operation. Attached Figure Description
[0021] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0022] Figure 1 This is a flowchart illustrating the method for synergistic optimization of tomato cold storage quality and refrigeration emissions provided in the embodiments of this application;
[0023] Figure 2 This is a schematic diagram of the structure of the time series prediction model based on the improved xPatch provided in the embodiments of this application;
[0024] Figure 3 This is a schematic diagram of the structure of the tomato cold storage quality and refrigeration emission synergistic optimization device provided in the embodiments of this application;
[0025] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation
[0026] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0027] Figure 1 This is a flowchart illustrating the method for synergistic optimization of tomato refrigeration quality and refrigeration emissions provided in this application, as shown below. Figure 1 As shown, the method includes the following steps:
[0028] S101. Collect and preprocess multi-source data characterizing the cold storage environment, tomato quality, and refrigeration equipment operation status.
[0029] S102. Perform correlation analysis on the preprocessed multi-source data to extract key characteristic parameters related to tomato quality and refrigeration process;
[0030] S103. Based on the extracted key feature parameters and the improved xPatch time-series prediction model, predict tomato quality-related indicators and refrigeration emission-related indicators.
[0031] S104. Construct an objective function related to the tomato quality indicators and the refrigeration emission indicators, and use the differential evolution algorithm to solve the objective function in a rolling manner to obtain the optimal control strategy for the refrigeration environment parameters.
[0032] S105. The optimal control strategy is adjusted and sent to the cold storage control system for execution.
[0033] The tomato cold storage quality and refrigeration emissions co-optimization method provided in this application improves data quality by collecting and preprocessing multi-source data characterizing the cold storage environment, tomato quality, and refrigeration equipment operation status, laying the foundation for the accuracy of subsequent model predictions. Correlation analysis of the preprocessed multi-source data extracts key feature parameters related to tomato quality and the refrigeration process, reducing redundant information in the model input and improving model training efficiency and prediction accuracy. Based on the extracted key feature parameters and an improved xPatch-based time-series prediction model, tomato quality-related indicators and refrigeration emissions-related indicators are predicted. The improved xPatch-based time-series prediction model can effectively... By capturing trends and seasonal variations in time-series data, prediction accuracy is improved. An objective function related to tomato quality and refrigeration emissions is constructed, and a differential evolution algorithm is used to solve this objective function in a rolling manner to obtain the optimal control strategy for the cold storage environment parameters. This achieves the goal of minimizing refrigeration energy consumption and emissions while ensuring tomato quality. The differential evolution algorithm has global search capabilities to find the optimal control strategy, and the rolling solution can adapt to changes in the environment and tomato state during cold storage, ensuring the effectiveness of the control strategy. By distributing the optimal control strategy to the cold storage control system for execution, automated intelligent control of the cold storage is achieved, reducing manual intervention. Overall, this results in extended tomato shelf life, reduced refrigeration energy consumption and emissions, and improved the intelligence level of cold storage operation.
[0034] S101. Collect and preprocess multi-source data characterizing the cold storage environment, tomato quality, and refrigeration equipment operation status.
[0035] Specifically, a sensor network deployed in the cold storage space, among tomato cargo, and within the refrigeration unit simultaneously collects multi-source data and other data characterizing the cold storage environment, tomato quality, and the operational status of the refrigeration equipment. Data characterizing the cold storage environment includes air temperature, relative humidity, and carbon dioxide concentration; data characterizing tomato quality includes tomato weight, internal fruit temperature, and internal humidity of the packaging; and data characterizing the equipment's operational status includes compressor voltage, current, power factor, and supply and return air temperatures. After multi-source data collection, necessary preprocessing is performed to improve data quality and lay a solid data foundation for the accuracy of subsequent model predictions.
[0036] The sensor network includes temperature and humidity sensors, carbon dioxide concentration sensors, electronic scales (for measuring tomato weight), internal fruit temperature sensors, and voltage, current, and power factor sensors for the refrigeration unit, all deployed within the cold storage facility.
[0037] In some embodiments, preprocessing includes timestamp alignment, outlier handling, missing value handling, calculation of tomato quality-related indicators, calculation of refrigeration emission-related indicators, and normalization.
[0038] Specifically, timestamp alignment can include aligning sensor data from different sources with 1-minute timestamps and resampling.
[0039] Outlier handling may specifically include: using the interquartile range method to identify outliers and uniformly marking outliers as missing.
[0040] Missing value handling can specifically include: using linear interpolation to fill in missing values for gradual parameters (such as air temperature), and using forward filling or nearest neighbor filling to fill in fluctuating parameters (such as compressor voltage and current).
[0041] One specific quality-related indicator for tomatoes is the water loss rate, which characterizes the quality deterioration of tomatoes during storage. The calculation of this indicator can be based on a mass sequence recorded by an electronic scale; the water loss rate at any point in time during the storage period is calculated using the following formula:
[0042]
[0043] in, The water loss rate at any point in time during the storage period. For the initial mass, For time points The measured mass.
[0044] Specifically, refrigeration emission-related indicators can be refrigeration emissions. The calculation of these indicators can include: calculating the cumulative electricity consumption of the chiller during the storage period based on voltage, current, and power factor collected by the cold storage operation monitoring system; and then combining this with the grid emission factor to obtain the first carbon refrigeration emission, satisfying the following conditions:
[0045]
[0046]
[0047] in, To calculate the cumulative electricity consumption, For time points voltage, For time points The current, cosφ For power factor, The sampling time interval, The first refrigeration emission, EF ele The power grid emission factor is (kgCO2 / kWh).
[0048] Furthermore, the calculated first carbon emission is added to the second carbon emission caused by tomato quality loss to obtain the total refrigeration emission per unit mass of tomatoes during the storage period. The second carbon emission caused by tomato quality loss is obtained by multiplying the water loss mass of tomatoes during the storage period by the carbon emission coefficient of the tomato production stage.
[0049] Normalization processing specifically includes: performing Min-Max normalization on each data point; integrating all normalized data with the calculated tomato quality-related indicators and refrigeration emission-related indicators; and further normalizing the data to construct a multi-source dataset for the synergistic optimization of tomato cold storage quality and refrigeration emissions. The processing procedure is as follows:
[0050]
[0051] in, The time series data after normalization. This is the original time series data. and They are respectively The maximum and minimum values.
[0052] S102. Perform correlation analysis on the preprocessed multi-source data to extract key characteristic parameters related to tomato quality and the refrigeration process.
[0053] Specifically, by performing correlation analysis on the preprocessed multi-source data, key feature parameters related to tomato quality and the refrigeration process can be extracted, which can reduce redundant information in the model input and improve model training efficiency and prediction accuracy.
[0054] In some embodiments, S102 may specifically include:
[0055] Distance correlation analysis is used to determine the degree of correlation between multi-source data and the prediction target, including linear correlation and nonlinear correlation.
[0056] Filter the key feature parameters whose correlation is greater than a preset threshold, and integrate them according to the time series format;
[0057] The prediction targets include tomato quality-related indicators and refrigeration emission-related indicators.
[0058] Specifically, the distance correlation analysis method was used to calculate the correlation between the preprocessed multi-source data and the two prediction targets: tomato quality-related indicators and refrigeration emission-related indicators. The correlation included linear and nonlinear correlation.
[0059] Based on the correlation results obtained from distance correlation analysis, key feature parameters highly correlated with the prediction target are selected, while feature parameters with correlation below a preset threshold are removed. The key feature parameters selected based on the preset threshold are then integrated in a unified time series format to construct the input feature dataset for the final time series prediction model.
[0060] S103. Based on the extracted key feature parameters and the improved xPatch time-series prediction model, predict tomato quality-related indicators and refrigeration emission-related indicators.
[0061] Specifically, the extracted key feature parameters are input into a pre-trained time-series prediction model based on the improved xPatch. The time-series prediction model based on the improved xPatch is used to predict tomato quality-related indicators and refrigeration emission-related indicators.
[0062] Figure 2 This is a schematic diagram of the structure of the time series prediction model based on the improved xPatch provided in the embodiments of this application, as shown below. Figure 2 As shown, the time series prediction model based on the improved xPatch includes a trend flow branch (multilayer perceptron) and a seasonal flow branch (lightweight temporal convolutional network). The trend flow branch is used to extract long-term trend features of multi-source data, and the seasonal flow branch is used to extract short-term periodic features of multi-source data. The features are fused through a two-stage gating and channel attention mechanism.
[0063] The construction and training process of the time series prediction model based on the improved xPatch is as follows:
[0064] Step a: Divide the preprocessed multi-source dataset into training set, validation set and test set in a ratio of 8:1:1.
[0065] Step b: Construct a time series prediction model based on the improved xPatch, with the model structure as follows: Figure 2 As shown, this time series prediction model first performs trend-seasonal decomposition on the input time series data, and then uses a trend flow branch containing a multilayer perceptron and a seasonal flow branch containing a lightweight temporal convolutional network for dual-path feature extraction. Finally, the features are fused through a two-stage gating and channel attention mechanism.
[0066] Step c: Using the training set, train the model using a joint loss function. During training, apply an optimizer (e.g., Adam), learning rate decay, and early stopping strategy, and optimize hyperparameters based on the validation set. Optionally, the joint loss function is a weighted joint loss function composed of smoothing loss and temporal difference loss.
[0067] Step d: Use the trained model to predict tomato quality-related indicators (specifically, tomato water loss rate) and refrigeration emission-related indicators (specifically, refrigeration emission amount), and use the mean absolute error and root mean square error indicators to compare the prediction results with the prediction results of multiple benchmark models.
[0068] It should be noted that the model training process can be carried out according to a certain time period. After the initial training is completed, the model can be trained and updated regularly based on real-time collected multi-source data.
[0069] S104. Construct an objective function related to the tomato quality indicators and the refrigeration emission indicators, and use the differential evolution algorithm to solve the objective function in a rolling manner to obtain the optimal control strategy for the refrigeration environment parameters.
[0070] Specifically, an objective function related to the tomato quality indicators and the refrigeration emission indicators is constructed. A differential evolution algorithm is used to solve the objective function in a rolling manner to obtain the optimal control strategy for the cold storage environment parameters. This achieves the goal of minimizing refrigeration energy consumption and emissions while ensuring tomato quality. The differential evolution algorithm has global search capabilities to find the optimal control strategy, and the rolling solution can adapt to changes in the environment and tomato state during cold storage, ensuring the effectiveness of the control strategy. By distributing the optimal control strategy to the cold storage control system for execution, automated intelligent control of the cold storage is achieved, reducing manual intervention.
[0071] In some embodiments, the objective function aims to minimize a weighted function of tomato quality-related indicators and refrigeration emission-related indicators, satisfying the following:
[0072]
[0073] in, The objective function value, For the refrigeration emissions-related indicators predicted by the model, These are the tomato quality-related indicators predicted by the model. and These are the weighting coefficients preset according to the warehousing strategy.
[0074] In some embodiments, before using the differential evolution algorithm to solve the objective function in S104, the following steps may be included:
[0075] Step a: Determine the decision variables of the objective function, including the setpoints of key characteristic parameters that can be controlled and characterize the cold storage environment and the operating status of the refrigeration equipment. From the key characteristic parameters extracted in S102, select the setpoints of key characteristic parameters that can be directly controlled by the cold storage control system and characterize the cold storage environment and the operating status of the refrigeration equipment. The setpoints of controllable key characteristic parameters characterizing the cold storage environment include setpoints for temperature, humidity, etc., and the setpoints of key characteristic parameters characterizing the operating status of the refrigeration equipment include the upper and lower limits of the fan operating frequency and the defrost trigger threshold, etc.
[0076] Step b: Determine the constraints of the optimization process, including cold storage environment constraints, refrigeration equipment safety constraints, and constraints on the variation range of control variables between adjacent control cycles. Cold storage environment constraints mainly refer to temperature and humidity constraints; the temperature and humidity must be maintained within the suitable storage range for tomatoes, for example, temperature maintained at 10-13℃ and humidity maintained at 85-95%. Refrigeration equipment safety constraints mean that the compressor start-stop interval and the upper and lower limits of the fan frequency must meet the equipment operation safety requirements, for example, the compressor start-stop interval should be greater than 15 minutes, and the fan frequency should be between 30-60Hz. Constraints on the variation range of control variables between adjacent control cycles mean that the variation range of control variables between adjacent control cycles must not exceed a preset threshold to avoid frequent equipment adjustments, for example, the temperature change between adjacent control cycles should not exceed 1℃, and the humidity change should not exceed 2%.
[0077] The process of using the differential evolution algorithm to solve the objective function in a rolling manner generally includes steps such as population initialization, iterative update, fitness evaluation, and output of the optimal solution, ultimately obtaining the optimal control strategy that satisfies the constraints.
[0078] The specific implementation process of using the differential evolution algorithm to solve the objective function in a rolling manner can include:
[0079] Initialization phase: At the beginning of each control cycle, an initial population is generated based on the current system state and the key feature parameters extracted by S102. Specifically, uniform sampling or low-difference sequence methods can be used to ensure the distribution of solutions.
[0080] Iterative phase: Mutation, crossover, and selection operations are performed on the population to gradually optimize the candidate solutions.
[0081] Fitness evaluation: Input the feature parameters corresponding to each candidate solution into the time series prediction model in S103 to predict the tomato quality-related indicators and refrigeration emission-related indicators corresponding to the control solution, and substitute them into the objective function to calculate the fitness, as shown in the following formula;
[0082]
[0083] in, J The objective function value, FitnessThe larger the value, the better the candidate solution.
[0084] Convergence criterion: Optimization is terminated when the number of iterations reaches the preset upper limit or when the fitness improvement is below the threshold for several consecutive generations.
[0085] Output the optimal solution: Select the candidate solution with the highest fitness as the optimal control strategy for this control cycle.
[0086] S105. The optimal control strategy is adjusted and sent to the cold storage control system for execution.
[0087] Specifically, after the control cycle ends, multi-source data is re-collected, and the training dataset of the model is updated based on the key feature parameters obtained in S102, and the next round of optimization is entered, forming a closed-loop rolling optimization mechanism of prediction-optimization-execution-feedback.
[0088] The following describes the tomato cold storage quality and refrigeration emission synergistic optimization device provided in this application. The tomato cold storage quality and refrigeration emission synergistic optimization device described below and the tomato cold storage quality and refrigeration emission synergistic optimization method described above can be referred to in correspondence.
[0089] Figure 3 This is a schematic diagram of the structure of the tomato cold storage quality and refrigeration emission synergistic optimization device provided in the embodiments of this application, as shown below. Figure 3 As shown, the device includes:
[0090] The data acquisition and preprocessing module 301 is used to acquire and preprocess multi-source data characterizing the cold storage environment, tomato quality and the operating status of refrigeration equipment.
[0091] The correlation analysis module 302 is used to perform correlation analysis on the preprocessed multi-source data and extract key characteristic parameters related to tomato quality and refrigeration process.
[0092] The model prediction module 303 is used to predict tomato quality-related indicators and refrigeration emission-related indicators based on the extracted key feature parameters and the improved xPatch time-series prediction model.
[0093] The optimization solution module 304 is used to construct an objective function related to the tomato quality-related indicators and the refrigeration emission-related indicators, and to use the differential evolution algorithm to solve the objective function in a rolling manner to obtain the optimal control strategy for the refrigeration environment parameters.
[0094] The strategy execution module 305 is used to send the optimal control strategy adjustment to the cold storage control system for execution.
[0095] The improved xPatch-based time-series prediction model includes a trend flow branch and a seasonal flow branch. The trend flow branch is used to extract long-term trend features from multi-source data, and the seasonal flow branch is used to extract short-term periodic features from multi-source data. The features are then fused through a two-stage gating and channel attention mechanism.
[0096] In some embodiments, the model prediction module 303 is specifically used for:
[0097] The preprocessed multi-source data is divided into training set, validation set and test set;
[0098] The improved xPatch-based time series prediction model is trained based on the training set and the joint loss function, and the hyperparameters are optimized based on the validation set.
[0099] Based on a trained time-series prediction model based on improved xPatch, tomato quality-related indicators and refrigeration emission-related indicators are predicted, and the prediction results are compared with the prediction results of at least one benchmark model.
[0100] In some embodiments, the training of the time-series prediction model based on the improved xPatch employs a joint loss function, which includes a smoothing loss term and a temporal difference loss term.
[0101] In some embodiments, the optimization solution module 304 is further configured to:
[0102] The decision variables for the objective function are determined, including the set values of key characteristic parameters that can be controlled and characterize the cold storage environment and the operating status of the refrigeration equipment.
[0103] Determine the constraints of the optimization process, including cold storage environment constraints, refrigeration equipment safety constraints, and control variable variation range constraints between adjacent control cycles.
[0104] In some embodiments, the correlation analysis module 302 is specifically used for:
[0105] Distance correlation analysis is used to determine the degree of correlation between multi-source data and the prediction target, including linear correlation and nonlinear correlation.
[0106] Filter the key feature parameters whose correlation is greater than a preset threshold, and integrate them according to the time series format;
[0107] The prediction targets include tomato quality-related indicators and refrigeration emission-related indicators.
[0108] In some embodiments, the tomato quality-related indicators include tomato water loss rate, and the refrigeration emission-related indicators include refrigeration emissions.
[0109] In some embodiments, the preprocessing includes timestamp alignment, outlier handling, missing value handling, calculation of tomato quality-related indicators, calculation of refrigeration emission-related indicators, and normalization.
[0110] It should be noted that the tomato cold storage quality and refrigeration emission synergistic optimization device provided in this application embodiment can realize all the method steps implemented in the above method embodiment and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiment and the beneficial effects will not be described in detail.
[0111] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application, such as... Figure 4 As shown, the electronic device may include: a processor 401, a communication interface 402, a memory 403, and a communication bus 804. The processor 401, communication interface 402, and memory 403 communicate with each other via the communication bus 404. The processor 401 can call logical instructions from the memory 403 to execute a method for the coordinated optimization of tomato refrigeration quality and refrigeration emissions. This method includes:
[0112] Collect and preprocess multi-source data characterizing the cold storage environment, tomato quality, and refrigeration equipment operation status.
[0113] Correlation analysis was performed on the preprocessed multi-source data to extract key characteristic parameters related to tomato quality and the refrigeration process;
[0114] Based on the extracted key feature parameters and the improved xPatch time-series prediction model, tomato quality-related indicators and refrigeration emission-related indicators are predicted.
[0115] An objective function related to the tomato quality indicators and the refrigeration emission indicators is constructed, and the objective function is solved by a differential evolution algorithm to obtain the optimal control strategy for the refrigeration environment parameters.
[0116] The optimal control strategy is adjusted and sent to the cold storage control system for execution.
[0117] Furthermore, the logical instructions in the aforementioned memory 403 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0118] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the tomato cold storage quality and refrigeration emission synergistic optimization method provided by the above methods, the method including:
[0119] Collect and preprocess multi-source data characterizing the cold storage environment, tomato quality, and refrigeration equipment operation status.
[0120] Correlation analysis was performed on the preprocessed multi-source data to extract key characteristic parameters related to tomato quality and the refrigeration process;
[0121] Based on the extracted key feature parameters and the improved xPatch time-series prediction model, tomato quality-related indicators and refrigeration emission-related indicators are predicted.
[0122] An objective function related to the tomato quality indicators and the refrigeration emission indicators is constructed, and the objective function is solved by a differential evolution algorithm to obtain the optimal control strategy for the refrigeration environment parameters.
[0123] The optimal control strategy is adjusted and sent to the cold storage control system for execution.
[0124] Furthermore, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method for synergistic optimization of tomato cold storage quality and refrigeration emissions provided by the methods described above, the method comprising:
[0125] Collect and preprocess multi-source data characterizing the cold storage environment, tomato quality, and refrigeration equipment operation status.
[0126] Correlation analysis was performed on the preprocessed multi-source data to extract key characteristic parameters related to tomato quality and the refrigeration process;
[0127] Based on the extracted key feature parameters and the improved xPatch time-series prediction model, tomato quality-related indicators and refrigeration emission-related indicators are predicted.
[0128] An objective function related to the tomato quality indicators and the refrigeration emission indicators is constructed, and the objective function is solved by a differential evolution algorithm to obtain the optimal control strategy for the refrigeration environment parameters.
[0129] The optimal control strategy is adjusted and sent to the cold storage control system for execution.
[0130] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0131] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.
Claims
1. A method for synergistic optimization of tomato cold storage quality and refrigeration discharge, characterized in that, include: Collect and preprocess multi-source data characterizing the cold storage environment, tomato quality, and refrigeration equipment operation status. Correlation analysis was performed on the preprocessed multi-source data to extract key feature parameters related to tomato quality and the refrigeration process. This included: using distance correlation analysis to determine the correlation between the multi-source data and the prediction target, where the correlation included linear and nonlinear correlation; screening key feature parameters with correlation greater than a preset threshold and integrating them in a time series format; wherein the prediction target included tomato quality-related indicators and refrigeration emission-related indicators. Based on the extracted key feature parameters and the improved xPatch time-series prediction model, tomato quality-related indicators and refrigeration emission-related indicators are predicted. An objective function related to the tomato quality indicators and the refrigeration emission indicators is constructed, and the objective function is solved by a differential evolution algorithm to obtain the optimal control strategy for the refrigeration environment parameters. The optimal control strategy is adjusted and sent to the cold storage control system for execution. The improved xPatch-based time-series prediction model includes a trend flow branch and a seasonal flow branch. The trend flow branch is used to extract long-term trend features from multi-source data, and the seasonal flow branch is used to extract short-term periodic features from multi-source data. The features are fused through a two-stage gating and channel attention mechanism. The tomato quality-related indicators include tomato water loss rate, and the refrigeration emission-related indicators include refrigeration emissions. Before employing the differential evolution algorithm to perform a rolling solution for the objective function, the method further includes: The decision variables for the objective function are determined, including the set values of key characteristic parameters that can be controlled and characterize the cold storage environment and the operating status of the refrigeration equipment. Determine the constraints of the optimization process, including cold storage environment constraints, refrigeration equipment safety constraints, and control variable variation range constraints between adjacent control cycles; The preprocessed multi-source data were subjected to correlation analysis to extract key characteristic parameters related to tomato quality and the refrigeration process.
2. The method for synergistic optimization of tomato cold storage quality and refrigeration emissions according to claim 1, characterized in that, The time-series prediction model based on extracted key feature parameters and improved xPatch predicts tomato quality-related indicators and refrigeration emission-related indicators, including: The preprocessed multi-source data is divided into training set, validation set and test set; The improved xPatch-based time series prediction model is trained based on the training set and the joint loss function, and the hyperparameters are optimized based on the validation set. Based on a trained time-series prediction model based on improved xPatch, tomato quality-related indicators and refrigeration emission-related indicators are predicted, and the prediction results are compared with the prediction results of at least one benchmark model.
3. The method for synergistic optimization of tomato cold storage quality and refrigeration emissions according to claim 1 or 2, characterized in that, The training of the time series prediction model based on the improved xPatch uses a joint loss function, which includes a smoothing loss term and a temporal difference loss term.
4. The method for synergistic optimization of tomato cold storage quality and refrigeration emissions according to claim 1, characterized in that, The preprocessing includes timestamp alignment, outlier handling, missing value handling, calculation of tomato quality-related indicators, calculation of refrigeration emission-related indicators, and normalization.
5. A device for synergistic optimization of tomato refrigeration quality and refrigeration emissions, characterized in that, include: The data acquisition and preprocessing module is used to collect and preprocess multi-source data that characterizes the cold storage environment, tomato quality, and the operating status of refrigeration equipment. The correlation analysis module is used to perform correlation analysis on the preprocessed multi-source data and extract key feature parameters related to tomato quality and the refrigeration process. This includes: using distance correlation analysis to determine the degree of correlation between the multi-source data and the prediction target, wherein the degree of correlation includes linear correlation and nonlinear correlation; screening key feature parameters whose degree of correlation is greater than a preset threshold and integrating them in a time series format; wherein the prediction target includes tomato quality-related indicators and refrigeration emission-related indicators. The model prediction module is used to predict tomato quality-related indicators and refrigeration emission-related indicators based on extracted key feature parameters and a time-series prediction model based on improved xPatch. The optimization solution module is used to construct objective functions related to the tomato quality indicators and the refrigeration emission indicators, and to use the differential evolution algorithm to solve the objective functions in a rolling manner to obtain the optimal control strategy for the refrigeration environment parameters. The strategy execution module is used to adjust and send the optimal control strategy to the cold storage control system for execution; The improved xPatch-based time-series prediction model includes a trend flow branch and a seasonal flow branch. The trend flow branch is used to extract long-term trend features from multi-source data, and the seasonal flow branch is used to extract short-term periodic features from multi-source data. The features are fused through a two-stage gating and channel attention mechanism. The tomato quality-related indicators include tomato water loss rate, and the refrigeration emission-related indicators include refrigeration emissions. The optimization solution module is also used for: The decision variables for the objective function are determined, including the set values of key characteristic parameters that can be controlled and characterize the cold storage environment and the operating status of the refrigeration equipment. Determine the constraints of the optimization process, including cold storage environment constraints, refrigeration equipment safety constraints, and control variable variation range constraints between adjacent control cycles; The preprocessed multi-source data were subjected to correlation analysis to extract key characteristic parameters related to tomato quality and the refrigeration process.
6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the method for synergistic optimization of tomato cold storage quality and refrigeration emissions as described in any one of claims 1 to 4.