Intelligent cold leakage prevention system for cold storage door

Abstract

The invention discloses an intelligent cold leakage prevention system for a cold storage door, and relates to the field of cold-chain logistics temperature control and energy conservation. According to the intelligent cold leakage prevention system for the cold storage door, different basic execution strategies are designed according to different areas of a door area, then corresponding cold leakage risk coefficient models are designed according to parameters of objects entering and exiting from the cold storage and environmental parameters when the cold storage door is opened, intelligent dynamic adjustment is conducted on the basic execution strategies, and a reinforcement learning optimization mechanism is designed; negative effects of energy consumption, temperature rise and cold leakage generated in the whole frequency conversion air curtain machine process are balanced by introducing a reward function, so that three important indexes of energy consumption, temperature rise and cold leakage are balanced, the energy consumption and the cold leakage are reasonably adjusted on the premise of ensuring that the temperature rise is within a reasonable index range, the comprehensive energy-saving effect is achieved, and the service life of the frequency conversion air curtain machine is prolonged. And meanwhile, zero-delay response in practical application can be realized, and object parameters and environment parameters are comprehensively introduced into the cold leakage risk coefficient model, so that complex practical scenes can be dealt with.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cold chain logistics temperature control and energy saving, in particular to an intelligent cold storage door anti-leakage cooling system. BACKGROUND

[0002] The current cold storage anti-leakage cooling technology mainly adopts fixed wind speed air curtain machine, basic infrared induction control and step adjustment air curtain.

[0003] And the fixed wind speed air curtain continuously outputs constant high-speed airflow during low-flow operation period or short door opening scene, resulting in a sharp increase in invalid energy consumption, and there is a problem of energy waste;

[0004] Although the infrared induction system can start and stop the air curtain through the door state, it cannot monitor the turbulence, air vortex and laminar disturbance caused by the pressure difference between the inside and outside air at the moment of opening the door in real time, resulting in that the air curtain intensity adjustment lags behind the airflow dynamic change, the leakage cooling risk is difficult to effectively inhibit, and there is a problem of response lag;

[0005] And the step adjustment technology is based on the step control mode of temperature difference threshold or preset time interval. Under complex working conditions such as multiple people or multiple vehicles entering and leaving in parallel, extreme weather causing temperature and humidity mutation, etc., the existing system cannot dynamically divide the high-leakage cooling risk area of the door area, and lacks differentiated air curtain coverage strategy according to the volume of moving objects, resulting in a 2-3 times increase in cold air loss rate compared with the regular scene. Especially when the external environment fluctuates violently, the system cannot adjust the air curtain intensity by real-time sensing airflow disturbance, nor can it optimize the airflow coverage range according to the object motion trajectory, ultimately causing the overall energy efficiency index to deviate seriously from the design requirement, seriously restricting the realization of the energy saving goal of the cold chain facility, and there is a problem of poor scene adaptability. SUMMARY

[0006] In view of the deficiencies of the prior art, the present application provides an intelligent cold storage door anti-leakage cooling system, which solves the problems of energy waste, response lag and poor scene adaptability of the existing cold storage anti-leakage cooling technology.

[0007] To achieve the above purpose, the present application realizes the following technical scheme: an intelligent cold storage door anti-leakage cooling system, comprising a data acquisition module, an intelligent analysis module and an execution module;

[0008] The data acquisition module acquires relevant key parameters during the opening process of the cold storage door;

[0009] Object parameters: moving speed v (m / s), volume V 物 (m 3 ), height from the ground h (m), number and trajectory;

[0010] Environmental parameters: internal and external temperature difference ΔT (℃), humidity difference ΔH (%), airflow speed U (m / s) and turbulent kinetic energy K;

[0011] The core technical logic of the intelligent analysis module includes multi-modal data fusion, dynamic segmented jet strategy and reinforcement learning optimization mechanism.

[0012] The multi-modal data fusion is to process the data obtained by the data acquisition module, calculate the object blocking air area S1 according to the volume V of the object, and the fixed opening area of the cold storage door is S2, and then design a calculation model of the cold leakage risk coefficient through data fusion.

[0013]

[0014] Wherein α, β, γ, δ are weight coefficients, and α+β+γ+δ=1;

[0015] The execution module is a variable frequency air curtain machine, and the basic execution strategy is

[0016] The top of the door area adopts a low-intensity strategy and is maintained at 10%-20% of the rated power to prevent hot air from penetrating;

[0017] The middle part of the door area adopts an adaptive speed regulation strategy and is maintained at 30%-70% of the rated power, and personnel and goods are transported;

[0018] The bottom of the door area adopts a high-intensity pulse and is maintained at 80%-100% of the rated power to cope with cold air sinking and forklift passing;

[0019] The dynamic segmented jet strategy is designed according to the cold leakage risk coefficient.

[0020]

[0021] Wherein P i Indicates the real-time power of the i-th section of the air curtain, G indicates the distance attenuation coefficient, and L indicates the distance between the object and the air curtain.

[0022] In this way, the variable frequency air curtain machine is controlled to dynamically adjust according to the cold leakage risk coefficient in each area.

[0023] Preferably, the data acquisition module includes an infrared sensor for distance measurement and a multi-parameter air flow module for air flow monitoring, the infrared sensor is arranged in an array on both sides of the cold storage door frame, and the multi-parameter air flow module integrates temperature sensor, humidity sensor, flow sensor and air pressure sensor.

[0024] The intelligent analysis module is an AI edge computer.

[0025] Preferably, the control execution process of the variable frequency air curtain machine is

[0026] S1: The cold storage door opens, and the infrared array scans the object's outline and trajectory, while the airflow module collects turbulence, temperature, humidity, and air pressure data;

[0027] S2, the AI ​​edge computer performs data fusion on the collected data and dynamically generates air curtain control instructions;

[0028] S3, the air curtain machine executes the corresponding airflow injection strategy for the door area according to the instruction;

[0029] S4. Real-time monitoring of the cooling leakage coefficient and judgment of the cooling leakage coefficient. When the cooling leakage coefficient does not exceed the threshold, the current optimization state is maintained. When the cooling leakage coefficient exceeds the threshold, a new air curtain control instruction is generated based on the collected data.

[0030] Preferably, in the step S4, when performing real-time monitoring of the cooling leakage coefficient, a cooling leakage evaluation model is introduced as follows;

[0031]

[0032] Where ρ represents the air density at the current temperature in the cold storage (kg / m 2 );c p represents the specific heat capacity of air (J / kg·k); A represents the effective leakage area of ​​the door area (m 2 ), which can be obtained by calculating the total area of ​​the door area minus the area blocked when the object enters and exits the door area; t represents the door opening time (s);

[0033] Based on the evaluation model, the cooling loss of the cold storage door during the door opening period is calculated in real time.

[0034] Preferably, in the control execution process of the variable frequency air curtain machine, by dynamically adjusting the control instructions of the air curtain machine, the algorithm and hardware are linked to achieve the effect of improving energy efficiency, and then by recording the average energy consumption performance in multiple cycles, the energy efficiency evaluation is obtained;

[0035]

[0036] Preferably, in the reinforcement learning optimization mechanism, a reward function is introduced to balance the negative effects of energy consumption, temperature rise and cooling leakage generated during the entire variable frequency air curtain process, which is specifically expressed as follows;

[0037] R=- ( ω1·Energy consumption + ω2·Temperature rise + ω3·Cooling leakage)

[0038] Among them, ω1+ω2+ω3=1, and the overall benefit performance of the dynamic segmented injection strategy is analyzed by the AI ​​edge computer, so that the respective weight distribution of ω1, ω2 and ω3 is updated and optimized every 24 hours.

[0039] Preferably, when calculating the cold leakage risk coefficient, when calculating the weight of the area of ​​the airflow blocked by the object, the cold storage door is divided into multiple sections in the height direction, including the bottom H1 of the door area, the middle H2 of the door area, and the top H3 of the door area. The risk score needs to be calculated separately for each section. At this time, the moving speed V, volume V 物 , height from the ground H 离地 , used to determine the position of the object in the door area, and thus calculate the proportion of each segment blocked by the object;

[0040]

[0041] At the bottom of the door area, when the height of the object from the ground is greater than H1, the occlusion ratio of this section is 0, and δ = 0;

[0042] In the middle of the gate area, when H 物 -H 离地 =H2, the occlusion ratio is 100%, and the δ weight should be increased at this time;

[0043] At the top of the goal area: When H 物 When ≤H1+H2, the object is not blocked, the occlusion ratio is 0, and δ=0.

[0044] Preferably, when calculating the risk score for each segment separately, other parameters including ΔT, ΔH and K are the same for the entire door area, and are all equivalently replaced and calculated according to the weight calculation method of the area of ​​airflow blocked by the object.

[0045] Preferably, the AI ​​edge computer processes sensor data in real time and generates control instructions by deploying a lightweight neural network model.

[0046] Preferably, in S2, when the AI ​​edge computer calculates the cooling leakage risk coefficient, it judges the top area and the middle area based on the cooling leakage risk coefficient. When the cooling leakage coefficient is lower than the dynamic threshold, the basic execution strategy is enabled. When the cooling leakage coefficient is higher than the dynamic threshold, the dynamic segmented injection strategy is enabled. The air curtain machine always uses pulse injection, especially for the door bottom area.

[0047] The present invention discloses an intelligent cold storage door anti-cold leakage system, which has the following beneficial effects:

[0048] 1. The intelligent cold storage door leak-proof cold system first divides the door area into the top, middle and bottom of the door area according to the application characteristics of the door area, applies different basic execution strategies to different areas, then designs a corresponding leak-proof cold risk coefficient model according to the object parameters and environmental parameters of the objects entering and leaving the cold storage when the cold storage door is opened, intelligently adjusts the basic execution strategy, further designs a reinforcement learning optimization mechanism on the basis of the intelligent analysis module, balances the negative effects of energy consumption, temperature rise and leak-proof cold generated in the whole variable frequency air curtain machine process by introducing a reward function, so as to balance the three important indicators of energy consumption, temperature rise and leak-proof cold, reasonably adjust energy consumption and leak-proof cold under the premise of ensuring that the temperature rise is within a reasonable index range, so as to achieve comprehensive energy-saving effect, and at the same time, it can realize zero delay response in actual application, and the leak-proof cold risk coefficient model can cope with complex actual scenes by comprehensively introducing object parameters and environmental parameters.

[0049] 2. The intelligent cold storage door leak-proof cold system designs an intelligent analysis module, normalizes the captured object parameters and environmental parameters, then designs a leak-proof cold risk coefficient calculation model through data fusion, and divides the door area into the top, middle and bottom of the door area according to the application characteristics of the door area, applies different basic execution strategies to different areas, then after capturing the door opening action, the leak-proof cold risk coefficient is calculated in real time by applying the leak-proof cold risk coefficient model with the introduction of each object parameter and environmental parameter, at this time, the leak-proof cold risk coefficient is further constrained on the basis of the execution strategy, realizing intelligent control of the variable frequency air curtain machine, so that the variable frequency air curtain machine is adaptively adjusted for each divided area of the door area, especially the pulse injection focuses on blocking the high-leakage cold area at the bottom of the door, thereby reducing the cold storage leakage. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0051] Figure 1 The function architecture diagram of the intelligent cold storage door leak-proof cold system of the present application;

[0052] Figure 2 The execution flow chart of the intelligent cold storage door leak-proof cold system of the present application;

[0053] Figure 3 The core technical logic diagram of the intelligent analysis module of the present application. DETAILED DESCRIPTION

[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0055] The embodiments of the present application provide an intelligent cold storage door anti-leakage system, thereby solving the problems of energy waste, delayed response and poor scene adaptability in existing cold storage anti-leakage technologies.

[0056] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0057] The embodiment of the present invention discloses an intelligent cold storage door cold leakage prevention system.

[0058] According to the attached Figure 1-3 As shown, it includes a data acquisition module, an intelligent analysis module and an execution module;

[0059] The data acquisition module obtains the key parameters related to the opening process of the cold storage door;

[0060] Object parameters: moving speed v (m / s), volume V 物 (m 3 ), height from the ground h (m), number and trajectory;

[0061] Environmental parameters: internal and external temperature difference ΔT (°C), humidity difference ΔH (%), air flow velocity U (m / s) and turbulent kinetic energy K;

[0062] The core technical logic of the intelligent analysis module includes multimodal data fusion, dynamic segmented injection strategy and reinforcement learning optimization mechanism;

[0063] Among them, multimodal data fusion processes the data obtained by the data acquisition module, calculates the airflow blocking area S1 of the object based on the volume V of the object, and the fixed opening area of ​​the cold storage door is S2. Then, the calculation model of the cold leakage risk coefficient is designed through data fusion:

[0064]

[0065] Among them, α, β, γ, and δ are weight coefficients, and α+β+γ+δ=1;

[0066] The execution module is a variable frequency air curtain machine, and its basic execution strategy is:

[0067] The top of the door area adopts a low-intensity strategy, maintaining at 1-20% of the rated power, to prevent hot air from seeping in;

[0068] The middle of the door area adopts an adaptive speed regulation strategy, maintaining at 3-70% of the rated power, for personnel passage and cargo handling;

[0069] The bottom of the door area adopts a high-intensity pulse, maintaining at 8-100% of the rated power, to cope with cold air sinking and forklift passage;

[0070] According to the cold leakage risk coefficient, a dynamic segmented injection strategy is designed;

[0071]

[0072] where P i represents the real-time power of the i-th section of the air curtain, G represents the distance attenuation coefficient, and L represents the distance between the object and the air curtain;

[0073] This achieves dynamic adjustment of the variable frequency air curtain machine in each area according to the cold leakage risk coefficient.

[0074] The following is a risk coefficient simulation calculation by assuming parameters, and the step-by-step calculation process is as follows:

[0075] If;

[0076] The inside and outside temperature difference ΔT = 40℃

[0077] The air flow speed in the cold storage u_in = 0.5m / s

[0078] The air flow speed outside the cold storage u_out = 2m / s

[0079] The turbulent kinetic energy k = 1.2m 2 / s 2

[0080] Object parameters: a forklift is passing through the cold storage door, with a shielding area of 1.5m 2 , the cold storage door area is 6m 2 , assuming the door width is 2m and the height is 3m;

[0081] At the same time, let T_max = 60℃, u_max = 10m / s, k_max = 5m 2 / s 2

[0082] Weight coefficients: α = 0.4, β = 0.3, γ = 0.2, δ = 0.1

[0083] Step 1, calculate the temperature difference term

[0084] Temperature difference term = α * (ΔT / T_max) = 0.4 * (40 / 60) = 0.2667;

[0085] Step two, calculate the air flow velocity difference term

[0086] Velocity difference = |u_in-u_out| = |0.5-2| = 1.5 m / s

[0087] Velocity difference term = β*(|u_in-u_out| / u_max) = 0.3*(1.5 / 10) = 0.045

[0088] Step three, calculate the turbulent kinetic energy term

[0089] Turbulent term = γ*(k / k_max) = 0.2*(1.2 / 5) = 0.048

[0090] Step four, calculate the object blocking area term

[0091] Blocking area ratio = object blocking area / door area = 1.5 / 6 = 0.25

[0092] Blocking term = δ*blocking area ratio = 0.1*0.25 = 0.025

[0093] Step five, calculate the total risk score

[0094] Risk score = 0.2667 + 0.045 + 0.048 + 0.025 = 0.3847

[0095] At the same time, according to the risk coefficient, the area can be divided;

[0096]

[0097] In this way, the variable frequency air curtain machine intelligently adjusts the power.

[0098] The data acquisition module includes an infrared sensor for distance measurement and a multi-parameter air flow module for air flow monitoring. The infrared sensor is arranged in an array on both sides of the cold storage door frame, and the multi-parameter air flow module integrates temperature sensor, humidity sensor, flow sensor and air pressure sensor;

[0099] The intelligent analysis module is an AI edge computer.

[0100] The control execution process of the variable frequency air curtain machine is;

[0101] S1, the cold storage door is opened, the infrared array scans the object contour and trajectory, and the air flow module collects turbulent flow, temperature and humidity, and air pressure data;

[0102] S2, the AI edge computer fuses the collected data and dynamically generates air curtain machine control instructions;

[0103] S3, the air curtain machine executes the corresponding airflow injection strategy for the door area according to the instruction;

[0104] S4. Real-time monitoring of the cooling leakage coefficient and judgment of the cooling leakage coefficient. When the cooling leakage coefficient does not exceed the threshold, the current optimization state is maintained. When the cooling leakage coefficient exceeds the threshold, a new air curtain control instruction is generated based on the collected data.

[0105] In step S4, when performing real-time monitoring of the cooling leakage coefficient, a cooling leakage evaluation model is introduced as follows;

[0106]

[0107] Where ρ represents the air density at the current temperature in the cold storage (kg / m 2 );c p represents the specific heat capacity of air (J / kg·k); A represents the effective leakage area of ​​the door area (m 2 ), which can be obtained by calculating the total area of ​​the door area minus the area blocked when the object enters and exits the door area; t represents the door opening time (s);

[0108] Based on the evaluation model, the cooling loss of the cold storage door during the door opening period is calculated in real time.

[0109] In the control execution process of the variable frequency air curtain machine, the control instructions of the air curtain machine are dynamically adjusted to achieve the linkage between the algorithm and the hardware to achieve the effect of improving energy efficiency. Then, by recording the average energy consumption performance over multiple cycles, the energy efficiency evaluation is obtained;

[0110]

[0111] In the reinforcement learning optimization mechanism, a reward function is introduced to balance the negative effects of energy consumption, temperature rise, and cooling leakage generated by the entire variable frequency air curtain process. The specific expression is as follows:

[0112] R=- ( ω1·Energy consumption + ω2·Temperature rise + ω3·Cooling leakage)

[0113] Among them, ω1+ω2+ω3=1, and the overall benefit performance of the dynamic segmented injection strategy is analyzed by the AI ​​edge computer, so that the respective weight distribution of ω1, ω2 and ω3 is updated and optimized every 24 hours.

[0114] When calculating the cold leakage risk coefficient, for the weight calculation of the area of ​​airflow blocked by the object, the cold storage door is divided into multiple sections in the height direction, including the bottom H1 of the door area, the middle H2 of the door area, and the top H3 of the door area. The risk score needs to be calculated separately for each section. At this time, the moving speed V, volume V 物 , height from the ground H 离地, for determining the position of the object in the door area, so as to calculate the proportion of each segment being blocked by the object;

[0115]

[0116] At the bottom of the door area, when the ground clearance of the object is greater than H1, the blocking proportion of this segment is 0, at this time δ = 0;

[0117] In the middle of the door area, when H 物 -H 离地 = H2, the blocking proportion is 100%, at this time the weight of δ at this time should be increased;

[0118] At the top of the door area: when H 物 ≤ H1+H2, the object is not blocked, the blocking proportion is 0, at this time δ = 0.

[0119] When calculating the risk score for each segment respectively, other parameters including ΔT, ΔH and K are the same for the whole door area, and are all calculated by equivalent replacement according to the object blocking air flow area weight calculation method.

[0120] The AI edge computer generates control instructions by deploying a lightweight neural network model to process sensor data in real time.

[0121] In S2, when the AI edge computer calculates the cold leakage risk coefficient, according to the cold leakage risk coefficient judgment in the top and middle areas, when the cold leakage coefficient is lower than the dynamic threshold, the basic execution strategy is enabled, and when the cold leakage coefficient is higher than the dynamic threshold, the dynamic segmented injection strategy is enabled, and the air curtain machine always uses pulse injection especially for the bottom area of the door.

[0122] Working principle; the intelligent cold storage door anti-leakage system, by installing infrared array sensor and multi-parameter air flow module on the cold storage door frame, to capture the cold storage door opening operation, while obtaining object parameters and environmental parameters when the cold storage door is opened, then through the design of intelligent analysis module, through the normalization processing of the captured object parameters and environmental parameters, then through data fusion design cold leakage risk coefficient calculation model, and according to the application characteristics of the door area, the door area is divided into top, middle and bottom, and different basic execution strategies are applied in different areas, then after capturing the opening action, with the import of each object parameter and environmental parameter, the cold leakage risk coefficient is calculated in real time by applying the cold leakage risk coefficient model, at this time according to the cold leakage risk coefficient, further constraint is carried out on the basis of execution strategy, realizing the intelligent control of the variable frequency air curtain machine, so that the variable frequency air curtain machine is adjusted adaptively for each divided area of the door area, especially the pulse injection is used to block the high cold leakage area at the bottom of the door, so as to reduce the cold leakage of the cold storage;

[0123] At the same time, a reinforcement learning optimization mechanism was further designed based on the intelligent analysis module. The reward function was introduced to balance the negative effects of energy consumption, temperature rise and cold leakage generated during the entire variable frequency air curtain process, thereby achieving a balance between the three important indicators of energy consumption, temperature rise and cold leakage. On the premise of ensuring that the temperature rise is within a reasonable indicator range, the energy consumption and cold leakage are reasonably adjusted to achieve a comprehensive energy-saving effect. At the same time, the entire system is based on the basic execution strategy as the prior framework, and the dynamic control method of the air curtain machine is designed on this basis, so as to achieve zero-delay response in actual application, and the cold leakage risk coefficient model is comprehensively introduced into the object parameters and environmental parameters, so as to cope with complex actual scenarios.

[0124] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent cold storage door anti-leakage system, characterized in that: It includes data acquisition module, intelligent analysis module and execution module; The data acquisition module obtains key parameters related to the opening process of the cold storage door; Object parameters: moving speed v (m / s), volume V 物 (m 3 ), height from the ground h (m), number and trajectory; Environmental parameters: internal and external temperature difference ΔT (°C), humidity difference ΔH (%), air flow velocity U (m / s) and turbulent kinetic energy K; The core technical logic of the intelligent analysis module includes multimodal data fusion, dynamic segmented injection strategy and reinforcement learning optimization mechanism; Among them, multimodal data fusion processes the data obtained by the data acquisition module, calculates the airflow blocking area S1 of the object based on the volume V of the object, and the fixed opening area of ​​the cold storage door is S2. Then, the calculation model of the cold leakage risk coefficient is designed through data fusion: Among them, α, β, γ, and δ are weight coefficients, and α+β+γ+δ=1; The execution module is a variable frequency air curtain machine, and its basic execution strategy is: A low-intensity strategy is adopted at the top of the door area, maintaining 10% to 20% of the rated power to prevent hot air from infiltrating; The middle section of the gate area adopts an adaptive speed regulation strategy, maintaining the rated power at 30% to 70%, which is suitable for personnel passage and cargo handling. The bottom of the door area uses high-intensity pulses, maintained at 80% to 100% of the rated power, to cope with the sinking of cold air and the passage of forklifts; Then, a dynamic segmented injection strategy is designed based on the cooling leakage risk factor; Among them, P i represents the real-time power of the i-th section of the wind curtain, G represents the distance attenuation coefficient, and L represents the distance between the object and the wind curtain; In this way, the variable frequency air curtain machine can be controlled to dynamically adjust according to the cold leakage risk coefficient in each area.

2. The intelligent cold storage door anti-cold leakage system according to claim 1 is characterized by: The data acquisition module includes an infrared sensor for distance measurement and a multi-parameter airflow module for airflow monitoring. The infrared sensors are arranged in an array on both sides of the cold storage door frame. The multi-parameter airflow module integrates a temperature sensor, a humidity sensor, a flow rate sensor and an air pressure sensor. The intelligent analysis module is an AI edge computer.

3. The intelligent cold storage door anti-cold leakage system according to claim 2 is characterized by: The control execution process of the variable frequency air curtain machine is as follows: S1: The cold storage door opens, and the infrared array scans the object's outline and trajectory, while the airflow module collects turbulence, temperature, humidity, and air pressure data; S2, the AI ​​edge computer performs data fusion on the collected data and dynamically generates air curtain control instructions; S3, the air curtain machine executes the corresponding airflow injection strategy for the door area according to the instruction; S4. Real-time monitoring of the cooling leakage coefficient and judgment of the cooling leakage coefficient. When the cooling leakage coefficient does not exceed the threshold, the current optimization state is maintained. When the cooling leakage coefficient exceeds the threshold, a new air curtain control instruction is generated based on the collected data.

4. The intelligent cold storage door anti-cold leakage system according to claim 3 is characterized by: In the step S4, when performing real-time monitoring of the cooling leakage coefficient, a cooling leakage evaluation model is introduced as follows; Where ρ represents the air density at the current temperature in the cold storage (kg / m 2 );c p represents the specific heat capacity of air (J / kg·k); A represents the effective leakage area of ​​the door area (m 2 ), which can be obtained by calculating the total area of ​​the door area minus the area blocked when the object enters and exits the door area; t represents the door opening time (s); Based on the evaluation model, the cooling loss of the cold storage door during the door opening period is calculated in real time.

5. The intelligent cold storage door anti-cold leakage system according to claim 3 is characterized by: In the control execution process of the variable frequency air curtain machine, the control instructions of the air curtain machine are dynamically adjusted to achieve the linkage between the algorithm and the hardware to achieve the effect of improving energy efficiency. Then, by recording the average energy consumption performance over multiple cycles, the energy efficiency evaluation is obtained; 6. The intelligent cold storage door anti-cold leakage system according to claim 5, characterized in that: In the reinforcement learning optimization mechanism, a reward function is introduced to balance the negative effects of energy consumption, temperature rise and cooling leakage generated during the entire variable frequency air curtain process. The specific expression is as follows: R=- ( ω1·Energy consumption + ω2·Temperature rise + ω3·Cooling leakage) Among them, ω1+ω2+ω3=1, and the overall benefit performance of the dynamic segmented injection strategy is analyzed by the AI ​​edge computer, so that the respective weight distribution of ω1, ω2 and ω3 is updated and optimized every 24 hours.

7. The intelligent cold storage door anti-cold leakage system according to claim 1 is characterized by: When calculating the cold leakage risk coefficient, for the weight calculation of the area of ​​airflow blocked by the object, the cold storage door is divided into multiple sections in the height direction, including the bottom H1 of the door area, the middle H2 of the door area, and the top H3 of the door area. The risk score needs to be calculated separately for each section. At this time, the moving speed V, volume V 物 , height from the ground H 离地 , used to determine the position of the object in the door area, and thus calculate the proportion of each segment blocked by the object; At the bottom of the door area, when the height of the object from the ground is greater than H1, the occlusion ratio of this section is 0, and δ = 0; In the middle of the gate area, when H 物 -H 离地 =H2, the occlusion ratio is 100%, and the δ weight should be increased at this time; At the top of the goal area: When H 物 When ≤H1+H2, the object is not blocked, the occlusion ratio is 0, and δ=0.

8. The intelligent cold storage door anti-cold leakage system according to claim 7, characterized in that: When calculating the risk score for each segment separately, other parameters including ΔT, ΔH and K are the same for the entire door area and are all calculated by equivalent replacement according to the weight calculation method of the area of ​​airflow blocked by the object.

9. The intelligent cold storage door anti-cold leakage system according to claim 2, characterized in that: The AI ​​edge computer processes sensor data in real time and generates control instructions by deploying a lightweight neural network model.

10. The intelligent cold storage door anti-cold leakage system according to claim 3, characterized in that: In S2, when the AI ​​edge computer calculates the cooling leakage risk coefficient, it judges the top area and the middle area based on the cooling leakage risk coefficient. When the cooling leakage coefficient is lower than the dynamic threshold, the basic execution strategy is enabled. When the cooling leakage coefficient is higher than the dynamic threshold, the dynamic segmented injection strategy is enabled. The air curtain machine always uses pulse injection, especially for the door bottom area.

Images