An intelligent monitoring and management system for refrigerators based on the Internet of Things

Abstract

The application discloses a kind of based on Internet of Things's refrigerated cabinet intelligent monitoring management system, it is related to intelligent monitoring field, comprising: management terminal module, for as management terminal, real-time acquisition refrigerated cabinet each partition temperature regulating component's control instruction, including compressor start-stop signal, damper opening and refrigerant flow data;Type identification module is used to obtain the category information of each partition storage goods by RFID label scanning or image recognition technology, establishes the mapping database of goods and partition;Weight distribution module is used to trigger when multiple goods exist in the same partition, call preset priority rule library, identify the weight score of key goods;The present application can effectively inhibit quality decline from source before the quality of goods is about to deteriorate, and then reduce the loss caused by improper storage conditions, improve the overall preservation effect and economic benefit of goods.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology, specifically to an intelligent monitoring and management system for refrigerated cabinets based on the Internet of Things. Background Technology

[0002] As consumers place increasingly higher demands on the quality and safety of fresh food, vaccines, and high-end pharmaceuticals, the traditional cold chain industry is moving towards intelligent management. Furthermore, thanks to the maturity of IoT, big data, and AI technologies, companies are pursuing energy-efficient and intelligent cold chain solutions. During the deployment of commercial refrigerated display cases, spoilage due to product deterioration represents a significant hidden cost, with improper refrigeration being a major contributing factor. Reducing spoilage can directly improve profitability.

[0003] Traditional multi-zone refrigerators control each zone as an independent system. When the temperature of one zone is adjusted, the temperature of adjacent zones will be disturbed due to heat transfer, leading to increased system energy consumption, decreased temperature control accuracy, and even a chain reaction that affects the preservation of items in other zones. Even if the temperature is stable, the items will still gradually deteriorate. The system cannot accurately sense this to set the optimal temperature value and lacks in-depth data analysis and autonomous decision-making. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] In view of the above-mentioned shortcomings of the existing technology, the present invention provides an intelligent monitoring and management system for refrigerated cabinets based on the Internet of Things, which can effectively solve the problems of the existing technology.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] This invention discloses an intelligent monitoring and management system for refrigerated cabinets based on the Internet of Things, comprising:

[0009] The management terminal module is used as a management terminal to collect control commands for the temperature regulation components of each zone of the refrigerator in real time, including compressor start / stop signals, damper opening and refrigerant flow data;

[0010] The type identification module is used to obtain the category information of the items stored in each partition through RFID tag scanning or image recognition technology, and to establish a mapping database between items and partitions;

[0011] The weight allocation module is triggered when multiple types of items exist in the same partition. It calls the preset priority rule library to identify the weight score of key items. The priority rule library includes dynamic weight factors. When a preset special identifier is detected, the weight of the item is automatically raised to the highest level.

[0012] The status prediction module is used to identify the current temperature, humidity and spoilage parameters of key items, and predict the rate of quality change curve of key items in the future within a preset period through a preset quality prediction model.

[0013] The intervention decision module is used to compare each rate point in the predicted quality change rate curve with a preset threshold. When a certain rate point exceeds the safety threshold, an initial temperature adjustment command is generated.

[0014] The impact analysis module is used to obtain the temperature coefficients of the target area and its adjacent areas, and to calculate the temperature fluctuation coefficients of the adjacent areas after the implementation of the temperature adjustment command generated by the interference decision module.

[0015] The instruction generation module is used to combine the initial temperature adjustment instruction of the target area with the temperature fluctuation coefficient of the adjacent area, and use the PID algorithm to output the optimal temperature setpoint after compensation of the initial temperature adjustment instruction.

[0016] The adjustment and execution module is used to control the variable frequency compressor, electronic expansion valve and zone damper of the temperature regulation components to perform temperature adjustment according to the optimal temperature setpoint.

[0017] Furthermore, the working logic of the type recognition module is as follows:

[0018] When an item is placed in a specific area of ​​the refrigerator, a wide-angle camera pre-deployed at the top of that area is triggered to capture a top-down RGB image and a depth image of the item.

[0019] The acquired images are preprocessed, including denoising by Gaussian filtering and background segmentation based on depth image information, in order to extract clean object image regions.

[0020] The preprocessed images of the items are input into a recognition model built using a deep learning convolutional neural network algorithm. This model is based on the ResNet-50 architecture and is trained on a large image dataset containing several types of refrigerated items to extract multi-level visual features of the items and output the corresponding item category labels and confidence scores.

[0021] When the confidence level of the recognition model output is lower than the preset threshold, image re-acquisition is triggered; for recognition results with a confidence level higher than the preset threshold, the item category information is associated with its partition number and updated to the item-partition mapping database, while the image sample of the item is recorded.

[0022] Furthermore, the state prediction module uses a distributed wireless temperature and humidity sensor network pre-deployed in each section of the refrigerator to synchronously collect real-time temperature and humidity data of the microenvironment where the key items are located at a set sampling frequency. It interacts with the type identification module and the weight allocation module to obtain the category information of the identified key items and identify the pre-stored spoilage parameters of each type of item.

[0023] Furthermore, the spoilage parameters identified by the state prediction module include: water activity threshold, optimal spoilage temperature range, and color change rate constant.

[0024] Furthermore, the construction process of the quality prediction model in the state prediction module is as follows:

[0025] The historical data acquisition module collects quality change data of various refrigerated items under different temperature, humidity and time conditions. The quality change data includes the measured value of the spoilage temperature, color change and water activity change of the items.

[0026] The collected historical data is preprocessed, including data cleaning to remove outliers, data normalization to eliminate the influence of units, and extraction of key feature variables related to quality changes.

[0027] Using machine learning algorithms as a framework, preprocessed historical data as the training set, temperature, humidity, and spoilage parameters as input features, and quality change rate as the output target, the model is trained. During the model training process, the model parameters are adjusted through iterative optimization algorithms to minimize the prediction error, and cross-validation is used to evaluate the model's generalization ability. The trained model is then deployed to the state prediction module.

[0028] Furthermore, the initial temperature adjustment command generation process in the influence analysis module is as follows: based on the difference between the value of the rate point exceeding the safety threshold and the safety threshold, a predefined temperature and quality change rate mapping table is queried, and the required temperature adjustment amount is calculated. This mapping table is established through experimental data fitting and reflects the influence relationship between temperature change and quality change rate. The generated initial temperature adjustment command includes the target zone identifier, temperature adjustment direction, and adjustment amount, and is output to the influence analysis module.

[0029] Furthermore, the calculation process of the temperature fluctuation coefficient of adjacent regions in the influence analysis module includes the following steps:

[0030] Step 61: Obtain the thermal resistance coefficient and thermal capacity coefficient of the target area and adjacent areas through the pre-stored partition thermodynamic parameter database. The thermodynamic parameter database is established through the refrigerator design parameters and experimental calibration data, including the thermal conductivity of the material of each partition, the space volume and the heat exchange area with adjacent areas.

[0031] Step 62: Based on the obtained thermal resistance coefficient and thermal capacity coefficient, simulate the effect of temperature change in the target area on heat transfer in adjacent areas after the implementation of the temperature adjustment command. The simulation process uses Fourier's heat conduction law to calculate the heat flow driven by the temperature gradient and uses the energy conservation equation to identify the instantaneous temperature response curves of adjacent areas.

[0032] Step 63: Based on the instantaneous temperature response curves of adjacent regions obtained from the simulation, extract the temperature change data and calculate the temperature fluctuation coefficient. The temperature fluctuation coefficient is defined as the product of the maximum deviation of the predicted temperature from the initial set temperature and the average rate of change.

[0033] Furthermore, the formula for calculating the optimal temperature setpoint in the instruction generation module is as follows:

[0034] ;

[0035] In the formula, This represents the optimal temperature setting after compensation. This represents the initial temperature setting for the target partition, which is the system's preset partition reference temperature. This represents the initial temperature adjustment for the target zone, generated by the interference decision module. It is the required temperature adjustment calculated when the rate of quality change exceeds a threshold. This represents the temperature fluctuation coefficient between adjacent regions. The proportional control coefficient, obtained through PID algorithm tuning, reflects the response strength of the current error. This represents the integral control coefficient, which serves as a gain parameter for eliminating historical accumulated errors. Representing the differential control coefficient, it serves as a gain parameter to suppress future error variations. Represents real-time temperature difference error. Representing a time variable, serving as a real-time timestamp for system operation. Represents the current sampling point number. Representing the Real-time temperature difference error of each sample. This represents a fixed sampling period, measured in seconds.

[0036] Furthermore, the adjustment execution module is interconnected with a feedback calibration module via a wireless network. The feedback calibration module is used to collect actual temperature data through a preset temperature sensor network, receive temperature adjustment feedback, correct the quality prediction model parameters, and push early warning information and control records to the management terminal module, supporting remote parameter configuration via APP.

[0037] Furthermore, the management terminal module is interconnected with the type identification module, weight allocation module, and state prediction module via a wireless network; the interference decision module is interconnected with the state prediction module and impact analysis module via a wireless network; and the instruction generation module is interconnected with the impact analysis module and adjustment execution module via a wireless network.

[0038] (III) Beneficial Effects

[0039] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects:

[0040] 1. By continuously monitoring the real-time freshness of items, identifying and predicting the rate of quality change of key items, and dynamically providing the most suitable storage temperature, the potential risk of spoilage can be predicted in advance before obvious deterioration occurs, and the temperature setting can be adjusted in time. This can effectively suppress quality decline from the source, thereby reducing losses caused by improper storage conditions and improving the overall preservation effect and economic benefits of items.

[0041] 2. By introducing the temperature fluctuation coefficient of adjacent areas, the interference to neighboring areas is quantitatively considered when calculating the optimal setpoint, thereby making a globally optimal decision, ensuring the overall stability of the system, and intervening in temperature control only when there is a risk to quality, avoiding unnecessary low-temperature operation, and achieving energy saving and consumption reduction. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0043] Figure 1 This is a schematic diagram of the framework of the present invention;

[0044] Figure 2 This is a flowchart illustrating the calculation process of the temperature fluctuation coefficient between adjacent regions in this invention.

[0045] The labels in the diagram represent: 1. Management terminal module; 2. Type identification module; 3. Weight allocation module; 4. State prediction module; 5. Intervention decision module; 6. Impact analysis module; 7. Instruction generation module; 8. Adjustment execution module; 9. Feedback calibration module. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0047] The present invention will be further described below with reference to embodiments.

[0048] This embodiment presents an IoT-based intelligent monitoring and management system for refrigerated display cases, such as... Figure 1 As shown, it includes:

[0049] The management terminal module 1 is used as a management terminal to collect control commands for the temperature adjustment components of each zone of the refrigerator in real time, including compressor start / stop signals, damper opening and refrigerant flow data.

[0050] Type identification module 2 is used to obtain category information of items stored in each partition through RFID tag scanning or image recognition technology, and establish a mapping database between items and partitions; the working logic of type identification module 2 is as follows:

[0051] When an item is placed in a specific area of ​​the refrigerator, a wide-angle camera pre-deployed at the top of that area is triggered to capture a top-down RGB image and a depth image of the item.

[0052] The acquired images are preprocessed, including denoising by Gaussian filtering and background segmentation based on depth image information, in order to extract clean object image regions.

[0053] The preprocessed images of the items are input into a recognition model built using a deep learning convolutional neural network algorithm. This model is based on the ResNet-50 architecture and is trained on a large image dataset containing several types of refrigerated items to extract multi-level visual features of the items and output the corresponding item category labels and confidence scores.

[0054] When the confidence level of the recognition model output is lower than the preset threshold, the image is re-acquired or the item is marked as "pending confirmation". The system administrator is notified through the management terminal module 1 to perform manual annotation. For recognition results with a confidence level higher than the preset threshold, the item category information is associated with its partition number and updated to the item-partition mapping database. At the same time, the image sample of the item is recorded for incremental learning and optimization of the subsequent model, so that the system can continuously optimize and adapt to new item types and ensure the intelligent evolution of the system.

[0055] The weight allocation module 3 is triggered when multiple types of items exist in the same partition. It calls the preset priority rule library to identify the weight score of key items. The priority rule library includes dynamic weight factors. When a preset special identifier is detected, the weight of the item is automatically increased to the highest level.

[0056] The status prediction module 4 is used to identify the current temperature, humidity, and spoilage parameters of key items, and predict the quality change rate curve of key items within a preset period through a preset quality prediction model. The status prediction module 4 synchronously collects real-time temperature and humidity data of the microenvironment in which the key items are located at a set sampling frequency through a distributed wireless temperature and humidity sensor network pre-deployed in each section of the refrigerated cabinet. It interacts with the type identification module 2 and the weight allocation module 3 to obtain the category information of the identified key items and identify the pre-stored spoilage parameters of each type of item. The spoilage parameters include: water activity threshold, optimal spoilage temperature range, and color change rate constant.

[0057] Interference decision module 5 is used to compare each rate point in the predicted quality change rate curve with a preset threshold. When a certain rate point exceeds the safety threshold, an initial temperature adjustment command is generated.

[0058] The impact analysis module 6 is used to obtain the temperature coefficients of the target area and its adjacent areas, and to calculate the temperature fluctuation coefficients of the adjacent areas after the implementation of the temperature adjustment command generated by the interference decision module 5. The initial temperature adjustment command is generated as follows: based on the difference between the value of the rate point exceeding the safety threshold and the safety threshold, a predefined temperature and quality change rate mapping table is queried, and the required temperature adjustment amount is calculated. This mapping table is established by fitting experimental data and reflects the influence relationship between temperature change and quality change rate. The generated initial temperature adjustment command includes the target zone identifier, temperature adjustment direction and adjustment amount, and is output to the impact analysis module 6 for subsequent processing.

[0059] Instruction generation module 7 is used to combine the initial temperature adjustment instruction of the target area with the temperature fluctuation coefficient of the adjacent area, and use a PID algorithm to output the optimal temperature setpoint after compensation of the initial temperature adjustment instruction.

[0060] The formula for calculating the optimal temperature setpoint is:

[0061] ;

[0062] In the formula, This represents the optimal temperature setting after compensation. This represents the initial temperature setting for the target partition, which is the system's preset partition reference temperature. The initial temperature adjustment amount for the target zone is generated by the interference decision module 5. It is the required temperature adjustment amount calculated when the rate of quality change exceeds a threshold. This represents the temperature fluctuation coefficient between adjacent regions. The proportional control coefficient, obtained through PID algorithm tuning, reflects the response strength of the current error. This represents the integral control coefficient, which serves as a gain parameter for eliminating historical accumulated errors. Representing the differential control coefficient, it serves as a gain parameter to suppress future error variations. Represents real-time temperature difference error. Representing a time variable, serving as a real-time timestamp for system operation. Represents the current sampling point number. Representing the Real-time temperature difference error of each sample. This represents a fixed sampling period, measured in seconds.

[0063] Adjustment module 8 is used to control the variable frequency compressor, electronic expansion valve and zone damper of the temperature regulation components to perform temperature adjustment according to the optimal temperature set value.

[0064] The adjustment execution module 8 is connected to the feedback calibration module 9 via a wireless network. The feedback calibration module 9 is used to collect actual temperature data through a preset temperature sensor network, receive temperature adjustment feedback, correct the quality prediction model parameters, and push early warning information and control records to the management terminal module 1, supporting remote parameter configuration via APP.

[0065] The management terminal module 1 is interconnected with the type identification module 2, weight allocation module 3 and state prediction module 4 via a wireless network. The intervention decision module 5 is interconnected with the state prediction module 4 and the impact analysis module 6 via a wireless network. The instruction generation module 7 is interconnected with the impact analysis module 6 and the adjustment execution module 8 via a wireless network.

[0066] Compared with existing technologies, this embodiment uses an intelligent identification and quality prediction model for items to make predictive adjustments based on the real-time quality change trend of items, and combines the thermal coupling effect between zones for collaborative optimization. This effectively improves the operating energy efficiency and intelligence level of the entire refrigeration system while ensuring the safety of items and significantly reducing spoilage and loss.

[0067] At other levels, this embodiment also provides a calculation process for the temperature fluctuation coefficient of adjacent regions, such as... Figure 2 As shown, it includes the following steps:

[0068] Step 61: Obtain the thermal resistance coefficient and thermal capacity coefficient of the target area and adjacent areas through the pre-stored partition thermodynamic parameter database. The thermodynamic parameter database is established through the refrigerator design parameters and experimental calibration data, including the thermal conductivity of the material of each partition, the space volume and the heat exchange area with adjacent areas.

[0069] Step 62: Based on the obtained thermal resistance coefficient and thermal capacity coefficient, simulate the effect of temperature change in the target area on heat transfer in adjacent areas after the implementation of the temperature adjustment command. The simulation process uses Fourier's heat conduction law to calculate the heat flow driven by the temperature gradient and uses the energy conservation equation to identify the instantaneous temperature response curves of adjacent areas.

[0070] Step 63: Based on the instantaneous temperature response curves of adjacent areas obtained from the simulation, extract the temperature change data and calculate the temperature fluctuation coefficient. The temperature fluctuation coefficient is defined as the product of the maximum deviation of the predicted temperature from the initial set temperature and the average rate of change, which is used to quantify the degree of temperature instability in adjacent areas after the implementation of the temperature adjustment command.

[0071] Compared with existing technologies, by converting the thermal interference of adjacent areas into a quantifiable temperature fluctuation coefficient, the improvement from isolated zone temperature control to coordinated global temperature control is achieved, effectively avoiding system oscillation and reducing unnecessary energy consumption.

[0072] This embodiment provides a process for constructing a quality prediction model as follows:

[0073] The historical data acquisition module collects quality change data of various refrigerated items under different temperature, humidity and time conditions. The quality change data includes the measured value of the spoilage temperature, color change and water activity change of the items.

[0074] The collected historical data is preprocessed, including data cleaning to remove outliers, data normalization to eliminate the influence of units, and extraction of key feature variables related to quality changes.

[0075] The model was trained using machine learning algorithms as a framework, preprocessed historical data as a training set, temperature, humidity and spoilage parameters as input features, and the rate of quality change as the output target.

[0076] During model training, the model parameters are adjusted through iterative optimization algorithms to minimize the prediction error, and the model's generalization ability is evaluated using cross-validation.

[0077] The trained model is deployed to the state prediction module 4 to predict the rate of change of the quality of key items in real time. The model is also incrementally learned and its parameters are updated regularly based on actual monitoring data to maintain prediction accuracy.

[0078] In summary, this invention provides real-time sensing of the quality change trends of the stored items, enabling temperature regulation to move beyond simply maintaining a set value. Instead, it directly aims to delay food spoilage and maintain optimal quality. By introducing analysis of the temperature fluctuation coefficient of neighboring zones, the temperature adjustment of the target zone can predict and compensate for the thermal impact on adjacent areas, thereby avoiding system oscillations that may be caused by local optimization. Combined with dynamic fine-tuning of the PID algorithm, the invention maximizes the shelf life of all food in the cabinet while minimizing global energy consumption, achieving a balance between quality assurance and operational efficiency.

[0079] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention 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 will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An Internet of Things-based intelligent monitoring and management system for a refrigerated cabinet, characterized in that, The application relates to a refrigerated cabinet temperature adjustment method and device. The management terminal module is used for collecting control instructions of temperature adjustment components of each partition of the refrigerated cabinet in real time as a management terminal. The type identification module is used for obtaining category information of stored goods in each partition and establishing a mapping database of the goods and the partitions. The weight allocation module is used for triggering when multiple types of goods exist in the same partition, calling a preset priority rule library, and identifying a weight score of a key good. The state prediction module is used for identifying a current temperature, humidity and spoilage parameter of the key good, and predicting a quality change rate curve of the key good in a future preset period through a preset quality prediction model. The interference decision module is used for comparing each rate point in the predicted quality change rate curve with a preset threshold value, and generating an initial temperature adjustment instruction when a certain rate point exceeds a safety threshold value. The influence analysis module is used for obtaining a temperature coefficient of a neighboring area of a target area, and calculating a temperature fluctuation coefficient of the neighboring area after implementation of the temperature adjustment instruction generated by the interference decision module. The instruction generation module is used for combining the initial temperature adjustment instruction of the target area and the temperature fluctuation coefficient of the neighboring area, and outputting an optimal temperature setting value after compensation of the initial temperature adjustment instruction through a PID algorithm. The adjustment execution module is used for controlling the temperature adjustment components to perform temperature adjustment according to the optimal temperature setting value.

2. The intelligent monitoring and management system for refrigerators based on the Internet of Things according to claim 1, characterized in that, The working logic of the type identification module is as follows: When goods are put into a specific area of the refrigerated cabinet, a wide-angle camera pre-deployed on the top of the area is triggered to collect a top-view RGB image and a depth image of the goods. The collected images are preprocessed, and background segmentation is performed based on the depth image information to extract a pure good image area. The preprocessed good image is input into an identification model constructed through a deep learning convolutional neural network algorithm to extract multi-level visual features of the goods and output corresponding good category labels and confidence. When the confidence output by the identification model is lower than a preset threshold value, image re-collection is triggered. 3.The intelligent monitoring and management system of the refrigerated cabinet based on the Internet of Things according to claim 1, characterized in that, For an identification result with a confidence higher than the preset threshold value, the good category information is associated with a partition number where the good is located, and is updated to the mapping database of the goods and the partitions.

4. The intelligent monitoring and management system for refrigerators based on the Internet of Things according to claim 1, characterized in that, The state prediction module synchronously collects real-time temperature and humidity data of a microenvironment where a key good is located through a distributed wireless temperature and humidity sensor network pre-deployed in each partition of the refrigerated cabinet at a set sampling frequency, interacts with the type identification module and the weight allocation module to obtain category information of the identified key good, and identifies pre-stored spoilage parameters of each type of good.

5. The intelligent monitoring and management system for refrigerators based on the Internet of Things according to claim 1, characterized in that, The spoilage parameters identified by the state prediction module include a water activity threshold value, an optimal spoilage temperature interval and a color change rate constant. The construction process of the quality prediction model in the state prediction module is as follows: A historical data collection module collects quality change data of multiple refrigerated goods under different temperature, humidity and time conditions; The collected historical data are preprocessed; A machine learning algorithm is used as a framework, the preprocessed historical data are used as a training set, temperature, humidity and spoilage parameters are used as input features, and a quality change rate is used as an output target for model training. Deploy the trained model to a state prediction module.

6. The intelligent monitoring and management system for refrigerators based on the Internet of Things according to claim 1, characterized in that, The initial temperature adjustment instruction generation process in the influence analysis module is as follows: according to the difference between the value of the rate point exceeding the safety threshold and the safety threshold, the pre-defined temperature and quality change rate mapping table is queried, the required temperature adjustment amount is calculated, the generated initial temperature adjustment instruction includes the target partition identification, the temperature adjustment direction and the adjustment amount, and is output to the influence analysis module. 7.The intelligent monitoring and management system of the refrigerated cabinet based on the Internet of Things according to claim 1, characterized in that, The calculation process of the temperature fluctuation coefficient of the adjacent area in the influence analysis module includes the following steps: Step 61: Obtain the thermal resistance coefficient and the heat capacity coefficient of the target area and the adjacent area through the pre-stored partition thermodynamic parameter database; Step 62: Based on the obtained thermal resistance coefficient and heat capacity coefficient, simulate the influence of the temperature change of the target area after the implementation of the temperature adjustment instruction on the heat transfer of the adjacent area, and identify the instantaneous temperature response curve of the adjacent area; Step 63: According to the simulated instantaneous temperature response curve of the adjacent area, extract the temperature change data and calculate the temperature fluctuation coefficient. 8.The intelligent monitoring and management system of a cold storage cabinet based on an Internet of Things according to claim 1, characterized in that, The calculation formula of the optimal temperature setting value in the instruction generation module is: ； In the formula, represents the optimal temperature setting value after compensation, represents the initial temperature setting value of the target partition, represents the initial temperature adjustment amount of the target partition, represents the temperature fluctuation coefficient of the adjacent area, represents the proportional control coefficient, represents the integral control coefficient, represents the differential control coefficient, represents the real-time temperature difference error, represents the time variable, represents the current sampling point sequence number, represents the real-time temperature difference error of the sampling, represents the fixed sampling period. 9.The intelligent monitoring and management system of a cold storage cabinet based on the Internet of Things according to claim 1, characterized in that, The adjustment execution module is connected with a feedback calibration module through a wireless network, and the feedback calibration module is used to collect actual temperature data through a pre-set temperature sensor network, receive feedback of temperature adjustment, and push early warning information and regulation records to the management terminal module. 10.The intelligent monitoring and management system of a cold storage cabinet based on an Internet of Things according to claim 1, characterized in that, The management terminal module, the type identification module, the weight allocation module and the state prediction module are connected through a wireless network, the interference decision module, the state prediction module and the influence analysis module are connected through a wireless network, and the instruction generation module, the influence analysis module and the adjustment execution module are connected through a wireless network.

Images