Intelligent cold chain integrated management platform and method

Abstract

This invention relates to the field of cold chain logistics technology, specifically disclosing a smart cold chain integrated management platform and method. The platform includes an ethylene sensor, a temperature sensor, an edge computing gateway, an industrial internet platform, and a multi-unit collaborative controller. The edge computing gateway constructs a maturity evolution curve based on ethylene concentration values ​​and uploads this curve to the industrial internet platform through edge-cloud collaboration. The industrial internet platform constructs an advanced temperature setting curve based on the maturity evolution curve and thermal inertia compensation time. The edge computing gateway generates an advanced control command sequence based on the advanced temperature setting curve and the temperature value inside the warehouse. The multi-unit collaborative controller controls the operation of the refrigeration units according to the advanced control command sequence. This invention, through the collaboration of edge computing and the cloud platform, predicts the trend of cargo maturity changes, adjusts the refrigeration strategy in advance, compensates for the thermal inertia lag of the refrigeration system, and matches the refrigeration system's response lag with the time window of quality changes, thereby reducing energy consumption and cargo damage rate.

Description

Technical Field

[0001] This invention relates to the field of cold chain logistics technology, and in particular to an intelligent integrated cold chain management platform and method. Background Technology

[0002] In the cold chain logistics sector, there is an inherent contradiction between the lag in the response of refrigeration systems and the rapid changes in cargo quality. Refrigeration systems have significant thermal inertia, with a substantial time delay between the issuance of control commands and the actual change in temperature within the storage facility. Meanwhile, goods such as climacteric fruits experience rapid quality changes during ripening, especially near the climacteric threshold where ethylene concentration rises sharply, leading to rapid changes in ripeness. With the rapid development of the Industrial Internet and edge computing technologies, how to utilize an edge-cloud collaborative architecture to achieve intelligent control of cold chain equipment has become an important direction for technological upgrading in the industry.

[0003] Existing cold chain management systems typically employ a fixed temperature setpoint control strategy, responding only to current temperature deviations and failing to detect changes in cargo status, let alone predict trends in cargo quality and adjust refrigeration strategies accordingly. Current systems generally use centralized cloud platforms to process all data, resulting in significant delays in control command response and failing to meet real-time requirements. Simultaneously, the lack of edge computing capabilities prevents timely local data processing, hindering accurate compensation for the thermal inertia of the refrigeration system. This "lagging response" control mode leads to a mismatch between the refrigeration strategy and the time window for quality changes: by the time the refrigeration system begins to respond, the cargo quality has already changed, resulting in energy waste and cargo quality loss. Therefore, existing technologies struggle to reconcile the contradiction between the delayed response of the refrigeration system's thermal inertia and the rapid changes in cargo quality.

[0004] Therefore, this invention proposes an integrated intelligent cold chain management platform and method. Summary of the Invention

[0005] This invention provides an integrated intelligent cold chain management platform and method. By sensing the trend of goods maturity changes through ethylene concentration, and using thermal inertia compensation time to shift the maturity evolution curve over time, an advanced temperature setting curve is constructed and an advanced control command sequence is generated. This transforms the refrigeration system from responding to temperature deviations with lag to predicting quality changes in advance, enabling precise matching between the refrigeration system's response lag and the time window of goods quality changes, thereby reducing energy consumption and cargo loss rate.

[0006] This invention provides an integrated intelligent cold chain management platform, comprising: Ethylene sensor, used to collect ethylene concentration values ​​in the cold chain environment in real time; Temperature sensors are used to collect real-time temperature values ​​inside the cold chain environment. Radio frequency identification (RFID) readers are used to read product category identifiers. The edge computing gateway is connected to the ethylene sensor, temperature sensor, and RFID reader, respectively. The edge computing gateway is configured as follows: Based on the change of ethylene concentration over time, a maturity evolution curve is constructed, which characterizes the trend of maturity level change over time. The industrial internet platform connects with the edge computing gateway through end-cloud collaboration. The industrial internet platform is used to construct an advanced temperature setting curve based on the maturity evolution curve and the preset thermal inertia compensation time. The advanced temperature setting curve is a sequence of temperature setting values ​​that are ahead of the maturity level setting to compensate for the thermal inertia of the refrigeration system. The edge computing gateway is also used to generate a sequence of advanced control commands based on the advanced temperature setting curve and the temperature value inside the refrigeration system. The sequence of advanced control commands is a set of control commands generated ahead of the temperature deviation inside the refrigeration system to compensate for the thermal inertia of the refrigeration system. The multi-unit collaborative controller, connected to the edge computing gateway, is used to control the operation of each chiller unit according to the sequence of advanced control commands.

[0007] Furthermore, when constructing the maturity evolution curve, the edge computing gateway uses the maturity level as the initial condition and the ethylene concentration value as the basis for the change in the maturity evolution rate. It outputs the maturity evolution curve by iteratively solving the maturity evolution differential equation. The rate of change of maturity level in the maturity evolution differential equation is positively correlated with the difference between the maturity level and the maximum maturity level, as well as the ethylene concentration.

[0008] Furthermore, before outputting the maturity evolution curve, the edge computing gateway first identifies whether the maturity level is at the critical point of the breathing leap. The edge computing gateway calculates the acceleration of the maturity evolution curve. When the acceleration of the maturity evolution curve remains positive within a preset time window and the growth rate of the ethylene concentration exceeds a preset growth rate threshold, the maturity level is determined to be at the critical point of the breathing jump. When the edge computing gateway determines that it is at the critical point of the breathing transition, it adjusts the maturity evolution rate constant in the maturity evolution differential equation to a multiple of its original value and outputs the final maturity evolution curve.

[0009] Furthermore, when constructing the advanced temperature setting curve, the industrial internet platform performs a time shift on the maturity evolution curve using a preset thermal inertia compensation time. The maturity level at the corresponding time point after the shift is mapped to the optimal storage temperature. After subtracting the thermal inertia advanced compensation amount from the optimal storage temperature, the advanced temperature setting curve is output.

[0010] Furthermore, the thermal inertia compensation time is determined using a system identification method; During the platform initialization phase, the edge computing gateway inputs a step control signal to the refrigeration system and collects the response curve of the temperature value inside the refrigeration chamber. Taking the input time of the step control signal as the starting point and the time point when the temperature value inside the refrigeration chamber reaches the set percentage of the steady-state value as the ending point, the time difference between the starting point and the ending point is used as the thermal inertia compensation time.

[0011] Furthermore, when the edge computing gateway generates the advance control command sequence, it inputs the deviation between the advance temperature setpoint and the temperature value in the warehouse into the proportional-integral-derivative controller to obtain the first control component, multiplies the maturity change rate by a preset maturity change rate weighting coefficient to obtain the second control component, and superimposes the first control component and the second control component and multiplies it by the thermal inertia advance compensation coefficient to output the advance control command sequence.

[0012] Furthermore, it also includes: The frost compensation unit is used to acquire images of the evaporator surface and output frost thickness values, mapping the frost thickness values ​​to a heat exchange efficiency attenuation coefficient. The edge computing gateway receives the heat exchange efficiency attenuation coefficient, multiplies it by the advance control command sequence, and corrects the advance control command sequence.

[0013] Furthermore, it also includes: The blockchain evidence storage unit is used to package ethylene concentration values, storage temperature values, maturity evolution curves, advanced temperature setting curves, and advanced control command sequences into evidence storage data blocks, and then upload the evidence storage data blocks to the blockchain network after digitally signing them. Before constructing the maturity evolution curve, the edge computing gateway obtains historical evidence data blocks stored in the blockchain evidence storage unit, extracts the historical ethylene concentration value sequence with the same product category identifier and the corresponding historical maturity level sequence from the historical evidence data block, and refits the maturity evolution rate constant in the maturity evolution differential equation with the historical ethylene concentration value sequence as input and the historical maturity level sequence as output.

[0014] Furthermore, it also includes: The digital twin mapping unit is used to receive ethylene concentration values, storage temperature values, maturity evolution curves, advanced temperature setting curves, and advanced control command sequences, construct a three-dimensional geometric model of the cold chain space, and map the maturity evolution curves, advanced temperature setting curves, advanced control command sequences, and storage temperature values ​​onto the three-dimensional geometric model to generate a dynamic visualization interface. The dynamic visualization interface displays the advanced-lag relationship between the maturity evolution curve and the storage temperature response curve.

[0015] This invention provides a smart cold chain integrated management method, applicable to any of the above-mentioned smart cold chain integrated management platforms, comprising: Real-time collection of ethylene concentration and temperature values ​​in the cold chain environment, and reading of cargo category identification; Based on the change of ethylene concentration over time, a maturity evolution curve is constructed, which characterizes the trend of maturity level change over time. Based on the maturity evolution curve and the preset thermal inertia compensation time, an advanced temperature setting curve is constructed. The advanced temperature setting curve is a sequence of temperature setting values ​​that are ahead of the maturity level setting to compensate for the thermal inertia of the refrigeration system. Based on the advanced temperature setpoint curve and the temperature value inside the refrigeration unit, an advanced control command sequence is generated. The advanced control command sequence is a control command generated ahead of the temperature deviation inside the refrigeration unit to compensate for the thermal inertia of the refrigeration system. The operation of each refrigeration unit is controlled according to the sequence of advance control commands.

[0016] The advantages of this invention compared to existing technologies are as follows: Existing cold chain management systems rely solely on temperature sensors to collect warehouse temperatures and control the start and stop of refrigeration units based on preset fixed temperature thresholds, employing a lag-response control mode. This technical solution cannot perceive the actual maturity state of goods, nor can it predict the changing trends in goods quality. This results in a severe time window misalignment between the lag in the thermal inertia response of the refrigeration system and the rapid changes in goods quality. By the time the refrigeration system begins to respond, the goods quality has already undergone irreversible deterioration. This invention introduces an ethylene sensor to perceive the maturity of goods in real time, constructs a maturity evolution curve to predict quality change trends, and uses thermal inertia compensation time to shift the maturity evolution curve over time, constructing an advanced temperature setting curve so that the temperature target precedes quality changes. Then, an edge computing gateway generates an advanced control command sequence to compensate for the thermal inertia lag of the refrigeration system, and finally, a multi-unit collaborative controller executes the control commands. This invention achieves a fundamental shift in refrigeration strategy from delayed response to proactive prediction, enabling precise matching of the refrigeration system's response lag with the time window of changes in cargo quality. This significantly reduces the energy consumption of the refrigeration system and minimizes cargo quality loss due to temperature lag. At the same time, it achieves an organic combination of real-time edge response and global optimization in the cloud through an edge-cloud collaborative architecture.

[0017] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is an architecture diagram of the intelligent cold chain integrated management platform in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the generation of the advance control instruction sequence in an embodiment of the present invention. Detailed Implementation

[0020] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0021] refer to Figure 1 and Figure 2 This invention provides an embodiment of an integrated intelligent cold chain management platform, comprising: Ethylene sensor, used to collect ethylene concentration values ​​in the cold chain environment in real time; Temperature sensors are used to collect real-time temperature values ​​inside the cold chain environment. Radio frequency identification (RFID) readers are used to read product category identifiers. The edge computing gateway is connected to the ethylene sensor, temperature sensor, and RFID reader, respectively. The edge computing gateway is configured as follows: Based on the change of ethylene concentration over time, a maturity evolution curve is constructed, which characterizes the trend of maturity level change over time. The industrial internet platform connects with the edge computing gateway through end-cloud collaboration. The industrial internet platform is used to construct an advanced temperature setting curve based on the maturity evolution curve and the preset thermal inertia compensation time. The advanced temperature setting curve is a sequence of temperature setting values ​​that are ahead of the maturity level setting to compensate for the thermal inertia of the refrigeration system. The edge computing gateway is also used to generate a sequence of advanced control commands based on the advanced temperature setting curve and the temperature value inside the refrigeration system. The sequence of advanced control commands is a set of control commands generated ahead of the temperature deviation inside the refrigeration system to compensate for the thermal inertia of the refrigeration system. The multi-unit collaborative controller, connected to the edge computing gateway, is used to control the operation of each chiller unit according to the sequence of advanced control commands.

[0022] In this embodiment, the ethylene sensor is installed above or to the side of the goods storage area inside the cold storage. The ethylene sensor is an electrochemical ethylene gas sensor. The ethylene sensor continuously collects the ethylene concentration value in the air inside the cold storage at a sampling frequency of once per minute. The unit of the ethylene concentration value is one part per million. The ethylene sensor transmits the collected ethylene concentration value to the edge computing gateway in real time through the data bus.

[0023] In this embodiment, a platinum resistance temperature sensor is used as the temperature sensor. The temperature sensor is installed in multiple different locations inside the cold storage, including near the return air vent and in the center of the goods storage area. The temperature sensor continuously collects the internal temperature value of the cold storage at a sampling frequency of once per minute. The unit of the internal temperature value is degrees Celsius. The temperature sensor transmits the collected internal temperature value to the edge computing gateway in real time through the data bus.

[0024] In this embodiment, the cargo category identifier is a unique identifier used to identify the category to which the cold chain stored goods belong. An RFID reader is installed at the entrance of the cold storage or in the cargo storage area. When goods enter the storage area, the RFID reader reads the RFID tag attached to the cargo packaging. The RFID tag pre-stores the cargo category identifier, which includes identifiers for bananas, apples, mangoes, strawberries, etc. The RFID reader transmits the read cargo category identifier to the edge computing gateway via a data bus. The edge computing gateway then queries a preset category parameter table based on the cargo category identifier to obtain preset parameters such as the initial maturity level, maturity evolution rate constant, optimal storage temperature for each maturity level, and respiratory hopping threshold criterion parameters corresponding to that category. The preset category parameter table is stored in the non-volatile memory of the edge computing gateway. The table records various parameters corresponding to different product category identifiers. For the banana category identifier, the initial maturity level is 0.2, the maturity evolution rate constant is 0.02 per hour, the optimal storage temperature is 13 degrees Celsius for a maturity level of 0.2, 15 degrees Celsius for a maturity level of 0.5, and 17 degrees Celsius for a maturity level of 0.8. The preset time window for the respiratory jump critical point criterion is 30 minutes, and the preset growth rate threshold is 10% per hour.

[0025] In this embodiment, the maturity evolution curve is a curve characterizing the trend of maturity level changes over time. The edge computing gateway uses maturity level as the initial condition and ethylene concentration as the basis for the change rate of maturity evolution. It iteratively solves the maturity evolution differential equation and outputs the maturity evolution curve. The maturity evolution differential equation adopts a differential equation based on the logistic growth model. The rate of change of maturity level in the differential equation is positively correlated with the difference between the maturity level and the maximum maturity level and the ethylene concentration. The edge computing gateway uses the Euler method to iteratively solve the maturity evolution differential equation. The Euler method discretizes continuous time into equally spaced time steps. Within each time step, the maturity level at the next moment is calculated based on the current maturity level and the ethylene concentration. The complete sequence of maturity level changes over time is obtained by recursion, which is the maturity evolution curve.

[0026] In this embodiment, the thermal inertia compensation time refers to the time delay required from the issuance of the refrigeration system control command to the actual change in the temperature inside the refrigeration chamber. This time delay reflects the magnitude of the thermal inertia of the refrigeration system. The thermal inertia compensation time is predetermined by the system identification method, which adopts the step response method. During the platform initialization phase, the edge computing gateway inputs a step control signal to the refrigeration system and collects the response curve of the temperature value inside the refrigeration chamber. Taking the input time of the step control signal as the starting point and the time point at which the temperature value inside the refrigeration chamber reaches the set percentage of the steady-state value as the ending point, the time difference between the starting point and the ending point is taken as the thermal inertia compensation time. For a typical first-order thermodynamic system, the set percentage is 63.2%, which corresponds to the defined value of the time constant of the first-order system.

[0027] In this embodiment, the industrial internet platform obtains the maturity evolution curve from the edge computing gateway. The industrial internet platform performs a time shift on the maturity evolution curve using a preset thermal inertia compensation time, that is, shifts the entire maturity evolution curve to the left by the length of the thermal inertia compensation time, resulting in a shifted maturity evolution curve. The industrial internet platform obtains the optimal storage temperature corresponding to each maturity level for the product category from a preset category parameter table, maps the maturity level corresponding to each time point of the shifted maturity evolution curve to the optimal storage temperature corresponding to that maturity level, and obtains the basic temperature setting curve. The industrial internet platform subtracts the thermal inertia advance compensation amount from the temperature value at each time point of the basic temperature setting curve, and outputs the advance temperature setting curve. The thermal inertia advance compensation amount is calculated using an empirical formula. The input parameters of the empirical formula include the thermal inertia compensation time, the rated cooling power of the refrigeration unit, and the heat capacity of the storage body.

[0028] In this embodiment, the thermal inertia of the refrigeration system refers to the physical characteristic that the temperature inside the cold storage cannot immediately respond to changes after the refrigeration system receives a control command, resulting in a delay. The thermal inertia of the refrigeration system is jointly determined by the heat storage characteristics of the cold storage enclosure structure, the start-stop response time of the refrigeration unit, the heat exchange efficiency of the air cooler, and the air circulation rate inside the cold storage. Thermal inertia manifests as a measurable and quantifiable time delay between the issuance of the control command and the actual change in the temperature inside the cold storage. This time delay is the thermal inertia compensation time.

[0029] In this embodiment, the maturity level is a dimensionless numerical value used to quantify the maturity of goods. The maturity level ranges from zero to one, where zero represents that the goods are completely immature and one represents that the goods are completely mature. For climacteric fruits, there is a non-linear positive correlation between the maturity level and the ethylene concentration value. The higher the ethylene concentration value, the higher the maturity level. The rate of change of the maturity level is positively correlated with the difference between the current maturity level and the maximum maturity level and the ethylene concentration value. The initial maturity level and maturity evolution rate constant of different goods categories are stored through a preset category parameter table.

[0030] In this embodiment, the advanced temperature setting curve is a sequence of temperature setpoints that are ahead of the maturity level setting to compensate for the thermal inertia of the refrigeration system. The advanced temperature setting curve is a curve with time as the horizontal axis and temperature setpoint as the vertical axis. Each point on the curve represents the temperature target value that should be set at the current time point. Unlike the traditional method of setting the temperature directly based on the current maturity level, the advanced temperature setting curve sets the temperature target in advance based on the prediction of the future change trend of the maturity level, so that the temperature target is ahead of the actual change of the maturity level of the goods. The time difference between the advanced temperature setting curve and the maturity evolution curve is equal to the thermal inertia compensation time, thereby utilizing the time window of the refrigeration system response delay to synchronize the actual cooling effect of the refrigeration system with the change of the maturity level of the goods.

[0031] In this embodiment, the edge computing gateway obtains the advanced temperature setpoint curve from the industrial internet platform and the real-time temperature value inside the warehouse from the temperature sensor. In each control cycle, the edge computing gateway calculates the deviation between the temperature setpoint value of the advanced temperature setpoint curve at the current moment and the current temperature value inside the warehouse, and inputs the deviation into the proportional-integral-derivative controller. The proportional-integral-derivative controller outputs the first control component. The edge computing gateway calculates the maturity change rate, which is obtained by the slope of the tangent line of the maturity evolution curve at the current moment. The maturity change rate is multiplied by a preset maturity change rate weighting coefficient to obtain the second control component. The edge computing gateway superimposes the first control component and the second control component and multiplies it by the thermal inertia advance compensation coefficient to output the advance control command sequence. The advance control command sequence is a sequence with time as the horizontal axis and control command value as the vertical axis. The control command value corresponds to the operating frequency or start / stop status of the refrigeration unit.

[0032] In this embodiment, the multi-unit collaborative controller is connected to the edge computing gateway. The multi-unit collaborative controller receives the advanced control command sequence output by the edge computing gateway. The multi-unit collaborative controller parses the control command value corresponding to each time point in the advanced control command sequence. The control command value includes the operating frequency setpoint and start / stop status setpoint of each chiller unit. The multi-unit collaborative controller allocates the total cooling capacity demand to each chiller unit according to the current operating status, cumulative operating time and energy efficiency ratio of each chiller unit. The allocation principle is to prioritize the operation of chiller units with high energy efficiency ratios and balance the cumulative operating time of each unit. The multi-unit collaborative controller sends the allocated operating frequency and start / stop status to the inverters and contactors of each chiller unit. Each chiller unit performs operating frequency adjustment or start / stop switching according to the received instructions.

[0033] Furthermore, when constructing the maturity evolution curve, the edge computing gateway uses the maturity level as the initial condition and the ethylene concentration value as the basis for the change in the maturity evolution rate. It outputs the maturity evolution curve by iteratively solving the maturity evolution differential equation. The rate of change of maturity level in the maturity evolution differential equation is positively correlated with the difference between the maturity level and the maximum maturity level, as well as the ethylene concentration.

[0034] In this embodiment, when constructing the maturity evolution curve, the edge computing gateway uses maturity level as the initial condition and ethylene concentration as the basis for the rate of change in maturity evolution. It iteratively solves the maturity evolution differential equation and outputs the maturity evolution curve. The edge computing gateway first reads the product category identifier using an RFID reader and obtains the initial maturity level of the category from a preset category parameter table. The initial maturity level ranges from 0 to 1; for bananas, the initial maturity level is 0.2, and for apples, it is 0.3. The edge computing gateway uses the initial maturity level as the initial condition for the maturity evolution differential equation and uses the real-time collected ethylene concentration value sequence as the external input to the differential equation. The maturity evolution differential equation adopts an ordinary differential equation based on the logistic growth model. This differential equation describes the relationship between the rate of change of maturity level over time, the difference between the current maturity level and the maximum maturity level, and the ethylene concentration value. The edge computing gateway uses the Euler method to numerically solve the maturity evolution differential equation. The Euler method discretizes continuous time into equally spaced time steps. Within each time step, the maturity level at the next moment is calculated based on the current maturity level and ethylene concentration value. This process is repeated to obtain a complete sequence of maturity levels changing over time, which is the maturity evolution curve.

[0035] In this embodiment, the rate of change of maturity level in the maturity evolution differential equation is positively correlated with the difference between the current maturity level and the maximum maturity level, as well as the ethylene concentration. The specific form of the maturity evolution differential equation is: the rate of change of maturity level equals the maturity evolution rate constant multiplied by the difference between the current maturity level and the maximum maturity level, and then multiplied by the ethylene concentration. The maturity evolution rate constant is stored in a preset category parameter table. For different product categories, the value of the maturity evolution rate constant is different; for bananas, it is 0.02 per hour, and for apples, it is 0.01 per hour. The difference between the current maturity level and the maximum maturity level equals the maximum maturity level minus the current maturity level, where the maximum maturity level is 1. This differential equation reflects the ripening pattern of climacteric fruits, i.e., the higher the ethylene concentration, the faster the maturity change; the closer the current maturity level is to the maximum maturity level, the slower the maturity change, which conforms to the logistic growth characteristics of actual biological growth processes.

[0036] Furthermore, before outputting the maturity evolution curve, the edge computing gateway first identifies whether the maturity level is at the critical point of the breathing leap. The edge computing gateway calculates the acceleration of the maturity evolution curve. When the acceleration of the maturity evolution curve remains positive within a preset time window and the growth rate of the ethylene concentration exceeds a preset growth rate threshold, the maturity level is determined to be at the critical point of the breathing jump. When the edge computing gateway determines that it is at the critical point of the breathing transition, it adjusts the maturity evolution rate constant in the maturity evolution differential equation to a multiple of its original value and outputs the final maturity evolution curve.

[0037] In this embodiment, calculating the acceleration of the maturity evolution curve refers to taking the second derivative of the maturity evolution curve. The edge computing gateway first extracts a discrete sequence of maturity levels changing over time from the maturity evolution curve. It then performs a first-order difference operation on this discrete sequence to obtain a maturity change rate sequence, and further performs a first-order difference operation on this maturity change rate sequence to obtain a maturity change acceleration sequence. The sign of the maturity change acceleration indicates the direction of change in the maturity change rate; a positive acceleration indicates that the maturity change rate is accelerating, and a negative acceleration indicates that the maturity change rate is slowing down.

[0038] In this embodiment, the preset time window refers to a pre-defined continuous time period used to observe whether the trend of the maturity evolution curve remains stable. The value of the preset time window is determined according to the characteristics of the product category; for bananas, the preset time window is 30 minutes, and for apples, it is 40 minutes. The edge computing gateway continuously monitors the sign of the acceleration of maturity change within the preset time window.

[0039] In this embodiment, the growth rate of ethylene concentration refers to the relative change in ethylene concentration over a unit of time. The edge computing gateway calculates the first difference of the ethylene concentration value sequence to obtain the change in ethylene concentration value. This change in ethylene concentration value is then divided by the current ethylene concentration value, and finally by the time step to obtain the growth rate of ethylene concentration. The unit of the growth rate of ethylene concentration is per hour.

[0040] In this embodiment, the preset growth rate threshold refers to a pre-set critical value for the growth rate of ethylene concentration, used to determine whether the ethylene concentration is in a phase of rapid increase. The value of the preset growth rate threshold is determined according to the characteristics of the product category. For bananas, the preset growth rate threshold is 10% per hour, and for apples, it is 8% per hour. When the growth rate of the ethylene concentration exceeds the preset growth rate threshold, it indicates that the product has entered the critical stage of a rapid increase in ethylene concentration.

[0041] In this embodiment, the dual criteria for determining whether the maturity level is at the respiratory judder threshold are that the acceleration of the maturity evolution curve remains positive within a preset time window and the growth rate of the ethylene concentration exceeds a preset growth rate threshold. The edge computing gateway first determines whether the acceleration of maturity change remains greater than zero within the preset time window. If so, it further determines whether the growth rate of the ethylene concentration exceeds the preset growth rate threshold. If both conditions are met simultaneously, the current maturity level is determined to be at the respiratory judder threshold. This dual criterion effectively avoids misjudgments caused by sensor noise or short-term fluctuations.

[0042] In this embodiment, the multiple of the original value refers to a scaling factor that amplifies the maturity evolution rate constant. When the edge computing gateway determines that it is at the respiratory climax threshold, it multiplies the maturity evolution rate constant in the maturity evolution differential equation by this multiple. For banana and apple varieties, this multiple is 2. This adjustment reflects the physiological characteristic that the maturity evolution rate significantly accelerates after the respiratory climax threshold.

[0043] In this embodiment, the maturity evolution rate constant in the maturity evolution differential equation is adjusted to a multiple of its original value, and the final maturity evolution curve is output. After determining that it is at the respiratory hopping threshold, the edge computing gateway multiplies the maturity evolution rate constant in the maturity evolution differential equation by a multiple to obtain the adjusted maturity evolution rate constant. Using the current maturity level as a new initial condition, the maturity evolution differential equation is resolved using the adjusted maturity evolution rate constant, and the corrected maturity evolution curve is output starting from the respiratory hopping threshold. This corrected maturity evolution curve more accurately reflects the actual situation of rapid maturity changes after the respiratory hopping threshold.

[0044] Furthermore, when constructing the advanced temperature setting curve, the industrial internet platform performs a time shift on the maturity evolution curve using a preset thermal inertia compensation time. The maturity level at the corresponding time point after the shift is mapped to the optimal storage temperature. After subtracting the thermal inertia advanced compensation amount from the optimal storage temperature, the advanced temperature setting curve is output.

[0045] In this embodiment, the preset thermal inertia compensation time refers to the time delay required from the issuance of the refrigeration system control command to the actual temperature response change inside the cold storage, which is predetermined by the system identification method. During the platform initialization phase, the edge computing gateway inputs a step control signal to the refrigeration system. This step control signal is the command for the refrigeration unit to switch directly from the off state to rated frequency operation. The edge computing gateway synchronously starts a timer and continuously collects the temperature value inside the cold storage. The time difference between the input time of the step control signal and the time when the temperature inside the cold storage reaches 63.2% of its steady-state value is used as the thermal inertia compensation time. For a typical medium-sized cold storage, the thermal inertia compensation time is between 30 and 60 minutes.

[0046] In this embodiment, the optimal storage temperature refers to the storage temperature that maintains the best quality for a specific product category at a specific maturity level. The industrial internet platform stores a preset product category parameter table, which records the optimal storage temperature corresponding to each maturity level for different product categories. For bananas, maturity level 0.2 corresponds to an optimal storage temperature of 13 degrees Celsius, maturity level 0.5 corresponds to an optimal storage temperature of 15 degrees Celsius, and maturity level 0.8 corresponds to an optimal storage temperature of 17 degrees Celsius. For apples, maturity level 0.2 corresponds to an optimal storage temperature of 1 degree Celsius, maturity level 0.5 corresponds to an optimal storage temperature of 2 degrees Celsius, and maturity level 0.8 corresponds to an optimal storage temperature of 3 degrees Celsius.

[0047] In this embodiment, the thermal inertia advance compensation amount refers to the temperature compensation value set in advance to offset the influence of the thermal inertia of the refrigeration system. The industrial internet platform calculates the thermal inertia advance compensation amount using an empirical formula. The empirical formula is: thermal inertia advance compensation amount equals thermal inertia compensation time multiplied by the rated refrigeration power of the refrigeration unit divided by the heat capacity of the cold storage, and then multiplied by an empirical correction factor, which is between 0.5 and 1.5. For a typical medium-sized cold storage, the thermal inertia advance compensation amount is between 1 and 2 degrees Celsius.

[0048] In this embodiment, when constructing the advanced temperature setting curve, the maturity evolution curve is shifted over time by a preset thermal inertia compensation time. The maturity level at the corresponding time point after the shift is mapped to the optimal storage temperature. The advanced temperature setting curve is then output after subtracting the thermal inertia advanced compensation amount from the optimal storage temperature. The industrial internet platform first obtains the maturity evolution curve from the edge computing gateway and shifts it over time by a preset thermal inertia compensation time. That is, the maturity evolution curve is shifted to the left by the length of the thermal inertia compensation time, resulting in the shifted maturity evolution curve. The maturity level corresponding to each time point on the shifted maturity evolution curve represents the maturity level that will be reached after the thermal inertia compensation time. The industrial internet platform queries the optimal storage temperature corresponding to each maturity level for the product category from a preset category parameter table. The maturity level corresponding to each time point on the shifted maturity evolution curve is mapped to the optimal storage temperature corresponding to that maturity level, resulting in the basic temperature setting curve. The industrial internet platform subtracts the thermal inertia advance compensation from the temperature value at each time point of the basic temperature setpoint curve, outputting an advanced temperature setpoint curve. Through the dual adjustment of time shift and advance compensation, the advanced temperature setpoint curve enables the cooling strategy to anticipate quality changes and compensate for thermal inertia in advance.

[0049] Furthermore, the thermal inertia compensation time is determined using a system identification method; During the platform initialization phase, the edge computing gateway inputs a step control signal to the refrigeration system and collects the response curve of the temperature value inside the refrigeration chamber. Taking the input time of the step control signal as the starting point and the time point when the temperature value inside the refrigeration chamber reaches the set percentage of the steady-state value as the ending point, the time difference between the starting point and the ending point is used as the thermal inertia compensation time.

[0050] In this embodiment, the platform initialization phase refers to the initial operation phase after the intelligent cold chain integrated management platform is first installed and deployed or after the cold storage system undergoes major maintenance. During the platform initialization phase, no goods are stored in the cold storage. The edge computing gateway executes the system identification program to calibrate the thermal inertia parameters of the refrigeration system. After the platform initialization phase is completed, the platform enters the normal operation phase, during which the calibration of thermal inertia compensation time is no longer repeated.

[0051] In this embodiment, the step control signal refers to a test signal indicating a sudden change in control command from one steady-state state to another. The step control signal sent by the edge computing gateway to the cooling system is a command that directly switches the cooling unit from the off state to the rated frequency operation. This command causes the cooling unit to jump from zero operating state to full-load operating state, and the amplitude of the step control signal is the rated operating frequency of the cooling unit.

[0052] In this embodiment, the response curve of the internal temperature value refers to the curve of the internal temperature of the cold storage changing over time under the excitation of a step control signal. The edge computing gateway starts a timer simultaneously with the issuance of the step control signal, continuously collecting the internal temperature value at a sampling frequency of once every 10 seconds. The temperature change process is recorded with time on the horizontal axis and the internal temperature value on the vertical axis, forming the response curve of the internal temperature value. The response curve starts from an initial steady-state temperature value, gradually decreases after a delay, and eventually tends towards a new steady-state temperature value.

[0053] In this embodiment, the steady-state value refers to the stable temperature value reached by the system after a sufficiently long period of time under the excitation of a step control signal. During the process of collecting the temperature response curves in the database, the edge computing gateway continuously calculates the temperature change at adjacent sampling points. When the temperature change at five consecutive sampling points is less than 0.1 degrees Celsius, the system is determined to have entered a steady state, and the temperature value of the last sampling point is taken as the steady-state value.

[0054] In this embodiment, the set ratio refers to the percentage threshold for determining the temperature change during thermal inertia compensation. The set ratio is set to 63.2%, a value derived from theoretical analysis of first-order systems. For a first-order thermodynamic system, the time required from a step input to the output reaching 63.2% of the steady-state value is the system's time constant, reflecting the magnitude of the system's inertia. The edge computing gateway multiplies the steady-state value by the set ratio to obtain the target temperature value, with the point in time when the temperature inside the storage reaches the target temperature value as the endpoint.

[0055] In this embodiment, during the platform initialization phase, a step control signal is input to the refrigeration system, and the response curve of the temperature value inside the cold storage is collected. The starting point is the time point when the step control signal is input, and the ending point is the time point when the temperature inside the cold storage reaches a set percentage of the steady-state value. The time difference between the starting and ending points is used as the thermal inertia compensation time. During the platform initialization phase, the edge computing gateway first ensures that there are no goods inside the cold storage and that the temperature inside is stable, recording the initial temperature value as the initial steady-state temperature value. The edge computing gateway sends a step control signal to the refrigeration system, instructing the refrigeration unit to switch from the off state to rated frequency operation. Simultaneously, a timer is started, and the input time point of the step control signal is recorded as the starting point. The edge computing gateway continuously collects the temperature value inside the cold storage at a sampling frequency of once every 10 seconds. When the temperature value inside the cold storage decreases from the initial steady-state temperature value to 63.2% of the difference between the initial and final steady-state temperatures, the current time point is recorded as the ending point. The edge computing gateway calculates the time difference between the ending and starting points and stores this time difference as the thermal inertia compensation time in the non-volatile memory of the edge computing gateway for use during normal platform operation.

[0056] Furthermore, when the edge computing gateway generates the advance control command sequence, it inputs the deviation between the advance temperature setpoint and the temperature value in the warehouse into the proportional-integral-derivative controller to obtain the first control component, multiplies the maturity change rate by a preset maturity change rate weighting coefficient to obtain the second control component, and superimposes the first control component and the second control component and multiplies it by the thermal inertia advance compensation coefficient to output the advance control command sequence.

[0057] In this embodiment, the advanced temperature setpoint refers to the temperature target value corresponding to the advanced temperature setpoint curve at the current moment. The edge computing gateway reads the temperature setpoint corresponding to the current time point from the advanced temperature setpoint curve in each control cycle. This temperature setpoint has been adjusted for time shifting and thermal inertia advance compensation, and is used to compensate for the thermal inertia of the refrigeration system, thus exceeding the temperature target set for the maturity level.

[0058] In this embodiment, the proportional-integral-derivative (PID) controller is a feedback controller that controls based on a combination of proportional, integral, and derivative parameters of the deviation. The PID controller receives the deviation between the leading temperature setpoint and the warehouse temperature as input. The proportional component outputs a control quantity based on the current deviation magnitude, the integral component outputs a control quantity based on the accumulated deviation, and the derivative component outputs a control quantity based on the rate of change of the deviation. The weighted sum of the three outputs yields the first control component. The proportional, integral, and derivative coefficients of the PID controller are pre-tuned through on-site testing.

[0059] In this embodiment, the first control component is the feedback control quantity output by the proportional-integral-derivative controller. The magnitude and direction of the first control component depend on the deviation between the advanced temperature setpoint and the temperature inside the refrigeration unit. When the temperature inside the refrigeration unit is higher than the advanced temperature setpoint, the first control component is positive, instructing the refrigeration unit to increase the cooling capacity. When the temperature inside the refrigeration unit is lower than the advanced temperature setpoint, the first control component is negative, instructing the refrigeration unit to decrease the cooling capacity. The absolute value of the first control component is positively correlated with the magnitude of the deviation.

[0060] In this embodiment, the maturity change rate refers to the speed at which the maturity level changes over time. The edge computing gateway extracts the maturity level at the current moment and the maturity level at the previous moment from the maturity evolution curve, calculates the difference between the maturity levels at the two moments, divides it by the time interval, and obtains the maturity change rate. The unit of the maturity change rate is per hour. A positive maturity change rate indicates that the maturity level is increasing, and the larger the positive value, the faster the maturity level is increasing.

[0061] In this embodiment, the preset maturity change rate weighting coefficient is a preset coefficient used to adjust the contribution of the maturity change rate in feedforward control. The preset maturity change rate weighting coefficient ranges from 0.1 to 0.5. For goods with a faster maturity change rate, such as bananas, the preset maturity change rate weighting coefficient is 0.4, and for goods with a slower maturity change rate, such as apples, the preset maturity change rate weighting coefficient is 0.2. This weighting coefficient is pre-tuned through on-site debugging or historical data fitting.

[0062] In this embodiment, the second control component is a feedforward control quantity obtained by weighting the maturity change rate. The edge computing gateway multiplies the current maturity change rate by a preset maturity change rate weighting coefficient to obtain the second control component. The introduction of the second control component enables the control command to respond in advance to the maturity change trend, increasing the cooling capacity in advance when the maturity changes rapidly, thus achieving feedforward compensation.

[0063] In this embodiment, the thermal inertia lead compensation coefficient is a preset coefficient used to adjust the overall amplitude of the lead control command. The value of the thermal inertia lead compensation coefficient ranges from 0.8 to 1.2. When the thermal inertia of the refrigeration system is large, the thermal inertia lead compensation coefficient is greater than 1; when the thermal inertia of the refrigeration system is small, the thermal inertia lead compensation coefficient is less than 1. The thermal inertia lead compensation coefficient is determined through system identification experiments during the platform initialization phase, ensuring that the lead control command can fully compensate for the response delay caused by thermal inertia.

[0064] In this embodiment, when generating the advanced control command sequence, the deviation between the advanced temperature setpoint and the warehouse temperature is input into a proportional-integral-derivative (PID) controller to obtain a first control component. The maturity change rate is multiplied by a preset maturity change rate weighting coefficient to obtain a second control component. The first and second control components are superimposed and multiplied by a thermal inertia advanced compensation coefficient to output the advanced control command sequence. The edge computing gateway executes the following calculation steps in each control cycle: First, it reads the advanced temperature setpoint at the current time point from the advanced temperature setpoint curve and reads the current warehouse temperature from the temperature sensor, calculating the deviation between the advanced temperature setpoint and the warehouse temperature. The edge computing gateway inputs this deviation into the PID controller, which outputs the first control component. The edge computing gateway calculates the maturity change rate at the current moment from the maturity evolution curve and multiplies the maturity change rate by a preset maturity change rate weighting coefficient to obtain the second control component. The edge computing gateway adds the first and second control components to obtain the total control component and multiplies the total control component by the thermal inertia advanced compensation coefficient to obtain the final advanced control command. The edge computing gateway arranges the advance control commands calculated in each control cycle in chronological order to form an advance control command sequence. This sequence combines the stability of feedback control with the predictability of feedforward control, enabling the refrigeration system to respond to current temperature deviations and anticipate future changes in maturity.

[0065] Furthermore, it also includes: The frost compensation unit is used to acquire images of the evaporator surface and output frost thickness values, mapping the frost thickness values ​​to a heat exchange efficiency attenuation coefficient. The edge computing gateway receives the heat exchange efficiency attenuation coefficient, multiplies it by the advance control command sequence, and corrects the advance control command sequence.

[0066] In this embodiment, acquiring evaporator surface images refers to obtaining visual data on the frost distribution on the evaporator surface through an image acquisition unit. The frost compensation unit includes an image acquisition module, which consists of an infrared thermal imaging camera and a visible light camera. Both the infrared thermal imaging camera and the visible light camera are installed directly in front of the evaporator inside the cold storage, with their lenses aimed at the evaporator surface. The infrared thermal imaging camera acquires a temperature distribution image of the evaporator surface, showing that the temperature in the frost-covered area is significantly lower than the temperature in the frost-free area. The visible light camera acquires a visible light image of the evaporator surface, showing that the frost layer appears as a white or semi-transparent covering. The infrared thermal imaging camera and the visible light camera synchronously acquire evaporator surface images at a sampling frequency of once every 30 minutes and transmit the acquired temperature distribution image and visible light image to the edge computing gateway.

[0067] In this embodiment, the output frost thickness value refers to the frost compensation unit converting the evaporator surface image into quantified frost thickness data using an image analysis algorithm. The frost compensation unit runs a dual-stream fusion network. The first stream of the dual-stream fusion network receives the visible light image and extracts the visible light feature vector, while the second stream receives the temperature distribution image and extracts the temperature feature vector. The fusion layer of the dual-stream fusion network performs pixel-level concatenation of the visible light and temperature feature vectors and inputs them into the fully connected layer. The fully connected layer outputs the frost thickness value of the evaporator surface. The unit of the frost thickness value is millimeters, and the value ranges from 0 to 5 millimeters. During the training phase, the dual-stream fusion network uses evaporator image pairs labeled with frost thickness values ​​as input and the predicted frost thickness value as output. Network parameters are optimized by minimizing the mean square error between the predicted and labeled frost thickness values.

[0068] In this embodiment, mapping the frost thickness value to the heat exchange efficiency attenuation coefficient means that the frost compensation unit converts the frost thickness value into a coefficient reflecting the degree of decrease in heat exchange efficiency through a preset mapping relationship. The frost compensation unit stores a mapping table between the frost thickness value and the heat exchange efficiency attenuation coefficient. The mapping table is obtained through experimental calibration. During the experiment, frost layers of different thicknesses are created on the evaporator surface, and the heat exchange efficiency at each frost thickness is measured. The attenuation coefficient is calculated based on the heat exchange efficiency in the frost-free state. A frost thickness of 0 mm corresponds to a heat exchange efficiency attenuation coefficient of 0, a frost thickness of 1 mm corresponds to a heat exchange efficiency attenuation coefficient of 0.1, a frost thickness of 2 mm corresponds to a heat exchange efficiency attenuation coefficient of 0.3, a frost thickness of 3 mm corresponds to a heat exchange efficiency attenuation coefficient of 0.5, a frost thickness of 4 mm corresponds to a heat exchange efficiency attenuation coefficient of 0.7, and a frost thickness of 5 mm or more corresponds to a heat exchange efficiency attenuation coefficient of 0.8.

[0069] In this embodiment, the heat exchange efficiency attenuation coefficient is a dimensionless coefficient reflecting the degree of decrease in heat exchange capacity of the evaporator due to frosting. The heat exchange efficiency attenuation coefficient ranges from 0 to 0.8, where 0 represents no decrease in evaporator heat exchange efficiency in a frost-free state, and 0.8 represents an 80% decrease in evaporator heat exchange efficiency in a severely frosted state. The larger the heat exchange efficiency attenuation coefficient, the more severe the evaporator frosting, and the stronger the cooling command is required to achieve the same cooling effect.

[0070] In this embodiment, multiplying the heat exchange efficiency attenuation coefficient by the advance control command sequence and correcting the advance control command sequence refers to the edge computing gateway introducing frosting compensation when generating the advance control command sequence. The edge computing gateway receives the heat exchange efficiency attenuation coefficient output by the frosting compensation unit, calculates the total control component, divides the total control component by 1 and subtracts the difference from the heat exchange efficiency attenuation coefficient to obtain the corrected total control component, and then multiplies the corrected total control component by the thermal inertia advance compensation coefficient to output the advance control command. The correction formula is that the corrected total control component equals the total control component divided by 1 and the heat exchange efficiency attenuation coefficient. This correction logic enables the advance control command to increase accordingly with the severity of frosting, compensating for the decrease in heat exchange efficiency caused by frosting, and ensuring that the actual cooling effect of the refrigeration system is consistent with expectations.

[0071] Furthermore, it also includes: The blockchain evidence storage unit is used to package ethylene concentration values, storage temperature values, maturity evolution curves, advanced temperature setting curves, and advanced control command sequences into evidence storage data blocks, and then upload the evidence storage data blocks to the blockchain network after digitally signing them. Before constructing the maturity evolution curve, the edge computing gateway obtains historical evidence data blocks stored in the blockchain evidence storage unit, extracts the historical ethylene concentration value sequence with the same product category identifier and the corresponding historical maturity level sequence from the historical evidence data block, and refits the maturity evolution rate constant in the maturity evolution differential equation with the historical ethylene concentration value sequence as input and the historical maturity level sequence as output.

[0072] In this embodiment, packaging ethylene concentration, storage temperature, maturity evolution curve, advanced temperature setting curve, and advanced control command sequence into a data block for evidence storage, and then digitally signing and uploading the data block to the blockchain network, refers to the blockchain evidence storage unit's trusted evidence storage of key data throughout the entire cold chain operation process. At the end of each preset evidence storage period, the blockchain evidence storage unit collects the ethylene concentration and storage temperature sequences from the edge computing gateway, the maturity evolution and advanced temperature setting curves generated by the industrial internet platform, and the advanced control command sequence generated by the edge computing gateway. This data is then packaged into a single data block according to a preset data format. The blockchain evidence storage unit uses the trusted execution environment built into the edge computing gateway to generate a local timestamp, uses the edge computing gateway's device digital certificate to perform an elliptic curve digital signature on the data block, and uploads the signed data block to the blockchain network through an edge-cloud collaborative channel. The blockchain network consists of multiple consortium blockchain nodes. Each node performs consensus verification on the received data block and writes it into the blockchain ledger. Once written, the data block is immutable and traceable.

[0073] In this embodiment, before constructing the maturity evolution curve, historical evidence data blocks stored in the blockchain evidence storage unit are obtained. From these historical evidence data blocks, historical ethylene concentration value sequences and corresponding historical maturity level sequences with the same product category identifier are extracted. Using the historical ethylene concentration value sequences as input and the historical maturity level sequences as output, the maturity evolution rate constant in the maturity evolution differential equation is refitted. This rate constant refers to the edge computing gateway's self-optimization of model parameters using historical data stored in the blockchain. Before constructing the maturity evolution curve for the current batch, the edge computing gateway first sends a query request to the blockchain evidence storage unit. The query request includes the current product category identifier and a preset time range, with the preset time range being the past 90 days. The blockchain evidence storage unit retrieves evidence data blocks that meet the query conditions from the blockchain network and returns them to the edge computing gateway. The edge computing gateway extracts historical ethylene concentration value sequences and corresponding historical maturity level sequences with the same product category identifier from the returned evidence data blocks. The historical maturity level sequences originate from the maturity evolution curves in the evidence data blocks. The edge computing gateway uses historical ethylene concentration sequences as input features and historical maturity level sequences as output labels. It refits the maturity evolution rate constant in the maturity evolution differential equation using the least squares method. The goal of the least squares method is to minimize the sum of squared residuals between the predicted maturity level and the historical maturity level. The edge computing gateway then uses the refitted maturity evolution rate constant to construct the maturity evolution curve for the current batch, achieving adaptive optimization of model parameters based on historical operating data.

[0074] In this embodiment, the maturity evolution rate constant is a key parameter in the maturity evolution differential equation, reflecting the maturation speed of a product category under specific conditions. The maturity evolution rate constant stores initial values ​​in a preset category parameter table; for bananas, the initial maturity evolution rate constant is 0.02 per hour, and for apples, it is 0.01 per hour. During platform operation, the maturity evolution rate constant is continuously refitted and updated using historical data stored on the blockchain. When the accumulated historical data for the same product category reaches more than 10 sets, the edge computing gateway initiates a parameter self-optimization process, replacing the original value with the refitted maturity evolution rate constant. This self-optimization mechanism enables the maturity evolution differential equation to adapt to the actual maturation patterns of different batches of goods and different seasonal environments, improving the accuracy of maturity prediction.

[0075] Furthermore, it also includes: The digital twin mapping unit is used to receive ethylene concentration values, storage temperature values, maturity evolution curves, advanced temperature setting curves, and advanced control command sequences, construct a three-dimensional geometric model of the cold chain space, and map the maturity evolution curves, advanced temperature setting curves, advanced control command sequences, and storage temperature values ​​onto the three-dimensional geometric model to generate a dynamic visualization interface. The dynamic visualization interface displays the advanced-lag relationship between the maturity evolution curve and the storage temperature response curve.

[0076] In this embodiment, constructing a three-dimensional geometric model of the cold chain space refers to the digital twin mapping unit creating a virtual three-dimensional spatial model based on the actual physical dimensions and layout of the cold storage. The digital twin mapping unit acquires architectural design drawings of the cold storage or collects point cloud data of the cold storage interior using a 3D laser scanner, extracting the length, width, and height dimensions of the cold storage, as well as the spatial coordinates of the refrigeration unit locations, air cooler locations, shelf layout, temperature sensor installation locations, and ethylene sensor installation locations. The digital twin mapping unit uses a 3D modeling engine to convert the above spatial data into a three-dimensional geometric model. The 3D geometric model includes physical objects such as walls, ceilings, floors, refrigeration units, air coolers, shelves, temperature sensor nodes, and ethylene sensor nodes. Each physical object has accurate position coordinates and appearance attributes in the 3D geometric model.

[0077] In this embodiment, the maturity evolution curve, advanced temperature setting curve, advanced control command sequence, and warehouse temperature value are mapped onto a three-dimensional geometric model to generate a dynamic visualization interface. The dynamic visualization interface displays the lead-lag relationship between the maturity evolution curve and the warehouse temperature response curve, meaning the digital twin mapping unit overlays real-time operational data onto the three-dimensional geometric model to form a visual display. The digital twin mapping unit obtains real-time ethylene concentration values, warehouse temperature values, maturity evolution curves, and advanced control command sequences from the edge computing gateway via an edge-cloud collaborative channel, and obtains the advanced temperature setting curve from the industrial internet platform. The digital twin mapping unit adds real-time data tags to each temperature sensor node and ethylene sensor node in the three-dimensional geometric model, and the data tags display the currently collected values. The digital twin mapping unit generates an independent curve display panel above or to the side of the three-dimensional geometric model. The curve display panel plots the maturity evolution curve, advanced temperature setting curve, and actual response curves of warehouse temperature values ​​changing over time on the same coordinate system, with time as the horizontal axis and value as the vertical axis. The dynamic visualization interface updates curves and label data once per second, allowing operators to intuitively observe the cold chain operation status.

[0078] In this embodiment, the lead-lag relationship between the maturity evolution curve and the warehouse temperature response curve refers to the time offset between the changing trend of the maturity evolution curve and the actual warehouse temperature response curve. The maturity evolution curve reflects the changing trend of the goods' maturity level over time, the advanced temperature setting curve is generated based on the maturity evolution curve after time shifting, and the warehouse temperature response curve reflects the actual changes in warehouse temperature over time. In the dynamic visualization interface, the operator can observe that the change time of the advanced temperature setting curve is earlier than the change time of the warehouse temperature response curve, and the time difference between the two is equal to the thermal inertia compensation time. When the maturity evolution curve begins to accelerate upward, the advanced temperature setting curve simultaneously declines ahead of time, and the warehouse temperature response curve begins to decline after the thermal inertia compensation time, forming a time match between the rapid increase phase of the maturity level and the early decline phase of the warehouse temperature. The dynamic visualization interface distinguishes three curves by color: the maturity evolution curve is drawn in green, the advanced temperature setting curve is drawn in blue, and the temperature response curve inside the warehouse is drawn in red. The thermal inertia compensation time interval is marked on the curve graph, clearly showing the effect of the advanced control strategy in matching the refrigeration system response lag with the quality change time window.

[0079] This invention provides an embodiment of an integrated intelligent cold chain management method, applicable to any of the above-mentioned integrated intelligent cold chain management platforms, comprising: Real-time collection of ethylene concentration and temperature values ​​in the cold chain environment, and reading of cargo category identification; Based on the change of ethylene concentration over time, a maturity evolution curve is constructed, which characterizes the trend of maturity level change over time. Based on the maturity evolution curve and the preset thermal inertia compensation time, an advanced temperature setting curve is constructed. The advanced temperature setting curve is a sequence of temperature setting values ​​that are ahead of the maturity level setting to compensate for the thermal inertia of the refrigeration system. Based on the advanced temperature setpoint curve and the temperature value inside the refrigeration unit, an advanced control command sequence is generated. The advanced control command sequence is a control command generated ahead of the temperature deviation inside the refrigeration unit to compensate for the thermal inertia of the refrigeration system. The operation of each refrigeration unit is controlled according to the sequence of advance control commands.

[0080] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A smart cold chain integrated management platform, characterized in that, include: Ethylene sensor, used to collect ethylene concentration values ​​in the cold chain environment in real time; Temperature sensors are used to collect real-time temperature values ​​inside the cold chain environment. Radio frequency identification (RFID) readers are used to read product category identifiers. The edge computing gateway is connected to the ethylene sensor, temperature sensor, and RFID reader, respectively. The edge computing gateway is configured as follows: Based on the change of ethylene concentration over time, a maturity evolution curve is constructed, which characterizes the trend of maturity level change over time. The industrial internet platform connects with the edge computing gateway through end-cloud collaboration. The industrial internet platform is used to construct an advanced temperature setting curve based on the maturity evolution curve and the preset thermal inertia compensation time. The advanced temperature setting curve is a sequence of temperature setting values ​​that are ahead of the maturity level setting to compensate for the thermal inertia of the refrigeration system. The edge computing gateway is also used to generate a sequence of advanced control commands based on the advanced temperature setting curve and the temperature value inside the refrigeration system. The sequence of advanced control commands is a set of control commands generated ahead of the temperature deviation inside the refrigeration system to compensate for the thermal inertia of the refrigeration system. The multi-unit collaborative controller, connected to the edge computing gateway, is used to control the operation of each chiller unit according to the sequence of advanced control commands.

2. The intelligent cold chain integrated management platform according to claim 1, characterized in that, When constructing the maturity evolution curve, the edge computing gateway uses the maturity level as the initial condition and the ethylene concentration value as the basis for the change in the maturity evolution rate. It outputs the maturity evolution curve by iteratively solving the maturity evolution differential equation. The rate of change of maturity level in the maturity evolution differential equation is positively correlated with the difference between the maturity level and the maximum maturity level, as well as the ethylene concentration.

3. The intelligent cold chain integrated management platform according to claim 2, characterized in that, Before outputting the maturity evolution curve, the edge computing gateway first identifies whether the maturity level is at the critical point of the breathing leap. The edge computing gateway calculates the acceleration of the maturity evolution curve. When the acceleration of the maturity evolution curve remains positive within a preset time window and the growth rate of the ethylene concentration exceeds a preset growth rate threshold, the maturity level is determined to be at the critical point of the breathing jump. When the edge computing gateway determines that it is at the critical point of the breathing transition, it adjusts the maturity evolution rate constant in the maturity evolution differential equation to a multiple of its original value and outputs the final maturity evolution curve.

4. The intelligent cold chain integrated management platform according to claim 1, characterized in that, When constructing the advanced temperature setting curve, the industrial internet platform performs a time shift on the maturity evolution curve using a preset thermal inertia compensation time. The maturity level at the corresponding time point after the shift is mapped to the optimal storage temperature. The advanced temperature setting curve is then output after subtracting the thermal inertia advanced compensation amount from the optimal storage temperature.

5. The intelligent cold chain integrated management platform according to claim 4, characterized in that, The thermal inertia compensation time is determined using a system identification method; During the platform initialization phase, the edge computing gateway inputs a step control signal to the refrigeration system and collects the response curve of the temperature value inside the refrigeration chamber. Taking the input time of the step control signal as the starting point and the time point when the temperature value inside the refrigeration chamber reaches the set percentage of the steady-state value as the ending point, the time difference between the starting point and the ending point is used as the thermal inertia compensation time.

6. The intelligent cold chain integrated management platform according to claim 1, characterized in that, When the edge computing gateway generates a sequence of advance control commands, it inputs the deviation between the advance temperature setpoint and the temperature in the warehouse into the proportional-integral-derivative controller to obtain the first control component, multiplies the maturity change rate by a preset maturity change rate weighting coefficient to obtain the second control component, and then superimposes the first control component and the second control component and multiplies it by the thermal inertia advance compensation coefficient to output the sequence of advance control commands.

7. The intelligent cold chain integrated management platform according to claim 1, characterized in that, Also includes: The frost compensation unit is used to acquire images of the evaporator surface and output frost thickness values, mapping the frost thickness values ​​to a heat exchange efficiency attenuation coefficient. The edge computing gateway receives the heat exchange efficiency attenuation coefficient, multiplies it by the advance control command sequence, and corrects the advance control command sequence.

8. The intelligent cold chain integrated management platform according to claim 1, characterized in that, Also includes: The blockchain evidence storage unit is used to package ethylene concentration values, storage temperature values, maturity evolution curves, advanced temperature setting curves, and advanced control command sequences into evidence storage data blocks, and then upload the evidence storage data blocks to the blockchain network after digitally signing them. Before constructing the maturity evolution curve, the edge computing gateway obtains historical evidence data blocks stored in the blockchain evidence storage unit, extracts the historical ethylene concentration value sequence with the same product category identifier and the corresponding historical maturity level sequence from the historical evidence data block, and refits the maturity evolution rate constant in the maturity evolution differential equation with the historical ethylene concentration value sequence as input and the historical maturity level sequence as output.

9. The intelligent cold chain integrated management platform according to claim 1, characterized in that, Also includes: The digital twin mapping unit is used to receive ethylene concentration values, storage temperature values, maturity evolution curves, advanced temperature setting curves, and advanced control command sequences, construct a three-dimensional geometric model of the cold chain space, and map the maturity evolution curves, advanced temperature setting curves, advanced control command sequences, and storage temperature values ​​onto the three-dimensional geometric model to generate a dynamic visualization interface. The dynamic visualization interface displays the advanced-lag relationship between the maturity evolution curve and the storage temperature response curve.

10. A smart cold chain integrated management method, applied to the smart cold chain integrated management platform according to any one of claims 1 to 9, characterized in that, include: Real-time collection of ethylene concentration and temperature values ​​in the cold chain environment, and reading of cargo category identification; Based on the change of ethylene concentration over time, a maturity evolution curve is constructed, which characterizes the trend of maturity level change over time. Based on the maturity evolution curve and the preset thermal inertia compensation time, an advanced temperature setting curve is constructed. The advanced temperature setting curve is a sequence of temperature setting values ​​that are ahead of the maturity level setting to compensate for the thermal inertia of the refrigeration system. Based on the advanced temperature setpoint curve and the temperature value inside the refrigeration unit, an advanced control command sequence is generated. The advanced control command sequence is a control command generated ahead of the temperature deviation inside the refrigeration unit to compensate for the thermal inertia of the refrigeration system. The operation of each refrigeration unit is controlled according to the sequence of advance control commands.

Images