Food quick-freezing temperature and humidity cooperative control method and system based on edge calculation

Abstract

The invention discloses a food quick-freezing temperature and humidity cooperative control method and system based on edge calculation, and relates to the technical field related to data processing, and the method comprises the steps: configuring an edge box in a quick-freezing cabin, integrating a surface temperature array, a humidity probe and an anemometer, and executing the data collection in the quick-freezing cabin; obtaining a quick-frozen food label, reading equipment power of a quick-freezing cabin, and configuring constraint conditions; configuring a cooperative control channel including the cooperative control channel of the first target and the second target; and configuring a cooperative control channel by using a quick-frozen food label, taking an initial data set as an initial condition, executing three-stage temperature and humidity control optimization under the constraint of a constraint condition, establishing a cooperative control parameter, and then performing quick-freezing cabin control management. The technical problem that in the prior art, the quick-freezing efficiency and the product quality are poor due to the fact that data transmission cloud decision response is high in delay and temperature and humidity cannot be optimized in a collaborative mode is solved, and the technical effects of improving the quick-frozen food quality and the quick-freezing efficiency and reducing energy consumption are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a food quick-freezing temperature and humidity collaborative control method and system based on edge computing. BACKGROUND

[0002] In the food quick-freezing process, multiple environmental parameters such as temperature, humidity, and wind speed directly affect the quick-freezing effect, for example, the size, distribution uniformity of ice crystals, and the integrity of food tissue. If not properly controlled, it is easy to cause cell structure damage, juice loss, and texture deterioration, which seriously affects product quality. The existing quick-freezing system adopts a centralized control architecture and relies on an upper computer or a cloud platform for data analysis and decision-making. However, it has the disadvantages of high response delay, insufficient real-time performance, limited data transmission bandwidth, and the like, and is difficult to meet the high-precision control requirements of temperature and humidity dynamic changes in the quick-freezing process, especially in the case of coexistence of multiple food types and large fluctuations in equipment power. It is unable to achieve multi-objective collaborative optimization, resulting in increased energy consumption, prolonged freezing time, or inconsistent product quality.

[0003] Therefore, in the related art at the present stage, there are technical problems of high response delay of data transmission and cloud decision-making, inability to collaboratively optimize temperature and humidity, and poor quick-freezing efficiency and product quality. SUMMARY

[0004] The present application provides a food quick-freezing temperature and humidity collaborative control method and system based on edge computing, which solves the technical problems of high response delay of data transmission and cloud decision-making, inability to collaboratively optimize temperature and humidity, and poor quick-freezing efficiency and product quality in the prior art, and achieves the technical effects of improving the quality of quick-frozen food, quick-freezing efficiency, and reducing energy consumption.

[0005] The present application provides a food quick-freezing temperature and humidity collaborative control method based on edge computing, which comprises: configuring an edge box in a quick-freezing cabin, the edge box being integrated with a surface temperature array, a humidity probe, and an anemometer; using the edge box to perform data collection in the quick-freezing cabin and establish an initial data set; obtaining a quick-frozen food label placed in the quick-freezing cabin and reading the equipment power of the quick-freezing cabin; configuring constraint conditions according to the quick-frozen food label and the equipment power; configuring a collaborative control channel, the collaborative control channel being a collaborative control channel comprising a first target and a second target, wherein the first target is a primary target comprising a maximum tissue integrity target and a minimum ice crystal median diameter target, and the second target is a secondary target comprising a shortest freezing time target and a minimum unit energy consumption target; after configuring the collaborative control channel using the quick-frozen food label, using the initial data set as an initial condition to perform three-stage temperature and humidity control optimization under the constraint of the constraint conditions, establishing collaborative control parameters, and performing quick-freezing cabin control management according to the collaborative control parameters.

[0006] In a possible implementation, the food quick-freezing temperature and humidity collaborative control method based on edge computing further performs the following processing: after field extraction is performed on the quick-freezing food label, a control constraint matrix is established by using a field extraction result to establish a food material type constraint, a unit volume constraint, and a target quality priority constraint; a weight factor of a control target in a collaborative control channel is nonlinearly compensated by using the control constraint matrix, and collaborative control channel configuration is completed; after three-stage stage trigger thresholds are set by using the quick-freezing food label and an initial data set, three-stage temperature and humidity control optimization is sequentially performed, where the three stages are a primary coagulation stage, a penetration sensitivity stage, and a freezing stage, and a fitting result of a current stage is taken as an initial input of a next stage after optimization control of each stage is completed; three-stage temperature and humidity control optimization is performed by using the first target and the second target configured to be completed, and collaborative control parameters are established.

[0007] In a possible implementation, the food quick-freezing temperature and humidity collaborative control method based on edge computing further performs the following processing: the first target and the second target are compensated and corrected according to a surface temperature and humidity control constraint of the primary coagulation stage, a first-stage local target function is established, the first-stage local target function includes a surface temperature control term, a humidity control term in the cabin, and an energy consumption control term; the initial data set is taken as an initial condition, a second-stage trigger threshold is taken as a control target under constraint conditions, control optimization is performed by using the first-stage local target function, a first-stage control optimization result is established, and a first-stage control fitting result is generated; and collaborative control parameters are established according to the first-stage control optimization result and the first-stage control fitting result.

[0008] In a possible implementation, the food quick-freezing temperature and humidity collaborative control method based on edge computing further performs the following processing: the first target and the second target are compensated and corrected according to an ice crystal growth constraint and a core temperature constraint of the penetration sensitivity stage, a second-stage local target function is established, the second-stage local target function includes a minimum median diameter evaluation term, a tissue integrity evaluation term, a core temperature control term, and an energy consumption control term; the first-stage control fitting result is taken as an initial condition, a third-stage trigger threshold is taken as a control target under constraint conditions, control optimization is performed by using the second-stage local target function, a second-stage control optimization result is established, and a second-stage control fitting result is generated; and collaborative control parameters are established according to the first-stage control optimization result, the first-stage control fitting result, the second-stage control optimization result, and the second-stage control fitting result.

[0009] In a possible implementation, the edge computing-based food quick-freezing temperature and humidity coordinated control method further performs the following processing: Compensating and correcting the first and second objectives based on the ice crystal growth constraints and core temperature constraints of the penetration-sensitive stage, establishing a second-stage local objective function. This second-stage local objective function includes a median diameter minimization evaluation term, a tissue integrity evaluation term, a core temperature control term, and an energy consumption control term. Using the first-stage control fitting result as initial conditions, and under constraint conditions, using the third-stage trigger threshold as the control objective, performing control optimization using the second-stage local objective function, establishing the second-stage control optimization result, and generating the second-stage control fitting result. Finally, establishing coordinated control parameters based on the first-stage control optimization result, the first-stage control fitting result, the second-stage control optimization result, and the second-stage control fitting result.

[0010] In a possible implementation, the edge computing-based food quick-freezing temperature and humidity coordinated control method further performs the following processing: raw temperature, humidity, and wind speed data are acquired by the edge computing node built into the edge box, filtering and outlier processing are performed, and a data time series is established after synchronization with timestamps; a three-dimensional temperature distribution grid is constructed using a temperature data array, and humidity and wind speed data are mapped to the food location to establish a local environmental state vector; the edge computing node performs rapid feature calculation based on the three-dimensional temperature distribution grid and the local environmental state vector, the rapid feature calculation including temperature gradient calculation, humidity gradient calculation, and wind speed distribution analysis, and outputs an initial dataset.

[0011] In a possible implementation, the edge computing-based food quick-freezing temperature and humidity coordinated control method further performs the following processes: executing quick-freezing chamber control with the coordinated control parameters, and setting response nodes based on the coordinated control parameters; activating edge boxes at the response nodes to perform response verification and establish response compensation data; and adaptively adjusting the parameters of the coordinated control parameters according to the response compensation data.

[0012] In a possible implementation, the edge computing-based food quick-freezing temperature and humidity coordinated control method further performs the following processing: configuring a compensation target based on the response compensation data; reading the node interval of the response node, identifying control anomalies in the quick-freezing chamber based on the node interval and the response compensation data, and establishing control anomaly compensation; and adaptively adjusting the parameters of the coordinated control parameters based on the compensation target and the control anomaly compensation.

[0013] In a possible implementation, the edge computing-based food quick-freezing temperature and humidity coordinated control method further performs the following processing: performing compensation early warning identification on the response compensation data; if the compensation early warning identification result is an early warning trigger result, then generating an early warning reset instruction, and reconstructing the coordinated control parameters according to the early warning reset instruction.

[0014] This application also provides a food quick-freezing temperature and humidity coordinated control system based on edge computing. The system includes: a data acquisition module for configuring an edge box in the quick-freezing chamber, the edge box integrating a surface temperature array, humidity probe, and anemometer, and using the edge box to perform data acquisition within the quick-freezing chamber to establish an initial dataset; a constraint configuration module for acquiring the labels of quick-frozen foods placed in the quick-freezing chamber and reading the equipment power of the quick-freezing chamber, and configuring constraint conditions according to the quick-frozen food labels and equipment power; a control channel configuration module for configuring a coordinated control channel, the coordinated control channel including a first objective and a second objective, wherein the first objective is the primary objective, including maximizing tissue integrity and minimizing the median ice crystal diameter, and the second objective is the secondary objective, including minimizing the freezing time and minimizing unit energy consumption; and a temperature and humidity control optimization module for configuring the coordinated control channel using the quick-frozen food labels, using the initial dataset as initial conditions, performing a three-stage temperature and humidity control optimization under the constraints, establishing coordinated control parameters, and performing quick-freezing chamber control management according to the coordinated control parameters.

[0015] This application proposes a food quick-freezing temperature and humidity collaborative control method and system based on edge computing. An edge box is configured in the quick-freezing chamber, integrating a surface temperature array, humidity probe, and an anemometer to perform data acquisition within the chamber. The system acquires the labels of quick-frozen foods and reads the equipment power of the quick-freezing chamber, configuring constraints. Collaborative control channels are configured, including channels for the first and second objectives. Using the quick-frozen food labels to configure these channels, the initial dataset is used as initial conditions. Under the constraints, a three-stage temperature and humidity control optimization is performed to establish collaborative control parameters, and then the quick-freezing chamber is controlled and managed. This solves the technical problems of high data transmission latency and inability to collaboratively optimize temperature and humidity in existing technologies, leading to poor quick-freezing efficiency and product quality. It achieves the technical effects of improving the quality and efficiency of quick-freezing foods while reducing energy consumption. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 A schematic diagram of the process for a food quick-freezing temperature and humidity coordinated control method based on edge computing provided in this application embodiment.

[0018] Figure 2 A schematic diagram of the edge computing-based temperature and humidity coordinated control system for food quick-freezing provided in this application embodiment.

[0019] Figure labeling: Data acquisition module 10, constraint configuration module 20, control channel configuration module 30, temperature and humidity control optimization module 40. Detailed Implementation

[0020] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0023] This application provides an edge computing-based method for coordinated temperature and humidity control in food quick-freezing, such as... Figure 1 As shown, the method includes: Step S100: An edge box is configured in the quick-freezing chamber. The edge box integrates a surface temperature array, a humidity probe, and an anemometer. Data acquisition in the quick-freezing chamber is performed using the edge box to establish an initial dataset.

[0024] Preferably, an embedded industrial control device with computing capabilities, namely an edge box, is configured on-site in the blast freezer. The edge box integrates a surface temperature array, a humidity probe, and an anemometer. The surface temperature array is a sensor group composed of multiple temperature sensors arranged in a row, used to simultaneously measure the temperature distribution at different locations within the blast freezer, especially on the surface of frozen food, thereby acquiring spatial temperature field information. The humidity probe measures the humidity in the air within the blast freezer, and the anemometer measures the speed of the refrigeration airflow within the blast freezer. Data acquisition within the blast freezer is performed using the edge box; that is, the edge box drives its integrated sensors to collect spatial temperature data, humidity data, and wind speed data at a set frequency. The data is then preprocessed to obtain a structured dataset as the initial dataset.

[0025] Furthermore, step S100 also includes step S110, where the edge computing node built into the edge box acquires the collected raw data of temperature, humidity, and wind speed, performs filtering and outlier processing, and establishes a data time series after synchronization with timestamps; step S120, a three-dimensional temperature distribution grid is constructed using the temperature data array, and the humidity and wind speed data are mapped to the food location to establish a local environmental state vector; step S130, the edge computing node performs fast feature calculation based on the three-dimensional temperature distribution grid and the local environmental state vector, the fast feature calculation including temperature gradient calculation, humidity gradient calculation, and wind speed distribution analysis, and outputs an initial dataset.

[0026] Preferably, the edge computing node built into the edge box reads the raw temperature, humidity, and wind speed data from the electronic interfaces of the surface temperature array, humidity probe, and anemometer connected to it. Then, it applies moving average filtering and Kalman filtering to process the data, suppressing or eliminating high-frequency, random noise fluctuations. Next, it performs outlier processing, including using the standard deviation method and interquartile range method to identify and remove obviously erroneous and unreasonable measurement values. Then, it synchronizes the raw temperature, humidity, and wind speed data through timestamps, aligning all data to a unified time axis to ensure that each data sample of temperature, humidity, and wind speed corresponds to the same time point, ultimately forming a data time series, that is, each time point corresponds to a complete set of temperature, humidity, and wind speed data.

[0027] Preferably, the surface temperature array provides temperature data from multiple different spatial points within the quick-freezing chamber. Using this discrete temperature data, spatial interpolation algorithms such as Kriging interpolation and inverse distance weighted interpolation are employed to estimate the temperature value of any virtual point within the entire quick-freezing chamber. This digitizes the entire quick-freezing chamber into a model composed of countless tiny three-dimensional grids, with each grid assigned a temperature value, thus forming a three-dimensional temperature distribution grid. Humidity and wind speed data are then mapped to the food's location. Humidity probes and anemometers provide measurements representing the overall conditions of a specific area within the chamber. Based on the installation locations of the humidity probes and anemometers, spatial correlation algorithms are used to associate the measured humidity and wind speed values ​​with the area where the frozen food is located. For example, if video A is primarily within the measurement range of anemometer S1, then the wind speed in the area where food A is located is the reading of anemometer S1. Furthermore, a local environmental state vector is created for each quick-frozen food item, comprehensively describing the complete state of the environment in which the food is located, including the food surface temperature, the humidity of the air surrounding the food, and the wind speed on the food surface.

[0028] Preferably, at the edge computing nodes, rapid feature calculations are performed based on a three-dimensional temperature distribution grid and local environmental state vectors to extract key indicators that more directly reflect the process state and quality impact. These include temperature gradient calculation, humidity gradient calculation, and wind speed distribution analysis. Specifically, temperature gradient calculation refers to calculating the temperature change per unit distance, revealing whether the cooling is uniform in different areas of the quick-freezing chamber or the temperature difference between the food surface and core. Humidity gradient calculation refers to calculating the humidity change per unit distance, reflecting the intensity of water vapor migration and condensation / sublimation. Wind speed distribution analysis refers to analyzing the statistical characteristics of wind speed in the entire quick-freezing chamber or around the food, such as maximum, minimum, average, and uniformity index, to assess the efficiency of cold air flow. The final output is an initial dataset, which includes the preprocessed raw data time series, a three-dimensional temperature distribution grid model, the local environmental state vector of each food, and the calculated temperature gradient, humidity gradient, and wind speed distribution.

[0029] Step S200: Obtain the label of the frozen food placed in the quick-freezing chamber, read the equipment power of the quick-freezing chamber, and configure constraint conditions according to the frozen food label and equipment power.

[0030] Preferably, the labels of frozen foods in the quick-freezing compartment are obtained through identification using RFID, QR codes, barcodes, etc., and may include food category, physical specifications, target quality grade, and initial state. Among them, food category may include seafood, meat, fruits and vegetables, pastries, and prepared dishes, etc., and different categories of food have different thermal properties; physical specifications may include the size, thickness, weight, and bulk density of the food, which affect the difficulty and time of heat penetration; target quality grade may include premium or standard grade, and initial state may include initial temperature and surface moisture content; the current operating power of the quick-freezing compartment refrigeration equipment is obtained through the equipment's control system bus, including the power of the refrigeration compressor, fan, and defrost heater. Based on the constraints of the frozen food label and equipment power configuration, namely boundary conditions and limits, these mainly include: the maximum allowable cooling rate limiting the upper limit of the temperature drop speed; the temperature safety window ensuring that the temperature quickly passes through the danger temperature zone within a specified time; the maximum ice crystal formation zone passing time limiting the time it spends in the temperature range most prone to producing large ice crystals; the surface and core temperature gradient limiting excessive temperature difference; the freezing time limit used to limit the maximum time of the entire quick-freezing process; the maximum cooling capacity constraint ensuring that the cooling load cannot exceed the limit that the current equipment compressor power can provide; the wind speed and air volume constraint requiring that the target wind speed must be within the wind speed range that the fan power can generate; and the dehumidification / humidification capacity constraint ensuring that the humidity control setpoint is within the power regulation capability of the defrost heater, steam humidifier, and other actuators.

[0031] Step S300: Configure a collaborative control channel. The collaborative control channel includes a first objective and a second objective. The first objective is the primary objective, including maximizing tissue integrity and minimizing the median diameter of ice crystals. The second objective is the secondary objective, including the shortest freezing time and the minimum unit energy consumption.

[0032] Preferably, the optimization objectives are determined based on the frozen food label, and then a collaborative control channel is configured. This collaborative control channel is the optimization channel for the decision-making logic, including a collaborative control channel for the first objective and a collaborative control channel for the second objective. The first objective is the primary objective, including maximizing tissue integrity and minimizing the median ice crystal diameter. The second objective is the secondary objective, including minimizing the freezing time and minimizing unit energy consumption. For example, for high-end sashimi, the first objective, quality, has a very high weight; for ordinary vegetables, the second objective, efficiency and energy consumption, has a higher weight. Specifically, maximizing tissue integrity means minimizing the damage to the food's cell structure during the freezing process, i.e., minimizing the cell rupture rate caused by ice crystal expansion and mechanical stress. This is achieved by establishing a control channel that correlates with the temperature history and cooling rate. The model indirectly evaluates the ice crystals; minimizing the median ice crystal diameter aims to minimize the average size of ice crystals formed inside the food. Smaller and more uniformly distributed ice crystals cause less puncture and damage to cell structures, resulting in better quality after thawing. An ice crystal growth model is established to correlate temperature control parameters with predicted ice crystal size. The ice crystal growth model is strongly correlated with freezing rate and time to pass through the maximum ice crystal formation zone; the shortest freezing time aims to minimize the total time it takes for the food to drop from the initial temperature to the desired core temperature while meeting the first objective. However, pursuing the fastest cooling may lead to cold shock or excessive temperature gradients, which could damage tissue integrity and increase ice crystal size; the minimum unit energy consumption aims to minimize the electrical energy consumed in processing a unit weight of food.

[0033] Step S400: After configuring the collaborative control channel using the frozen food label, the initial dataset is used as the initial condition. Under the constraints, a three-stage temperature and humidity control optimization is performed to establish collaborative control parameters. The frozen food compartment is then controlled and managed according to the collaborative control parameters.

[0034] Step S400 further includes step S410, after extracting fields from the frozen food label, establishing food type constraints, unit volume constraints, and target quality priority constraints using the field extraction results to form a control constraint matrix; step S420, using the control constraint matrix to perform nonlinear compensation on the weight factors of the control targets within the collaborative control channel to complete the collaborative control channel configuration; step S430, after setting the stage trigger thresholds for the three stages using the frozen food label and the initial dataset, executing the three-stage temperature and humidity control optimization in sequence, wherein the three stages are the initial condensation stage, the penetration sensitivity stage, and the freezing stage, and after the optimization control of each stage is completed, the fitting result of the current stage is used as the initial input for the next stage; step S440, using the configured first target and second target to execute the three-stage temperature and humidity control optimization to establish collaborative control parameters.

[0035] Preferably, the frozen food label is processed by extracting fields, parsing the structured information in the label string, and determining the field extraction results, including ingredient type, unit volume, and target quality priority. Then, based on the field extraction results, constraints on ingredient type, unit volume, and target quality priority are established. Specifically, unique biothermodynamic parameters are retrieved from the knowledge base as ingredient type constraints, such as the critical temperature difference for cold shock rupture of blueberries, the protein denaturation temperature window of meat, and the starch aging sensitive temperature zone of pastries. The difficulty coefficient of heat penetration is calculated based on the size, shape, and loading of the food, and converted into unit volume constraints on the expected freezing time and required cooling intensity. For example, a large-volume steak requires a longer expected freezing time and stronger surface heat transfer than small blueberries. The quality requirements for premium and standard grades are quantified as the degree of stringency towards the primary objective, determining the target quality priority constraints. For example, very strict upper limits for ice crystal size and lower limits for tissue integrity are generated based on premium quality, while relatively lenient constraints are generated based on standard quality. Finally, the ingredient type constraints, unit volume constraints, and target quality priority constraints are organized into a structured data table to obtain a control constraint matrix.

[0036] Preferably, the weight factors of the control objectives within the collaborative control channel are nonlinearly compensated using a control constraint matrix. Specifically, the weight factors of the control objectives within the collaborative control channel are intelligently adjusted according to the strictness of each constraint in the control constraint matrix. If the unit volume constraint calculates that the current food has extremely high heat penetration difficulty, the weight of the shortest freezing time objective is appropriately increased to ensure that the solution to shorten the time is sought more actively without compromising quality. If the target quality priority constraint is extremely strict, the weights of tissue integrity and median ice crystal diameter are further strengthened to ensure that efficiency and economy objectives will never sacrifice quality. Finally, the collaborative control channel configuration is completed to optimize the temperature and humidity control of quick-frozen foods.

[0037] Preferably, based on the food label and initial dataset, stage trigger thresholds are set for the three stages of quick-freezing, i.e., defining clear switching conditions. The three stages are the initial freezing stage, the penetration-sensitive stage, and the freezing stage. When the food surface temperature reaches its freezing point and an initial ice shell forms, the process transitions from the initial freezing stage to the penetration-sensitive stage; when the food core temperature passes through the maximum ice crystal formation zone, the process transitions from the penetration-sensitive stage to the freezing stage. Quick-freezing optimization control is executed in stages, first addressing the optimization problem of the initial freezing stage. After this stage is completed, the process moves to the penetration-sensitive stage, and so on. Furthermore, after each stage's optimization control is completed, the fitting result of the current stage is used as the initial input for the next stage, meaning the final state determined in the previous stage is used as the starting condition for the next stage's optimization, ensuring the continuity and global optimality of the entire optimization process. Finally, the configured first and second objectives are used to perform three-stage temperature and humidity control optimization. Within the initial freezing, penetration-sensitive, and freezing stages, based on the boundaries defined by the control constraint matrix, the optimal combination of temperature, humidity, wind speed, and other setpoints for the current stage is determined, outputting complete collaborative control parameters to achieve precise control.

[0038] Furthermore, step S440 also includes step S441, compensating and correcting the first objective and the second objective based on the surface temperature and humidity control constraints of the initial condensation stage, and establishing a first-stage local objective function, which includes a surface temperature control term, a cabin humidity control term, and an energy consumption control term; step S442, using the initial dataset as the initial condition, and under the constraint conditions, using the second-stage trigger threshold as the control objective, performing control optimization using the first-stage local objective function, establishing the first-stage control optimization result, and generating the first-stage control fitting result; step S443, establishing collaborative control parameters based on the first-stage control optimization result and the first-stage control fitting result.

[0039] Preferably, the initial freezing stage refers to the process from when food is placed in the quick-freezing chamber until its surface temperature drops to the freezing point and a thin ice shell forms. This stage is used to quickly remove sensible heat and control the surface temperature and humidity of the food, preventing excessive evaporation of surface moisture that could lead to frostbite. Then, the first and second objectives are compensated and corrected based on the surface temperature and humidity control constraints of the initial freezing stage. Specifically, the surface temperature and humidity control constraints of the initial freezing stage are extracted from the control constraint matrix, including a maximum surface cooling rate constraint to prevent excessive cooling that could cause cold shock rupture of surface cells, and a minimum surface humidity constraint to prevent excessively dry air inside the chamber, which could lead to excessive sublimation of surface moisture and cause frostbite, such as surface whitening and deterioration of texture. For the first objective, since the food core is not yet frozen, the objective of minimizing the median ice crystal diameter is not applicable. Therefore, the compensation and correction of the first objective is to almost entirely focus on maximizing the integrity of the surface texture. The first objective, preventing surface frost damage and cracking, is to minimize the freezing time, which is reflected in the fastest possible surface frost formation. Minimizing unit energy consumption remains crucial. Therefore, a mathematical optimization function is constructed for the initial frost formation stage, including a weighted summation of surface temperature control, cabin humidity control, and energy consumption control. The surface temperature control minimizes the deviation between the current surface temperature and the ideal surface cooling curve, which is the curve that reaches the freezing point fastest and safest. The cabin humidity control minimizes the deviation between the current cabin humidity and the target humidity setpoint, which is set at a level that effectively inhibits excessive surface moisture sublimation. The energy consumption control minimizes the power consumption of the refrigeration and fan systems in this stage. The surface temperature and cabin humidity control have higher weights, while the energy consumption control has a lower weight.

[0040] Preferably, the initial temperature, initial humidity, and initial wind speed of the food in the initial dataset are used as initial conditions. Under constraints, the second-stage trigger threshold is used as the control objective, i.e., the average surface temperature reaching the freezing point of the food is used as the termination condition. The first-stage local objective function is used for control optimization, i.e., model predictive control is adopted. Under the premise of following all constraints, virtual control parameters such as compressor frequency, fan speed, and damper opening are repeatedly adjusted to simulate the initial condensation stage so that the value of the first-stage local objective function is minimized, and the time sequence of the optimal control command is obtained as the first-stage control optimization result. After executing the optimal control command, the state change curve is predicted as the first-stage control fitting result, which is used as the initial state for optimization in the penetration sensitive stage. Finally, the first-stage control optimization result and the first-stage control fitting result are integrated to generate cooperative control parameters.

[0041] Furthermore, step S443 also includes step A10, compensating and correcting the first and second objectives based on the ice crystal growth constraints and core temperature constraints of the penetration-sensitive stage, and establishing a second-stage local objective function. The second-stage local objective function includes a median diameter minimization evaluation term, a tissue integrity evaluation term, a core temperature control term, and an energy consumption control term. Step A20, using the first-stage control fitting result as the initial condition, and under the constraint conditions, using the third-stage trigger threshold as the control objective, performing control optimization using the second-stage local objective function, establishing the second-stage control optimization result, and generating the second-stage control fitting result. Step A30, establishing collaborative control parameters based on the first-stage control optimization result, the first-stage control fitting result, the second-stage control optimization result, and the second-stage control fitting result.

[0042] Preferably, the penetration-sensitive stage refers to the process from the start of ice formation on the food surface to the point where the core temperature passes through the maximum ice crystal formation zone. The core objective is to strictly control ice crystal growth and core temperature changes, minimizing the damage to cell structure caused by large ice crystals, ensuring small and uniform ice crystals and food tissue integrity, while also considering the freezing speed. Then, the first and second objectives are compensated and corrected based on the ice crystal growth constraints and core temperature constraints of the penetration-sensitive stage. Among them, the ice crystal growth constraint refers to the maximum allowable passage time constraint set to prevent excessive ice crystal growth, that is, the control strategy to be found to ensure that the residence time of the food core temperature in the maximum ice crystal formation zone is shorter than the critical value; the core temperature constraint refers to the limitation on the cooling rate at the core of the food. Too fast a rate may lead to excessive internal and external temperature differences and stress cracks, while too slow a rate may lead to ice crystal growth.

[0043] Preferably, the first objective is compensated and corrected, including strengthening the minimization of the median ice crystal diameter and maximizing tissue integrity. That is, minimizing the median ice crystal diameter is corrected as the core optimization objective, while maximizing tissue integrity is specified as minimizing the mechanical damage to cells caused by ice crystal growth and temperature stress. The second objective is compensated and corrected to keep its weight at a reasonable level to prevent energy consumption from running out of control. Then, a mathematical optimization function for the penetration-sensitive stage is constructed to determine the local objective function of the second stage, namely, a weighted summation of the median diameter minimization evaluation term, the tissue integrity evaluation term, the core temperature control term, and the energy consumption control term. Among them, the median diameter minimization evaluation term is used to suppress excessively large ice crystals, ensure a delicate structure, and improve food quality; the tissue integrity evaluation term is used to maintain the fiber structure and texture as much as possible and reduce freezing damage; the core temperature control term is used to control the core cooling rate to avoid local freezing too fast or too slow, so as to balance quality and freezing time; the energy consumption control term is used to control power consumption within a reasonable range to avoid unnecessary energy waste. Among them, the median diameter minimization evaluation term and the tissue integrity evaluation term have higher weights, while the core temperature control term and the energy consumption control term are auxiliary optimization terms with lower weights.

[0044] Preferably, the predicted state at the end of the initial freezing stage is used as the initial condition, and the third-stage trigger threshold is used as the control objective, i.e., the core temperature reaching the freezing stage is used as the termination condition, such as the core temperature being stably lower than the lower limit of the maximum ice crystal formation zone. The second-stage local objective function is used for control optimization. That is, under constraints such as ice crystal growth, core temperature, and equipment power, the simulation calculation is performed to minimize the second-stage local objective function to simulate different temperature and humidity control strategies, predict the ice crystal size, cell damage, core cooling curve, and energy consumption under each strategy, and finally find the strategy with the highest comprehensive score. Then, the optimal control command sequence for the penetration sensitive stage is generated as the second-stage control optimization result. After executing the optimal control command sequence for the penetration sensitive stage, the state at the end of the stage is predicted as the control fitting result, and it is used as the initial condition for the freezing stage. Finally, the first-stage control optimization result, the first-stage control fitting result, the second-stage control optimization result, and the second-stage control fitting result are integrated to generate cooperative control parameters.

[0045] Furthermore, step A30 also includes step A31, which involves performing compensation and correction of the first and second objectives based on the freezing penetration constraints and energy consumption constraints of the freezing stage, and establishing a third-stage local objective function. The third-stage local objective function includes a core temperature evaluation term, a freezing penetration time minimization term, an energy consumption minimization term, and an in-cabin humidity control term. Step A32 involves using the second-stage control fitting result as the initial condition, and under the constraint conditions, using the fourth-stage trigger threshold as the control objective, and using the third-stage local objective function to perform control optimization, thereby establishing the third-stage control optimization result. Step A33 involves establishing collaborative control parameters based on the first-stage control optimization result, the second-stage control optimization result, and the third-stage control optimization result.

[0046] Preferably, the freezing stage lowers the core temperature of the food to the required final storage temperature. The core objective is to efficiently complete the process and minimize energy consumption while ensuring the food is completely frozen. Then, based on the freezing stage's freezing penetration constraints and energy consumption constraints, compensation and corrections are made for the first and second objectives. The freezing penetration constraint must ensure that the final core temperature of the food reaches or falls below the target endpoint temperature. The energy consumption constraint refers to ensuring that the energy consumption based on total production costs or equipment operation does not exceed the energy consumption limit. During the freezing stage, the ice crystal morphology and cell structure inside the food are basically finalized. The importance of the first objective decreases significantly, transforming into a protective objective, namely, preventing the recrystallization and enlargement of ice crystals due to temperature fluctuations. The second objective becomes absolutely dominant, with the shortest freezing time being the primary objective. Minimizing unit energy consumption becomes the core optimization objective. A mathematical optimization function is constructed specifically for the freezing phase, namely, a weighted summation of the core temperature evaluation term, the freezing time minimization term, the energy consumption minimization term, and the cabin humidity control term to obtain the third-stage local objective function. Among them, the core temperature evaluation term is used to minimize the deviation between the predicted final core temperature and the target endpoint temperature; the freezing time minimization term is a penalty term positively correlated with the duration of this phase, seeking the fastest possible cooling strategy to shorten the time and improve equipment turnover; the energy consumption minimization term is a penalty term positively correlated with the power consumption of equipment such as refrigeration compressors and fans in this phase; the cabin humidity control term is used to maintain the cabin humidity at a reasonably low level to prevent excessive frost formation on the evaporator surface.

[0047] Preferably, the predicted state at the end of the penetration-sensitive stage optimization is used as the initial condition, and the fourth stage trigger threshold is used as the control target, i.e., the final endpoint temperature at which the core temperature of the food reaches the process requirements is used as the termination condition. Under the constraints of freezing through and energy consumption, the local objective function of the third stage is minimized for simulation calculation to simulate the advantages and disadvantages of different strategies. For example, the energy consumption of the fan can be significantly saved by appropriately reducing the wind speed, and the evaporation temperature can be increased when the core temperature is close to the target. Finally, the strategy that is optimal in terms of time consumption and energy consumption under the condition of meeting the freezing through requirement is determined as the optimal control command sequence for the freezing stage, and the control optimization result of the third stage is generated. Finally, the control optimization results of the first stage, the second stage, and the third stage are integrated to establish collaborative control parameters, covering the control commands of the entire process from entering the chamber to leaving the chamber. These commands can be directly issued to the actuators at all levels of the quick-freezing chamber and realize the collaborative optimization of multiple parameters such as temperature, humidity, and wind, thereby ensuring the improvement of quick-frozen food quality, quick-freezing efficiency, and reduction of energy consumption.

[0048] Furthermore, step S400 also includes step S450, performing quick-freezing chamber control with the cooperative control parameters and setting a response node based on the cooperative control parameters; step S460, activating the edge box at the response node to perform response verification and establishing response compensation data; step S470, adaptively adjusting the parameters of the cooperative control parameters according to the response compensation data.

[0049] Preferably, the quick-freezing compartment control is executed using coordinated control parameters. Specifically, the controller within the edge box executes specific operational instructions from the coordinated control parameters according to time sequence and logical conditions. This includes adjusting the compressor's operating frequency and the evaporator's liquid supply valve opening to precisely control the cooling capacity and evaporation temperature; adjusting the fan speed and damper opening angle to control the airflow speed and uniformity within the compartment; and controlling the defrost heater's start / stop / power or the humidifier's atomization level to maintain the target humidity within the compartment. Simultaneously, response nodes are set based on the coordinated control parameters, i.e., verification points are set at preset time points or key state points in the coordinated control parameters. When the control execution reaches the set... When a response node is set, the edge box is activated to perform response verification, which involves collecting real-time temperature, humidity, and wind speed again and comparing them with the predicted values ​​of that node in the collaborative control parameters. The deviation between the actual and predicted values ​​is calculated to establish response compensation data. Then, the collaborative control parameters are dynamically fine-tuned based on the response compensation data, including analyzing whether the deviation exceeds the normal range, determining whether it is a random fluctuation or a persistent trend, and calculating the amount of adjustment to the collaborative control parameters based on the magnitude and direction of the deviation. For example, if the cooling rate is found to be slower than predicted, the power setpoint of the compressor is automatically increased slightly. Finally, a correction control command is generated and continues to be executed.

[0050] Furthermore, step S470 also includes step S471, configuring a compensation target based on the response compensation data; step S472, reading the node interval of the response node, identifying control anomalies in the quick-freezing chamber based on the node interval and the response compensation data, and establishing control anomaly compensation; step S473, adaptively adjusting the parameters of the collaborative control parameters based on the compensation target and control anomaly compensation.

[0051] Preferably, the analysis of response compensation data configures compensation targets. Response compensation data represents the deviation between the actual measured value and the predicted value at the response node. If the response compensation data is positive, it indicates that the cooling is too slow, and the compensation target is to moderately increase the cooling rate in subsequent stages. If the response compensation data is negative, it indicates that the cooling is too fast, and the compensation target is to moderately reduce the cooling rate in subsequent stages. The node interval of the response node is read, i.e., the time length between the current response node and the previous response node is determined. Then, based on the node interval and response compensation data, control anomalies in the quick-freezing chamber are identified. This involves calculating the trend and rate of deviation change by combining the magnitude of the deviation and the time interval, thereby determining the root cause of the anomaly, such as a small and stable deviation, a continuously increasing deviation, or a sudden and violent fluctuation in deviation. Control anomaly compensation is then established, including calculating the compensation amount to compensate for the current deviation and anomaly trend based on the identified anomaly type and severity. Finally, based on the compensation target and control anomaly compensation, adaptive adjustment of the collaborative control parameters is performed, i.e., determining the amount of modification to the control command sequence that has not yet been executed. This may include adjusting stage setpoints, adjusting equipment outputs, and adjusting time parameters.

[0052] Furthermore, step S470 also includes performing compensation warning identification on the response compensation data; if the compensation warning identification result is a warning trigger result, a warning reset instruction is generated, and the collaborative control parameters are reconstructed according to the warning reset instruction.

[0053] Preferably, the compensation warning identification refers to comparing the current deviation and trend with the preset safety threshold to determine whether it has deviated from the controllable range, such as the absolute value of the deviation exceeding the limit, the cumulative deviation exceeding the limit, or the rate of change of the deviation exceeding the limit. If the compensation warning identification result is a warning trigger result, it indicates that it is in an abnormal state, and a warning reset instruction is generated, including immediately stopping the current strategy and starting global replanning, including resetting the initial conditions, using the actual measurement values ​​of all sensors collected by the edge box at the current moment as the new initial dataset, rereading the frozen food label and the current power of the equipment, reconfiguring the constraints, and then performing a three-stage optimization to generate collaborative control parameters that match the current actual situation, thereby enhancing the reliability of the entire system, improving the quality and freezing efficiency of frozen food, and reducing energy consumption.

[0054] In the above text, refer to Figure 1 This paper describes in detail a food quick-freezing temperature and humidity coordinated control method based on edge computing according to embodiments of the present invention. Next, reference will be made to... Figure 2 This invention describes a food quick-freezing temperature and humidity coordinated control system based on edge computing according to an embodiment of the present invention.

[0055] The edge computing-based temperature and humidity coordinated control system for food quick-freezing according to embodiments of the present invention addresses the technical problems in existing technologies, such as high latency in data transmission and cloud-based decision-making response, and the inability to coordinate temperature and humidity optimization, leading to poor quick-freezing efficiency and product quality. It achieves the technical effects of improving the quality and efficiency of quick-freezing foods while reducing energy consumption. Figure 2 As shown, the edge computing-based food quick-freezing temperature and humidity coordinated control system includes: a data acquisition module 10, a constraint configuration module 20, a control channel configuration module 30, and a temperature and humidity control optimization module 40.

[0056] The data acquisition module 10 is used to configure an edge box in the quick-freezing chamber. The edge box integrates a surface temperature array, a humidity probe, and an anemometer. It uses the edge box to perform data acquisition within the quick-freezing chamber and establish an initial dataset. The constraint configuration module 20 is used to acquire the labels of the quick-frozen foods placed in the quick-freezing chamber and read the power of the equipment in the quick-freezing chamber. It configures constraint conditions based on the labels of the quick-frozen foods and the power of the equipment. The control channel configuration module 30 is used to configure a collaborative control channel. The collaborative control channel includes a first objective and a second objective. The first objective is the primary objective, including maximizing tissue integrity and minimizing the median diameter of ice crystals. The second objective is the secondary objective, including minimizing the freezing time and minimizing the unit energy consumption. The temperature and humidity control optimization module 40 is used to configure the collaborative control channel using the labels of the quick-frozen foods, and then, using the initial dataset as initial conditions, performs a three-stage temperature and humidity control optimization under the constraints to establish collaborative control parameters. It then performs control and management of the quick-freezing chamber based on the collaborative control parameters.

[0057] The specific configuration of the temperature and humidity control optimization module 40 will be described in detail below. The temperature and humidity control optimization module 40 further includes: extracting fields from the frozen food label; using the field extraction results to establish food type constraints, unit volume constraints, and target quality priority constraints to form a control constraint matrix; using the control constraint matrix to perform nonlinear compensation on the weight factors of the control targets within the collaborative control channel to complete the collaborative control channel configuration; using the frozen food label and the initial dataset to set the stage trigger thresholds for the three stages, and then executing the three-stage temperature and humidity control optimization sequentially, wherein the three stages are the initial condensation stage, the penetration sensitivity stage, and the freezing stage, and after each stage optimization control is completed, the fitting result of the current stage is used as the initial input for the next stage; using the configured first and second targets to execute the three-stage temperature and humidity control optimization and establish collaborative control parameters.

[0058] The specific configuration of the temperature and humidity control optimization module 40 will be described in detail below. The temperature and humidity control optimization module 40 further includes: compensating and correcting the first and second objectives based on the surface temperature and humidity control constraints of the initial condensation stage; establishing a first-stage local objective function, which includes a surface temperature control term, a cabin humidity control term, and an energy consumption control term; using the initial dataset as initial conditions, and under the constraints, using the second-stage trigger threshold as the control objective, performing control optimization using the first-stage local objective function, establishing the first-stage control optimization result, and generating the first-stage control fitting result; and establishing collaborative control parameters based on the first-stage control optimization result and the first-stage control fitting result.

[0059] The specific configuration of the temperature and humidity control optimization module 40 will be described in detail below. The temperature and humidity control optimization module 40 further includes: compensating and correcting the first and second objectives based on the ice crystal growth constraints and core temperature constraints of the penetration-sensitive stage; establishing a second-stage local objective function, which includes a median diameter minimization evaluation term, a tissue integrity evaluation term, a core temperature control term, and an energy consumption control term; using the first-stage control fitting result as the initial condition, and under the constraint conditions, using the third-stage trigger threshold as the control objective, performing control optimization using the second-stage local objective function, establishing the second-stage control optimization result, and generating the second-stage control fitting result; and establishing collaborative control parameters based on the first-stage control optimization result, the first-stage control fitting result, the second-stage control optimization result, and the second-stage control fitting result.

[0060] The specific configuration of the temperature and humidity control optimization module 40 will be described in detail below. The temperature and humidity control optimization module 40 further includes: compensating and correcting the first and second objectives based on the freezing-through constraints and energy consumption constraints of the freezing stage; establishing a third-stage local objective function, which includes a core temperature evaluation term, a freezing-through time minimization term, an energy consumption minimization term, and an in-cabin humidity control term; using the second-stage control fitting result as the initial condition, and under the constraint conditions, using the fourth-stage trigger threshold as the control objective, performing control optimization using the third-stage local objective function to establish the third-stage control optimization result; and establishing collaborative control parameters based on the first-stage control optimization result, the second-stage control optimization result, and the third-stage control optimization result.

[0061] The specific configuration of the data acquisition module 10 will be described in detail below. The data acquisition module 10 further includes: acquiring raw temperature, humidity, and wind speed data from edge computing nodes built into the edge box; performing filtering and outlier processing; and establishing a data time series after synchronization via timestamps; constructing a three-dimensional temperature distribution grid using a temperature data array; mapping humidity and wind speed data to the location of frozen food; and performing rapid feature calculations on the edge computing nodes based on the three-dimensional temperature distribution grid and the local environmental state vector. These rapid feature calculations include temperature gradient calculation, humidity gradient calculation, and wind speed distribution analysis, outputting an initial dataset.

[0062] The specific configuration of the temperature and humidity control optimization module 40 will be described in detail below. The temperature and humidity control optimization module 40 further includes: executing quick-freezing chamber control using the cooperative control parameters, and setting response nodes based on the cooperative control parameters; activating edge boxes at the response nodes to perform response verification and establish response compensation data; and adaptively adjusting the parameters of the cooperative control parameters according to the response compensation data.

[0063] The specific configuration of the temperature and humidity control optimization module 40 will be described in detail below. The temperature and humidity control optimization module 40 further includes: configuring a compensation target based on the response compensation data; reading the node interval of the response node, identifying control anomalies in the quick-freezing chamber based on the node interval and the response compensation data, and establishing control anomaly compensation; and adaptively adjusting the parameters of the coordinated control parameters based on the compensation target and the control anomaly compensation.

[0064] The specific configuration of the temperature and humidity control optimization module 40 will be described in detail below. The temperature and humidity control optimization module 40 further includes: performing compensation early warning identification on the response compensation data; if the compensation early warning identification result is an early warning trigger result, generating an early warning reset instruction, and reconstructing the collaborative control parameters according to the early warning reset instruction.

[0065] The edge computing-based food quick-freezing temperature and humidity coordinated control system provided in this embodiment of the invention can execute the edge computing-based food quick-freezing temperature and humidity coordinated control method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0066] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0067] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A food quick-freezing temperature and humidity co-control method based on edge computing, characterized in that, The method comprises: An edge box is configured in the quick-freezing cabin, the edge box is integrated with a surface temperature array, a humidity probe, and an anemometer, data collection in the quick-freezing cabin is performed by using the edge box, and an initial data set is established; A quick-frozen food label placed in the quick-freezing cabin is obtained, and equipment power of the quick-freezing cabin is read, and constraint conditions are configured according to the quick-frozen food label and the equipment power; A cooperative control channel is configured, the cooperative control channel is a cooperative control channel comprising a first target and a second target, wherein the first target is a primary target comprising a maximum tissue integrity target and a minimum ice crystal median diameter target, and the second target is a secondary target comprising a minimum freezing-through time target and a minimum unit energy consumption target; After the cooperative control channel is configured by using the quick-frozen food label, the initial data set is taken as an initial condition, three-stage temperature and humidity control optimization is performed under constraint conditions, cooperative control parameters are established, and the quick-freezing cabin is controlled and managed according to the cooperative control parameters.

2. The edge computing based food flash freezing temperature and humidity cooperative control method of claim 1, wherein, After the cooperative control channel is configured by using the quick-frozen food label, the initial data set is taken as an initial condition, three-stage temperature and humidity control optimization is performed under constraint conditions, cooperative control parameters are established, and the quick-freezing cabin is controlled and managed according to the cooperative control parameters. After the quick-frozen food label is field-extracted, a control constraint matrix is established by using a field extraction result, material type constraints, unit volume constraints, and target quality priority constraints are formed, and the cooperative control channel is configured by using the control constraint matrix to perform nonlinear compensation on weight factors of control targets in the cooperative control channel; After the quick-frozen food label and the initial data set are used to set stage trigger thresholds of three stages, three-stage temperature and humidity control optimization is performed in sequence, wherein the three stages are a primary coagulation stage, a penetration sensitive stage, and a freezing stage, and after optimization control in each stage is completed, a fitting result of the current stage is taken as an initial input of the next stage; Three-stage temperature and humidity control optimization is performed by using the configured first target and the second target, and cooperative control parameters are established. Three-stage temperature and humidity control optimization is performed by using the configured first target and the second target, and cooperative control parameters are established. 3.The food quick-freezing temperature and humidity co-ordinated control method based on edge computing according to claim 2, wherein, According to the surface temperature and humidity control constraints of the primary coagulation stage, the first target and the second target are compensated and modified, a first-stage local objective function is established, and the first-stage local objective function comprises a surface temperature control item, a cabin humidity control item, and an energy consumption control item; The initial data set is taken as an initial condition, a second-stage trigger threshold is taken as a control target under constraint conditions, control optimization is performed by using the first-stage local objective function, a first-stage control optimization result is established, and a first-stage control fitting result is generated; Cooperative control parameters are established according to the first-stage control optimization result and the first-stage control fitting result. Cooperative control parameters are established according to the first-stage control optimization result and the first-stage control fitting result.

4. The edge computing based food flash freezing temperature and humidity cooperative control method of claim 3, wherein, ​ According to the ice crystal growth constraint of the penetration sensitive stage and the core temperature constraint, the first target and the second target are compensated and corrected, a second stage local target function is established, and the second stage local target function includes a median diameter minimization evaluation item, an organization integrity evaluation item, a core temperature control item and an energy consumption control item; The first stage control fitting result is taken as an initial condition, a third stage trigger threshold is taken as a control target under the constraint of a constraint condition, the second stage local target function is used for control optimization, a second stage control optimization result is established, and a second stage control fitting result is generated; According to the first stage control optimization result, the first stage control fitting result, the second stage control optimization result and the second stage control fitting result, a collaborative control parameter is established.

5. The edge computing based food flash freezing temperature and humidity co-control method according to claim 4, characterized in that, The establishment of the collaborative control parameter according to the first stage control optimization result, the first stage control fitting result, the second stage control optimization result and the second stage control fitting result includes: According to the freeze-through constraint and the energy consumption constraint of the freezing stage, the first target and the second target are compensated and corrected, a third stage local target function is established, and the third stage local target function includes a core temperature evaluation item, a freeze-through time minimization item, an energy consumption minimization item and a cabin humidity control item; The second stage control fitting result is taken as an initial condition, a fourth stage trigger threshold is taken as a control target under the constraint of a constraint condition, the third stage local target function is used for control optimization, and a third stage control optimization result is established; According to the first stage control optimization result, the second stage control optimization result and the third stage control optimization result, a collaborative control parameter is established.

6. The edge computing based food flash freezing temperature and humidity co-control method of claim 1, wherein, The establishment of the initial data set by using the edge box to perform the quick freezing cabin data acquisition includes: The edge computing node built-in the edge box acquires the temperature, humidity and wind speed original data, performs filtering and outlier processing, and establishes a data time sequence after synchronization through a time stamp; A three-dimensional temperature distribution grid is constructed by using a temperature data array, and humidity and wind speed data are mapped to a food position to establish a local environment state vector; The edge computing node performs rapid feature calculation based on the three-dimensional temperature distribution grid and the local environment state vector, and the rapid feature calculation includes temperature gradient calculation, humidity gradient calculation and wind speed distribution analysis, and an initial data set is output.

7. The edge computing based food flash freezing temperature and humidity co-ordination control method of claim 1, wherein, The quick freezing cabin control management according to the collaborative control parameter includes: The quick freezing cabin control is performed according to the collaborative control parameter, and a response node is set based on the collaborative control parameter; The edge box is activated at the response node to perform response verification, and response compensation data is established; The parameter adaptive adjustment of the collaborative control parameter is performed according to the response compensation data. 8.The food quick-freezing temperature and humidity co-ordinated control method based on edge computing according to claim 7, wherein, The parameter adaptive adjustment of the collaborative control parameter according to the response compensation data includes: A compensation target is configured according to the response compensation data; The node interval of the response node is read, the control abnormality identification of the quick freezing cabin is performed according to the node interval and the response compensation data, and control abnormality compensation is established; The parameter adaptive adjustment of the collaborative control parameter is performed according to the compensation target and the control abnormality compensation. 9.The edge computing based food flash freezing temperature and humidity cooperative control method of claim 7, wherein, Before the parameter self-adaptive adjustment of the cooperative control parameter according to the response compensation data, the method comprises: Compensation early warning identification is performed on the response compensation data; If the compensation early warning identification result is an early warning triggering result, an early warning reset instruction is generated, and the cooperative control parameter is reconstructed according to the early warning reset instruction.

10. The food quick-freezing temperature and humidity co-ordinated control system based on edge computing, characterized in that, The system is used to implement the food quick-freezing temperature and humidity cooperative control method based on edge computing according to any one of claims 1 to 9, and the system comprises: A data acquisition module is configured with an edge box in the quick-freezing cabin, the edge box is integrated with a surface temperature array, a humidity probe and an anemometer, data acquisition in the quick-freezing cabin is performed by using the edge box, and an initial data set is established; A constraint configuration module is used to obtain a quick-freezing food label placed in the quick-freezing cabin and read the equipment power of the quick-freezing cabin, and constraint conditions are configured according to the quick-freezing food label and the equipment power; A control channel configuration module is used to configure a cooperative control channel, the cooperative control channel is a cooperative control channel comprising a first target and a second target, the first target is a main target, comprising a maximum tissue integrity target and a minimum ice crystal median diameter target, and the second target is a secondary target, comprising a shortest freezing-through time target and a minimum unit energy consumption target; A temperature and humidity control optimization module is used to perform three-stage temperature and humidity control optimization under the constraint conditions by taking the initial data set as an initial condition after the cooperative control channel is configured by using the quick-freezing food label, establish cooperative control parameters, and perform quick-freezing cabin control management according to the cooperative control parameters.

Images