A temperature monitoring and control method in cold chain commodity transportation process

Abstract

The present application relates to the technical field of cold chain commodity transportation, and particularly relates to a temperature monitoring and control method in the process of cold chain commodity transportation, comprising: acquiring real-time temperature data, generating a thermal response digital file of the cold chain commodity according to the real-time temperature data and a cold chain commodity category database; collecting real-time position and external temperature data of the cold chain commodity, and predicting a temperature threshold of the next period according to thermal capacity characteristic parameters reflected by the thermal response digital file; constructing a three-dimensional temperature field distribution map based on the temperature threshold of the next period and vibration data of a compartment where the cold chain commodity is located; generating a control instruction according to temperature abnormal area coordinates in the three-dimensional temperature field distribution map; and after calculating a control weight coefficient according to the control instruction and a preset energy consumption optimization objective function, executing a temperature control operation according to the control weight coefficient and the control instruction, so as to effectively realize balance between temperature monitoring and control and energy consumption optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cold chain commodity transportation, and in particular to a temperature monitoring and control method in the process of cold chain commodity transportation. BACKGROUND

[0002] In the field of cold chain commodity transportation, the accuracy of temperature monitoring and control is directly related to the core quality indicators of cold chain commodities. The existing mainstream solutions generally adopt a dual-path technology architecture: one is a PID feedback control mechanism based on fixed temperature thresholds, which collects discrete point data by arranging multiple temperature sensors in the vehicle cabin, starts the compressor for refrigeration compensation when the monitoring value exceeds the preset threshold, and realizes remote data transmission of temperature data combined with GPS positioning; the second is a predictive control model based on historical meteorological database, which analyzes the correlation between environmental temperature and humidity of historical transportation routes, and performs fuzzy matching operation on the pre-set segmented temperature control strategy library.

[0003] However, these technologies all have a fundamental defect, that is, the multi-dimensional data such as heat capacity characteristics of goods, environmental disturbance factors, and equipment operating state are in a fragmented state, resulting in a constant contradiction between temperature monitoring and control real-time and energy consumption optimization.

[0004] Therefore, how to balance temperature monitoring and control and energy consumption optimization is a problem that needs to be solved by those skilled in the art. SUMMARY

[0005] In order to balance temperature monitoring and control and energy consumption optimization, the present application provides a temperature monitoring and control method in the process of cold chain commodity transportation.

[0006] The temperature monitoring and control method in the process of cold chain commodity transportation provided by the present application adopts the following technical solution:

[0007] A temperature monitoring and control method in the process of cold chain commodity transportation, comprising:

[0008] Obtaining real-time temperature data, and generating a heat response digital file of the cold chain commodity according to the real-time temperature data and a cold chain commodity category database;

[0009] Collecting real-time position and external temperature data of the cold chain commodity, and predicting a temperature threshold of the next period according to the heat capacity characteristic parameters reflected by the heat response digital file;

[0010] Based on the temperature threshold of the next period and the vibration data of the compartment where the cold chain commodity is located, a three-dimensional temperature field distribution map is constructed;

[0011] Generating a control instruction according to the temperature abnormal area coordinates in the three-dimensional temperature field distribution map;

[0012] According to the control instruction and the preset energy consumption optimization objective function, the control weight coefficient is calculated, and then the temperature control operation is performed according to the control weight coefficient and the control instruction.

[0013] Further, the step of generating the thermal response digital archive of the cold-chain commodity according to the real-time temperature data and the cold-chain commodity category database comprises:

[0014] According to the real-time temperature data, the temperature change of the cold-chain commodity is obtained to generate temperature fluctuation data;

[0015] Based on the temperature fluctuation data, the temperature response rate and the thermal inertia index of the cold-chain commodity are calculated;

[0016] According to the thermal sensitivity level parameter in the cold-chain commodity category database, the temperature response rate and the thermal inertia index are combined to construct a commodity-temperature interaction response matrix;

[0017] The feature vector in the commodity-temperature interaction response matrix is extracted to generate a heat capacity feature map;

[0018] The heat capacity feature map and the historical temperature fluctuation data are associated and mapped to generate a thermal response digital archive, wherein the thermal response digital archive records a safety threshold gradient and a critical phase change point.

[0019] Further, according to the thermal capacity characteristic parameters reflected by the thermal response digital archive, the step of predicting the temperature threshold of the next period comprises:

[0020] According to the safety threshold gradient and the critical phase change point, a basic temperature change response curve is generated;

[0021] According to the transportation route of the cold-chain commodity, real-time road condition data and weather data are obtained to calculate a thermal disturbance influence factor;

[0022] Based on the basic temperature change response curve and the thermal disturbance influence factor, a dynamic heat conduction tensor is generated;

[0023] After the thermal propagation state of the next period is predicted by the dynamic heat conduction tensor to obtain a temperature deviation probability distribution, the temperature threshold of the next period is obtained according to the temperature deviation probability distribution.

[0024] Further, the step of obtaining the temperature threshold of the next period by predicting the thermal propagation state of the next period by the dynamic heat conduction tensor to obtain a temperature deviation probability distribution, and obtaining the temperature threshold of the next period according to the temperature deviation probability distribution comprises:

[0025] According to the eigenvalues of the dynamic heat conduction tensor, the thermal propagation state distribution field of the next period is simulated;

[0026] According to the instantaneous heat flux of each node reflected by the heat propagation state distribution field of the next period, after extracting the temperature variation feature vector, a temperature deviation evolution model is constructed to obtain a temperature deviation probability distribution;

[0027] Based on the temperature deviation probability distribution, the risk weight is adjusted to obtain a confidence coefficient;

[0028] Based on the temperature deviation probability distribution and the confidence coefficient, risk simulation is performed to obtain a temperature risk factor matrix, and then the temperature threshold of the next period is calculated according to the temperature risk factor matrix.

[0029] Further, based on the temperature threshold of the next period and the vibration data of the compartment where the cold chain commodity is located, the steps of constructing a three-dimensional temperature field distribution map include:

[0030] Modal decomposition is performed on the vibration data to obtain multi-scale vibration feature components;

[0031] According to the high-frequency vibration feature components in the multi-scale vibration feature components, a vibration entropy value matrix is generated;

[0032] After coupling the vibration entropy value matrix and the temperature threshold of the next period, a heat source constraint matching condition is constructed, and then the local heat source distribution vector is obtained by deducing the heat source constraint matching condition;

[0033] Obtain the compartment structure data of the compartment where the cold chain commodity is located, and based on the compartment structure data and the local heat source distribution vector, construct a three-dimensional temperature field distribution map.

[0034] Further, according to the temperature abnormal region coordinates in the three-dimensional temperature field distribution map, the steps of generating a control instruction include:

[0035] According to the temperature abnormal region coordinates, the corresponding thermal gradient response values are extracted from the three-dimensional temperature field distribution map, and then the refrigerant diffusion path matrix is obtained through field coupling;

[0036] Based on the refrigerant diffusion path matrix, the gas flow distribution strategy is solved, and then the multi-stage compressor variable frequency parameters are deduced according to the gas flow distribution strategy;

[0037] According to the multi-stage compressor variable frequency parameters and the structure constraint conditions of the compartment where the cold chain commodity is located, a control instruction is generated, wherein the control instruction includes refrigeration control parameters and ventilation control parameters.

[0038] Further, according to the control instruction and the preset energy consumption optimization objective function, the steps of calculating the control weight coefficient include:

[0039] According to the refrigeration control parameters and the ventilation control parameters, the energy consumption prediction value is calculated;

[0040] The preset real-time electricity price parameters in the energy consumption prediction value and the preset energy consumption optimization objective function are weighted to obtain an initial weight vector;

[0041] According to the transportation route of the cold-chain commodity, the initial weight vector is adjusted in weight to obtain a control weight coefficient, wherein the control weight coefficient includes a refrigeration power parameter and a ventilation distribution parameter.

[0042] Further, according to the control weight coefficient and the control instruction, the step of performing temperature control operation includes:

[0043] According to the preset refrigeration power distribution ratio, the refrigeration power parameter is adjusted to generate a refrigeration control signal, and according to the preset air door opening degree priority, the ventilation distribution parameter is adjusted to generate a directional ventilation signal;

[0044] According to the preset system temperature mode switching threshold and the real-time temperature data, a system mode switching command is outputted;

[0045] The refrigeration control signal, the directional ventilation signal and the system mode switching command are combined to generate a temperature control operation.

[0046] The beneficial effects achieved are:

[0047] The application provides a temperature monitoring and control method in the transportation process of cold-chain commodities, including: acquiring real-time temperature data, generating a thermal response digital archive of the cold-chain commodity according to the real-time temperature data and a cold-chain commodity category database; collecting real-time position and external temperature data of the cold-chain commodity, predicting a temperature threshold of the next period according to the heat capacity characteristic parameters reflected by the thermal response digital archive; constructing a three-dimensional temperature field distribution map based on the temperature threshold of the next period and vibration data of the compartment where the cold-chain commodity is located; generating a control instruction according to the temperature abnormal area coordinates in the three-dimensional temperature field distribution map; and after calculating the control weight coefficient according to the control instruction and a preset energy consumption optimization objective function, performing a temperature control operation according to the control weight coefficient and the control instruction.

[0048] That is, in the present application, the thermal response differences of different cold chain commodities are accurately distinguished by quantifying the thermal capacity characteristic parameters of cold chain commodities through thermal response digital archives, avoiding the monitoring redundancy of traditional unified threshold for high-sensitivity commodities in temperature monitoring, and generating temperature threshold of the next period based on real-time location and external temperature data, combined with thermal capacity characteristic parameters, to predictively monitor temperature instead of passive response, significantly reducing the invalid start and stop of the compressor caused by conservative temperature monitoring; then, the three-dimensional temperature field distribution map is constructed by reusing the compartment vibration data, converting the traditional interference source into a full spatial domain temperature monitoring signal without blind area, accurately positioning the coordinates of the temperature abnormal area such as the corner of the carriage, effectively improving the temperature monitoring accuracy, and combining the control instruction generated based on the temperature abnormal area coordinates with the energy consumption optimization objective function to dynamically calculate the control weight coefficient, and finally driving the execution unit to execute temperature control operation according to the control weight coefficient and the control instruction, by dynamically matching the local temperature control intensity with the heat capacity characteristics of the goods, the standard deviation of temperature fluctuation is reduced while the energy consumption is reduced, and the mutual exclusive dilemma of temperature monitoring accuracy, control real-time performance and energy efficiency is solved at no cost of hardware modification. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 is a flowchart of a temperature monitoring and control method for cold chain commodities in the present application;

[0050] Figure 2 is a flowchart of step S10 in the temperature monitoring and control method for cold chain commodities in the present application;

[0051] Figure 3 is a flowchart of step S20 in the temperature monitoring and control method for cold chain commodities in the present application;

[0052] Figure 4 is a time-temperature coordinate system diagram;

[0053] Figure 5 is a flowchart of step S30 in the temperature monitoring and control method for cold chain commodities in the present application;

[0054] Figure 6 is a flowchart of step S40 in the temperature monitoring and control method for cold chain commodities in the present application;

[0055] Figure 7 is a flowchart of step S50 in the temperature monitoring and control method for cold chain commodities in the present application. DETAILED DESCRIPTION

[0056] The present application will be further described in detail below. Figures 1-7 The present application will be further described in detail below.

[0057] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", "third" are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0058] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0059] The embodiment of the present application discloses a temperature monitoring and control method in the process of cold chain commodity transportation.

[0060] Please refer to Figure 1 In an embodiment of the present application, a temperature monitoring and control method in the process of cold chain commodity transportation includes steps S10-S50:

[0061] Step S10, acquiring real-time temperature data, generating a thermal response digital archive of cold chain commodities according to real-time temperature data and cold chain commodity category database.

[0062] By acquiring real-time temperature data and combining the cold chain commodity category database to generate a thermal response digital archive, the purpose is to break through the extensive mode of temperature control in traditional cold chain transportation, and to realize accurate quantitative modeling of the thermal sensitivity characteristics of different categories of commodities.

[0063] In the present embodiment, by dynamically capturing real-time temperature data of goods in the actual transportation environment, a thermal response digital archive is established for cold chain commodities of different categories, and a physical basis is provided for subsequent predictive temperature control. The core effect lies in upgrading the cold chain commodity from static temperature threshold management to a dynamic thermal characteristic adaptation mechanism, capturing the differential temperature response law of high-sensitive cold chain commodities such as vaccines and ordinary fruit and vegetable cold chain commodities, eliminating the risk of temperature control redundancy or loss of control due to insufficient understanding of the thermal inertia of cold chain commodities, and strictly matching the temperature control strategy with the inherent thermal demand of cold chain commodities, thereby avoiding excessive refrigeration or local temperature rise from the source, and causing damage and energy waste.

[0064] Step S20, collect real-time location and external temperature data of cold chain commodities, and predict the temperature threshold of the next period according to the heat capacity characteristic parameters reflected in the heat response digital archive.

[0065] Collecting real-time location and external temperature data of cold chain commodities, and combining with the accurately quantified heat capacity characteristic parameters in the heat response digital archive, the temperature threshold of the next period is predicted.

[0066] By fusing the heat capacity characteristics of cold chain commodities with external temperature data and real-time location, the temperature influence trend of the transportation path environment on cold chain commodities is accurately deduced, so as to generate a temperature threshold that adapts to the specific external conditions of the next period. The purpose is to identify the thermal risk of environmental mutation areas such as tunnels and high-altitude sections in advance, so that the refrigeration system can implement precise cold energy pre-regulation based on the actual heat demand of cold chain commodities and the external temperature change trend, eliminate the temperature control delay or resource mismatch problem caused by the mismatch between environmental temperature mutation and heat capacity characteristics of cold chain commodities, and realize dynamic optimization of temperature safety boundary and on-demand allocation of temperature control resources.

[0067] Step S30, based on the temperature threshold of the next period and the vibration data of the compartment where the cold chain commodities are located, a three-dimensional temperature field distribution map is constructed.

[0068] Based on the temperature threshold of the next period and the vibration data of the compartment where the cold chain commodities are located, a three-dimensional temperature field distribution map is constructed, aiming to break through the spatial limitations of traditional point temperature monitoring.

[0069] By cross-domain coupling the temperature threshold of the next period with real-time vibration data, the physical mechanism of vibration data forming the temperature field of cold chain commodities is deeply analyzed, so as to accurately reproduce the temperature gradient distribution form of cold chain commodities in the spatial dimension of the compartment. The purpose is to fuse discrete temperature monitoring data and vibration data into a full-space continuous thermal field model, accurately locate the temperature abnormal area coordinates caused by air flow obstruction or cold chain commodity stacking, provide spatial navigation basis for directional temperature control, and eliminate the monitoring missing problem caused by temperature sensor coverage blind area, realize accurate monitoring of the temperature state of cold chain commodities inside the compartment, and lay a spatial data foundation for fine temperature regulation.

[0070] Step S40, according to the temperature abnormal area coordinates in the three-dimensional temperature field distribution map, a control instruction is generated.

[0071] According to the temperature abnormal area coordinates in the three-dimensional temperature field distribution map, a control instruction is generated, aiming to realize the essential transition of cold chain transportation temperature control mode from extensive global coverage to fine spatial directional intervention.

[0072] By the temperature anomaly region coordinates, the spatial geometric characteristics of the temperature anomaly region are analyzed and the thermodynamic gradient data is analyzed, the abstract heat field distribution information is converted into control instructions for driving the refrigeration equipment actuator to act, which breaks through the inherent limitation of the traditional temperature control strategy for the overall compartment average response, directly implements directional quantization regulation and control for the local aggregation temperature change region of the cold chain commodity, strictly matches the high-precision temperature control behavior with the actual thermal anomaly spatial distribution of the cold chain commodity, eliminates the energy consumption redundancy consumption and local temperature control failure risk caused by global refrigeration, shortens the decision link from abnormal identification to execution response, provides spatial quantitative operation basis for refrigeration power optimization allocation and directional guidance of ventilation flow, realizes precise placement and efficient use of temperature control resources in the physical space dimension.

[0073] In step S50, according to the control instruction and the preset energy consumption optimization objective function, the control weight coefficient is calculated, and then the temperature control operation is performed according to the control weight coefficient and the control instruction.

[0074] According to the control instruction and the preset energy consumption optimization objective function, the control weight coefficient is calculated, and then the temperature control operation is performed according to the control weight coefficient and the control instruction.

[0075] By placing the actual action strength of the control instruction in the energy consumption game environment corresponding to the preset energy consumption optimization objective function, the contradiction between energy consumption and temperature control precision under the fixed power output mode is broken through, the temperature control operation is performed according to the control weight coefficient and the control instruction, so that the control instruction execution process can accurately respond to the local temperature change demand of the cold chain commodity and also consider the real-time electricity price policy and carbon emission constraint requirements, eliminate the problem of excessive energy consumption or insufficient temperature control response caused by global full power operation, realize the efficient compatibility of temperature control strength and low energy consumption, and provide an intelligent execution paradigm for the cold chain transportation system that considers economy, reliability and regulatory compliance.

[0076] The specific implementation is as follows:

[0077] Regarding step S10, referring to Figure 2 , the specific implementation can be realized by steps S11-S15:

[0078] In step S11, the temperature fluctuation data is generated according to the real-time temperature data and the temperature change of the cold chain commodity.

[0079] The temperature fluctuation data reflecting the dynamic change trajectory of the temperature of the cold-chain commodity is constructed by continuously collecting real-time temperature data of the cold-chain commodity during transportation, calculating the temperature change amount between adjacent time points based on a temperature reading sequence of a fixed time interval, and specifically characterized by a continuous function record of the amplitude, frequency and change trend of the temperature rise or fall of the cold-chain commodity per unit time. The core generation logic is to convert the discrete real-time temperature data into a smooth temperature change curve through a time series interpolation algorithm, and then extract the instantaneous temperature change rate of each time node through a first-order derivative calculation, and finally integrate the temperature fluctuation data containing the temperature change direction, change rate and fluctuation period.

[0080] In step S12, the temperature response rate and the thermal inertia index of the cold-chain commodity are calculated based on the temperature fluctuation data.

[0081] The temperature response rate and the thermal inertia index of the cold-chain commodity are calculated based on the temperature fluctuation data, which is achieved by performing time domain differentiation on the continuous temperature change curve in the temperature fluctuation data, extracting the instantaneous slope of the temperature change of the cold-chain commodity per unit time as the temperature response rate, and analyzing the time decay characteristics of the cold-chain commodity maintaining temperature stability under a specific external thermal disturbance through integral operation, and quantifying the ratio of temperature change delay length to external temperature change amplitude as the thermal inertia index.

[0082] In step S13, the product-temperature interactive response matrix is constructed according to the thermal sensitivity level parameter in the cold-chain commodity category database, combined with the temperature response rate and the thermal inertia index.

[0083] In this embodiment, the inherent thermal sensitivity level parameter of the cold-chain commodity is combined with the real-time calculated temperature response rate and thermal inertia index for multi-dimensional feature fusion, an associated mapping model is established with the thermal sensitivity level as the row vector and the combination of the temperature response rate and the thermal inertia index as the column vector, and the temperature behavior prediction value of different categories of cold-chain commodities under a specific thermal disturbance environment is quantified through matrix multiplication operation.

[0084] The product-temperature interactive response matrix is essentially a two-dimensional decision table, the row coordinates correspond to the thermal sensitivity level classification, such as A-level high sensitivity / B-level medium sensitivity / C-level low sensitivity, the column coordinates are composed of the intersection of the temperature response rate interval and the thermal inertia index interval, and the matrix element value represents the theoretical temperature change trajectory function index of a specific category of cold-chain commodity under corresponding thermal capacity characteristic parameters.

[0085] In step S14, the feature vectors in the product-temperature interactive response matrix are extracted to generate the thermal capacity feature map.

[0086] The feature vector in the product-temperature interaction response matrix is extracted to generate the heat capacity feature map. The orthogonal feature vector group corresponding to the maximum singular value of the product-temperature interaction response matrix is extracted by singular value decomposition operation, and the orthogonal feature vector group is mapped into the heat capacity feature map in the thermodynamic parameter space after tensor expansion in the Hilbert space.

[0087] The heat capacity feature map is essentially a thermodynamic state space mapping table that integrates the three-dimensional characteristics of the cold chain product heat sensitivity grade, temperature response rate and thermal inertia index. The horizontal axis represents the segmented interval of the heat sensitivity grade parameter, the vertical axis represents the value range of the composite function of the temperature response rate and the thermal inertia index, and the surface height corresponds to the heat capacity equivalent value of the cold chain product under a specific thermodynamic state.

[0088] In step S15, the heat capacity feature map and the historical temperature fluctuation data are associated and mapped to generate a heat response digital archive. The heat response digital archive records the safety threshold gradient and the critical phase change point.

[0089] By establishing the space-time correspondence between the thermodynamic state space coordinates in the heat capacity feature map and the key temperature change events in the historical temperature fluctuation data, such as rapid cooling and stage heating, the heat capacity equivalent value in the heat capacity feature map is fitted as a constraint boundary function of the historical temperature change trajectory using a nonlinear regression algorithm. Then, the safety threshold gradient representing the temperature safety evolution rule of the cold chain product is decoupled, which is the maximum temperature change slope allowed per unit time, and the critical phase change point indicating the state mutation risk of the cold chain product, such as the critical temperature for thawing frozen products.

[0090] The thermodynamic state space coordinates refer to a quantitative system that converts the thermodynamic characteristics of the cold chain product into digital space positioning parameters.

[0091] In this embodiment, the temperature response rate and the thermal inertia index are calculated from the temperature fluctuation data, the product-temperature interaction response matrix is constructed by combining the heat sensitivity grade parameter, the feature vector is extracted to generate the heat capacity feature map, and finally the heat response digital archive containing the safety threshold gradient and the critical phase change point is mapped. The heat response digital archive accurately describes the thermal behavior characteristics of the cold chain product, converts the temperature monitoring from a static threshold judgment to a dynamic thermal characteristic adaptation mechanism, eliminates temperature monitoring errors caused by insufficient thermal inertia awareness, realizes differentiated temperature monitoring strategies for high-sensitivity products such as vaccines and ordinary fruits and vegetables, effectively reduces the damage rate and invalid refrigeration energy consumption in cold chain transportation, avoids temperature sudden change damage to the cell structure of cold chain products by restricting the refrigeration rate based on the safety threshold gradient, and reduces the incidence of temperature rise exceeding the standard for cold chain products based on the critical phase change point.

[0092] Regarding step S20, as shown in Figure 3 , it can be implemented through steps S21-S24.

[0093] Step S21: Generate the basic temperature change response curve based on the safety threshold gradient and the critical phase transition point.

[0094] By using the safety threshold gradient as a constraint boundary condition for the rate of temperature change, and the critical phase transition point as a segmented control node for the risk of sudden changes in the state of cold chain commodities, a basic temperature change response curve of the temperature evolution of cold chain commodities over time is constructed in the thermodynamic state space.

[0095] The fundamental temperature response curve is essentially an idealized temperature change path in the time-temperature coordinate system that strictly follows the safety threshold gradient slope limit and avoids the forbidden zone of the critical phase transition point. Specifically: First, referring to... Figure 4 As shown, the critical phase transition point (i.e. Figure 4 In this context, d0) is used as a segmented anchor point to divide the temperature variation range, such as the freezing zone (i.e., Figure 4 a1) - transition region (i.e. Figure 4 a2) - Refrigerated area (i.e. Figure 4 (a3 in the middle), then according to the safety threshold gradient in each interval. The maximum allowable temperature change slope is calculated, and finally, a continuously differentiable temperature change curve that satisfies all constraints is generated using a cubic spline interpolation algorithm, i.e., the basic temperature change response curve. .in, The initial temperature. It is a time variable.

[0096] Step S22: Based on the transportation route of cold chain goods, obtain real-time road condition data and meteorological data, and calculate the thermal disturbance impact factor.

[0097] By using the geographical coordinates of the transportation route, real-time traffic data is obtained through the API interface of the traffic information platform. At the same time, meteorological data is collected by connecting to the meteorological satellite data interface. The vibration energy spectral density in the real-time traffic data is converted into the excitation force input term in the structural dynamics equation through a physical conversion formula. Then, the structural dynamics equation is solved to obtain the stress distribution inside the compartment. Meanwhile, the atmospheric temperature gradient and solar radiation parameters in the meteorological data are input into the heat conduction boundary condition equation to construct the boundary heat flow constraint. A two-way energy transfer model is constructed by coupling the vibration heat source term with the heat conduction boundary condition.

[0098] Specifically, the physical conversion formula is as follows:

[0099]

[0100] in, For vibration excitation force density, For vibrational energy spectral density, The mass density of the compartment. is the angular frequency of vibration. After the vibration excitation force density is calculated, the calculated vibration excitation force density is substituted into the structural dynamics equation:

[0101]

[0102] wherein, is the mass density of the van, is the instantaneous acceleration caused by vibration, is the vibration stress in the van.

[0103] The vibration heat source conversion formula is:

[0104]

[0105] wherein, is the vibration heat source, is the thermal elastic loss coefficient of the van material, is the vibration frequency, which converts the vibration stress into the vibration heat source.

[0106] The heat conduction boundary condition equation is:

[0107]

[0108] wherein, is the boundary heat flow constraint, is the heat exchange efficiency between the air flow and the surface of the van, is the external environment air temperature, is the measured temperature of the van shell, is the ability of the van surface to absorb solar radiation (generally set as 0.92 for black paint and 0.3 for white paint), is the solar radiation power received per unit area (generally set as 1000 on a sunny noon, 300 on a cloudy day, and 0 in a tunnel), is the ability of the van to radiate heat outward, is the blackbody radiation constant, is the equivalent temperature of atmospheric radiation to outer space.

[0109] At this time, the vibration heat source conversion formula and the heat conduction boundary condition equation are coupled to obtain a two-way energy transfer model. The two-way energy transfer model first converts the vibration stress into a vibration heat source and injects the vibration heat source into the source term of the heat conduction boundary condition equation. At the same time, the in the heat conduction boundary condition equation is fed back to the material elastic modulus function to correct the vibration stress Because , the vibration impact force of the van caused by the uneven road surface Under the same condition, the surface temperature of the compartment When there is a change, the temperature-dependent material properties will decrease, thereby increasing the vibration stress In this dynamic iterative equilibrium process, the disturbance of vibration frequency on thermal equilibrium is quantified as a 0.4-weighted vibration disturbance term based on the three-dimensional temperature field distribution, the influence of atmospheric temperature variation gradient on heat exchange is quantified as a 0.3-weighted temperature variation disturbance term, and the thermal penetration effect of solar radiation on the compartment wall is quantified as a 0.3-weighted radiation disturbance term. Finally, the thermal disturbance influence factor with a value range of 0 to 1 is generated through time integration operation. Among them, is the reference temperature, is the temperature coefficient of material properties, is the reference temperature, is the effective bearing area of the bolt connection.

[0110] The thermal disturbance influence factor represents the comprehensive interference intensity index of the external environment on the thermal stability of the compartment per unit time.

[0111] It should be noted that the time integration operation is:

[0112]

[0113] Among them, is the thermal disturbance influence factor, and is the time window, is the vibration stress term, is the material yield strength, which can be set to 200 MPa, is the atmospheric temperature variation term, is the solar radiation correction term.

[0114] Step S23, based on the basic temperature variation response curve and the thermal disturbance influence factor, generate a dynamic thermal conduction tensor.

[0115] First, construct an initial thermal conduction tensor by the safety threshold gradient in the basic temperature variation response curve. The initial thermal conduction tensor is determined by the material intrinsic thermal conductivity and the temperature gradient direction. Then, input the thermal disturbance influence factor into the material response model to dynamically correct the thermal conductivity. Finally, synthesize the dynamic thermal conduction tensor through the Kirchhoff transformation The diagonal elements of the dynamic thermal conduction tensor reflect the position-dependent thermal conductivity, and the non-diagonal elements represent the cross-direction thermal coupling strength.

[0116] The material response model is:

[0117]

[0118] Among them, To correct the thermal conductivity, Based on the basic thermal conductivity, This is the temperature gradient gain coefficient. This is the disturbance attenuation coefficient. Represents a unit tensor. Let the initial heat conduction tensor be... To control the correction intensity, in the tunnel scenario, the control correction intensity is set to 0.4.

[0119] Step S24: After rolling the prediction of the heat propagation state in the next time period by using the dynamic heat conduction tensor to obtain the temperature deviation probability distribution, the temperature threshold for the next time period is obtained based on the temperature deviation probability distribution.

[0120] Furthermore, step S24 can be specifically implemented through steps S25~S28:

[0121] Step S25: Based on the characteristic value of the dynamic heat conduction tensor, simulate the heat propagation state distribution field for the next time period.

[0122] By analyzing the dynamic heat conduction tensor Eigenvalue decomposition is performed to obtain the diagonal elements (i.e., eigenvalues) in the eigenvector matrix. The thermal state evolution equation is constructed by combining the initial temperature field vector, and the attenuation characteristics of thermal disturbances at different spatial frequencies are quantified. The global temperature distribution field for the next time period is obtained by solving the equation by finite element discretization.

[0123] It should be noted that the thermal state evolution equation is:

[0124]

[0125] in, for Temperature field distribution after time, The eigenvector matrix, For attenuation operators, It is the inverse of the eigenvector matrix. For the initial temperature field, the decay operator diagonal elements Quantify the attenuation characteristics of thermal disturbances at different spatial frequencies.

[0126] The specific method for obtaining the global temperature distribution field in the next time period by discretizing the equation using the finite element method is as follows: First, the cold chain compartment space is discretized into a finite number of element nodes, and the temperature of each node forms a node vector. At this time, the dynamic heat conduction tensor is decomposed into eigenvalues. , where the eigenvalue matrix diagonal elements The decay rate of different heat conduction modes is characterized, while the eigenvector matrix Q describes the spatial pattern of heat propagation, obtained through mode space transformation of the initial temperature field. and time evolution (exponential term accurately quantifying the energy decay ratio of each mode at the time step , finally reconstructing the temperature distribution of the physical space to obtain the global temperature distribution of the next period.

[0127] Step S26, according to the instantaneous heat flux of each node reflected by the heat propagation state distribution field of the next period, the temperature variation feature vector is extracted, and then a temperature deviation evolution model is constructed to obtain a temperature deviation probability distribution.

[0128] According to the instantaneous heat flux of each node in the heat propagation state distribution field of the next period, the temperature variation feature vector is extracted, and first the Fourier law is used to convert the discrete node temperature variation feature vector into an instantaneous heat flux density vector , then singular value decomposition is performed on the instantaneous heat flux density vector to obtain a temperature variation feature vector matrix, based on which a temperature deviation evolution model is constructed , after simulating the evolution trajectory under environmental disturbance 1000 times by the Monte Carlo method, the temperature deviation value distribution of each node is fitted into a temperature deviation probability distribution.

[0129] Step S27, based on the temperature deviation probability distribution, the risk weight is adjusted to obtain a confidence coefficient.

[0130] First, the temperature deviation value of each node output by the temperature deviation evolution model is fitted into a normal distribution, and a risk weight function is constructed based on its probability density function (wherein, is the risk weight, is the sensitivity coefficient, which can be set to 5; 8 is the nonlinear gain), which gives higher weight to high temperature deviation, and then generates the confidence coefficient through integral operation (wherein, is the confidence coefficient, is the safety threshold gradient, is the probability density function of the temperature deviation probability distribution, is the integral infinitesimal) The physical nature of which represents the compression mapping of the temperature overrun risk probability.

[0131] Step S28, combine the temperature deviation probability distribution and the confidence coefficient to simulate the risk, obtain the temperature risk factor matrix, and calculate the temperature threshold of the next period according to the temperature risk factor matrix.

[0132] ​1000 groups of random temperature deviation field samples are generated according to the temperature deviation probability distribution, then the confidence coefficient is injected as a failure probability compensation item into the risk simulation, the temperature risk factor of each spatial position is calculated, and finally the temperature risk factor matrix is integrated by all spatial nodes, the temperature risk factor matrix quantifies the local temperature rise out-of-control probability through the element value of the temperature risk factor, and the next period temperature threshold is calculated based on the temperature risk factor matrix, and the equation is wherein, is the next period temperature threshold, is the static threshold, is the shrinkage coefficient.

[0133] In this embodiment, the physical boundary of the temperature change of the cold chain commodity is constrained by the safety threshold variable gradient, the state mutation protection threshold is set in combination with the critical phase change point, the basic temperature change response curve is generated as a theoretical benchmark, the real-time road condition data and meteorological data in the transportation route are fused, the thermal disturbance influence factor is calculated to quantify the environmental disturbance intensity, the dynamic thermal conduction tensor is generated, the next period thermal propagation state is predicted by the dynamic thermal conduction tensor, the temperature deviation probability distribution is output, and finally the temperature threshold of the next period is dynamically generated according to the temperature deviation probability distribution, which effectively reduces the error of subsequent temperature control prediction.

[0134] Regarding step S30, as shown in Figure 5 , it can be implemented through steps S31-S34:

[0135] Step S31, modal decomposition is performed on the vibration data to obtain a multi-scale vibration feature component.

[0136] In this embodiment, the vibration data is decomposed by the empirical mode decomposition algorithm, specifically: first, the local extreme points of the vibration data are identified and the upper and lower envelope lines are constructed, the envelope mean curve m[n] is calculated, and the candidate component h[n]=a[n]-m[n] is separated out, and the iteration is screened until h[n] meets the intrinsic mode function condition, that is, the number difference between the extreme points and the zero-crossing points is not more than 1 and the local mean value is zero, then the output h[n] that meets the intrinsic mode function condition is output as the first-order multi-scale vibration feature classification c1[n], at this time, the vibration data a[n] is subtracted from the first-order multi-scale vibration feature classification c1[n] to obtain the residual term r1[n], the local extreme points of the residual term r1[n] are identified and the upper and lower envelope lines are constructed, and the second-order multi-scale vibration feature classification c2[n] is calculated, and so on, until the residual term is a monotonic trend term, the calculated multiple multi-scale vibration feature classifications are arranged in descending order of frequency to form the multi-scale vibration feature component.

[0137] Wherein, h[n] is a candidate component, a[n] is the original vibration data sequence formed after the local extreme point identification and the construction of upper and lower envelope lines of the vibration data.

[0138] Step S32, generating a vibration entropy value matrix according to the high-frequency vibration feature components in the multi-scale vibration feature components.

[0139] In this embodiment, the generation of the vibration entropy value matrix is realized by performing time-frequency energy analysis on the high-frequency vibration feature components in the multi-scale vibration feature components, specifically: the high-frequency vibration feature components are transformed by short-time Fourier transform to calculate a time-frequency energy density matrix , wherein represents the energy intensity of time i and frequency j, and then the energy probability distribution of each time window is calculated based on the definition of Shannon entropy . The vibration entropy value is generated , and after the above calculation is sequentially performed on each time window, the vibration entropy values corresponding to each time window calculated are integrated to generate a vibration entropy value matrix .

[0140] Wherein, is the time-frequency energy density matrix, T is the total number of time windows, and F is the total number of components of the multi-scale vibration feature components.

[0141] Step S33, coupling the vibration entropy value matrix and the temperature threshold of the next time period, constructing a heat source constraint matching condition, and deriving the local heat source distribution vector through the heat source constraint matching condition.

[0142] The vibration entropy value matrix and the temperature threshold of the next time period are input into a tensor product operation to obtain a joint constraint tensor , which is expanded into a heat source constraint matching condition through a Kronecker product .

[0143] When the heat source constraint matching condition in this embodiment derives the local heat source distribution vector , the following heat source constraint matching condition is solved:

[0144]

[0145]

[0146] Under the premise of meeting the heat conduction equation conversion condition , the objective function is minimized.The local heat source distribution vector is obtained by matching the heat source gradient and the vibration entropy gradient, and then the local heat source distribution vector is obtained by iteration calculation through the Lagrange multiplier method The heat source intensity of the space node i is represented.

[0147] Wherein, The coupling strength of the local heat source distribution vector and the vibration entropy matrix is, The material thermal elastic basic tensor is, The maximum allowed temperature rise is.

[0148] In step S34, the compartment structure data of the compartment where the cold chain commodity is located is obtained, and a three-dimensional temperature field distribution map is constructed based on the compartment structure data and the local heat source distribution vector.

[0149] By accessing the vehicle design database, the compartment geometric topology parameters such as length, width, height, layer position, material thickness distribution, etc. are extracted. In order to improve the accuracy of the structure data, the deformation error caused by assembly tolerance is corrected by combining real-time scanning point cloud data, and the compartment structure data is obtained.

[0150] Based on the length, width and height of the compartment, the domain boundary is defined, such as a rectangular domain boundary of 12.5m x width 2.4m x height 2.2m. The layer position divides the domain boundary into independent sub-regions, and the heat flow continuity condition is applied on each independent sub-region. The material thickness distribution is integrated with the weight of the equation discretization by modifying the unit dynamic heat conduction tensor. At the same time, the local heat source distribution vector is injected as an equation term, and the material attribute parameters, i.e. density, specific heat capacity and thermal conductivity, are loaded.

[0151] The time derivative term is discretized by using the back Euler format to generate the transient iteration equation:

[0152]

[0153] At this time, the non-steady-state heat conduction equation is discretized to obtain the element stiffness matrix K and the heat capacity matrix C, which are:

[0154] The continuous temperature field T is approximated by shape functions, and , The shape function of node j is, The shape function of node j is, The shape function is substituted into the non-steady-state heat conduction equation, and the residual is weighted and integrated in the calculation domain and forced to be zero, which is , which is reduced by Green's formula, which is + boundary term, and finally is calculated, wherein, The heat capacity matrix C isK is the element stiffness matrix.

[0155] Next, the element stiffness matrix K and the heat capacity matrix C are combined to form a global linear equation system:

[0156]

[0157] Apply convective heat transfer boundaries to the nodes on the outer surface of the chamber. Furthermore, a forced thermal equilibrium is established at the interlayer interface, and the global linear equations are iteratively solved until the residual norm is less than 1. Finally, the global node temperature vector T is output, and a dynamic three-dimensional temperature field distribution map is generated through spatial interpolation mapping.

[0158] in, For density, ρ is the specific heat capacity, k is the thermal conductivity. For the temperature field at time t, For time step, Let be the shape function of node i, t be the time variable, and the boundary term be the quantization of heat dissipation from the outer surface of the compartment. For the infinitesimal element when the volume integral is performed over the region Ω, Let be the rate of change of temperature over time at node j.

[0159] In this embodiment, multi-scale vibration feature components are extracted through modal decomposition of vibration data. A vibration entropy matrix is ​​generated using high-frequency vibration feature components to quantify the vibration data. The vibration entropy matrix is ​​coupled with the temperature threshold of the next time period to construct a heat source constraint matching condition. The local heat source distribution vector is derived through this heat source constraint matching condition to accurately locate vibration-induced heating risk points. Combined with the body structure data, a three-dimensional temperature field distribution map is constructed to realize spatial visualization of the internal thermal field of the body, reducing the temperature rise prediction error caused by local vibration.

[0160] Regarding step S40, refer to... Figure 6 As shown, this can be achieved through steps S41 to S43:

[0161] Step S41: Based on the coordinates of the temperature anomaly area, extract the corresponding thermodynamic gradient response value from the three-dimensional temperature field distribution map, and then obtain the refrigerant diffusion path matrix through field strength coupling.

[0162] In this embodiment, based on the coordinates of the temperature anomaly region, the core nodes of the anomaly are located in the three-dimensional temperature field distribution map. A vertex sequence of the closed boundary contour is generated using an isosurface extraction algorithm. Simultaneously, the temperature gradient magnitude of each node within the closed boundary contour is calculated, and the average value is taken as the thermodynamic gradient response value. Then, a field strength coupling model is used to physically correlate the thermodynamic gradient response value with the refrigerant pressure field, i.e., a coupling equation is established. Solving this equation yields the pressure gradient distribution. Then the Darcy's law is used The flow rate field q is calculated, and finally the flow rate field is streamline integrated to generate the refrigerant diffusion path matrix wherein k1 is the passing capacity of the refrigerant on the diffusion path, is the friction resistance of the refrigerant on the diffusion path, is a time microelement, is a thermal gradient conversion tensor.

[0163] After the air flow distribution strategy is solved based on the refrigerant diffusion path matrix, step S42, the multi-stage compressor variable frequency parameters are derived according to the air flow distribution strategy.

[0164] The refrigerant diffusion path matrix is analyzed, and the objective function is constructed, and the air flow distribution strategy satisfying the total air volume constraint and the air diffusion demand of each node greater than 0.7 is solved by linear programming.

[0165] According to the total air volume demand and the maximum pressure loss in the air flow distribution strategy, the multi-stage compressor variable frequency parameters are derived according to the compressor characteristic curve , wherein the basic frequency , the variable frequency increment , and finally the variable frequency parameter combination , , i.e. the multi-stage compressor variable frequency parameters, are output.

[0166] wherein is the total number of damper partition areas outputting refrigerant, is the flow resistance coefficient of the kth area air duct, is the target air volume of the kth area, is the deviation penalty weight coefficient, is the required air volume of the kth area, is the proportional gain coefficient of the compressor, is the basic air volume, i.e. the factory air volume under the steady state working condition.

[0167] Step S43, according to the multi-stage compressor variable frequency parameters and the structure constraint condition of the compartment where the cold chain goods are located, a control instruction is generated, wherein the control instruction includes refrigeration control parameters and ventilation control parameters.

[0168] The multi-stage compressor variable frequency parameters are input into the compressor characteristic equation , the output power is calculated, and then the refrigeration control parameters are generated in combination with the refrigerant flow demand .

[0169] Simultaneously, the geometric boundaries of the ventilation ducts are defined based on the length, width, and height dimensions in the compartment structure data; independent wind control subdomains are divided according to the partition positions; and the drag coefficient is corrected based on the material thickness distribution. This is achieved by solving a constrained optimization problem. Generate ventilation control parameters Ultimately, the cooling control parameters and ventilation control parameters are integrated to form a control command set, which drives the actuators to respond in a coordinated manner.

[0170] in, The actual opening degree of the damper in zone k. Let K be the target opening degree of the damper in zone k.

[0171] In this embodiment, the thermodynamic gradient response value corresponding to the coordinates of the temperature anomaly area is extracted, and a refrigerant diffusion path matrix is ​​generated through field strength coupling to quantify the optimal transport path of the refrigerant in the compartment space. Using this refrigerant diffusion path matrix, the airflow distribution strategy is solved to achieve the targeted delivery of refrigeration resources. Based on the airflow distribution strategy, the frequency conversion parameters of the multi-stage compressor are derived to dynamically adapt to the local heat load demand. Finally, combined with the structural constraints of the compartment, control commands containing refrigeration control parameters and ventilation control parameters are generated to improve the temperature adjustment response speed at the temperature anomaly point, improve the temperature adjustment accuracy, and avoid the problem of excessive energy consumption caused by overall temperature adjustment.

[0172] Regarding step S50, refer to... Figure 7 As shown, this can be achieved through steps S51 to S56:

[0173] Step S51: Calculate the predicted energy consumption value based on the cooling control parameters and ventilation control parameters.

[0174] Obtain the output power from the cooling control parameters Based on the ventilation control parameters and the wind resistance distribution parameters in the structural constraints of the enclosure, and based on the fan power consumption model... Calculate the total power consumption of the ventilation system. Then, the standby power consumption of the basic equipment is added to the total power consumption of the ventilation system. To obtain the predicted energy consumption value .

[0175] in, Let be the efficiency coefficient of the wind turbine in zone k. Let be the drag coefficient of the k-th zone.

[0176] Step S52: Perform weighted calculations on the predicted energy consumption value and the preset real-time electricity price parameter in the preset energy consumption optimization objective function to obtain the initial weight vector.

[0177] Energy consumption forecast Substituting into the first linear weighting function:

[0178]

[0179] Meanwhile, the preset real-time electricity price parameter is substituted into the second linear weighting function:

[0180]

[0181] An initial weight vector is obtained .

[0182] wherein, is a basic energy consumption value, is an energy consumption weight gain coefficient, is a benchmark electricity price, is an electricity price weight gain coefficient.

[0183] Step S53, according to the transportation route of the cold-chain commodity, the initial weight vector is adjusted in weight to obtain a control weight coefficient, wherein the control weight coefficient includes a refrigeration power parameter and a ventilation distribution parameter.

[0184] Firstly, the transportation route is analyzed to obtain a curvature radius, an altitude gradient, and a tunnel density of the transportation route, and according to the curvature radius , the altitude gradient , and the tunnel density a route feature vector is generated Then, the initial weight vector is input into an adjustment function with the route feature vector to obtain an adjusted weight .

[0185] Then, the adjusted weight is converted into a control weight coefficient through a weight mapping model, wherein the refrigeration power parameter , and the ventilation distribution parameter .

[0186] wherein, is a route influence coefficient matrix, which is taken as [0.15, -0.1, 0.2] in a conventional application scenario; is an energy consumption weight component, is an electricity price weight component, is a benchmark damper opening parameter, is a compartment structure air guide matrix, for example, the value of the corner area in the compartment is 0.8, and the value of the center area is 0.3.

[0187] Step S54, according to a preset refrigeration power distribution ratio, the refrigeration power parameter is adjusted to generate a refrigeration control signal, and according to a preset damper opening priority, the ventilation distribution parameter is adjusted to generate a directional ventilation signal.​​

[0188] It should be noted that the preset refrigeration power distribution ratio in the embodiment is a three-level refrigeration power distribution , and the preset damper opening priority .

[0189] According to the preset refrigeration power distribution ratio , the refrigeration power parameter is adjusted to generate a hierarchical refrigeration control signal , such as a first-level compressor frequency , a second-level compressor frequency , and a third-level compressor frequency .

[0190] At the same time, according to the preset damper opening priority, the ventilation distribution parameter is adjusted to generate a directional ventilation signal , such as a directional ventilation signal of the third-priority corner damper opening .

[0191] Step S55, according to the preset system temperature mode switching threshold and the real-time temperature data, outputting a system mode switching command.

[0192] It should be noted that the preset system temperature mode switching threshold in the embodiment is , wherein represents the energy-saving mode, which can be set to 2.0℃; represents the balanced mode, which can be set to 1.2℃; represents the high-precision mode, which can be set to 0.5℃.

[0193] First, input the real-time temperature data into the temperature rise deviation calculation formula to calculate the current maximum temperature rise deviation , and then determine the system mode switching command by judging the size relationship between the current maximum temperature rise deviation and the preset system temperature mode switching threshold.

[0194] When ≤ , the system mode switching command is to switch to the energy-saving mode; when < ≤ , the system mode switching command is to switch to the balanced mode; when > , the system mode switching command is to switch to the high-precision mode.

[0195] Step S56, combining the refrigeration control signal, the directional ventilation signal and the system mode switching command, to generate a temperature control operation.

[0196] determining the control intensity coefficient based on the system mode switching command weighting the stepped compressor frequency in the refrigeration control signal with the control intensity coefficient to obtain a modified refrigeration control signal ; meanwhile, weighting the directional ventilation signal with the control intensity coefficient to obtain a modified directional ventilation signal .

[0197] integrating the modified refrigeration control signal and the modified directional ventilation signal to output a temperature control operation, that is, generating control strategies corresponding to different fans and compressors in the same compartment to achieve temperature control of different regions in the compartment, so as to avoid the problem that the temperature control accuracy and energy consumption cannot be balanced due to the use of the same control strategy for the fans and compressors corresponding to different regions in the same compartment.

[0198] In the embodiment, the energy consumption prediction value is calculated by the refrigeration control parameter and the ventilation control parameter, and the initial weight vector is generated by weighting in combination with the preset real-time electricity price parameter. The control weight coefficient is obtained by dynamically adjusting the initial weight vector according to the transportation route. The control weight coefficient is adjusted based on the preset refrigeration power distribution ratio and the preset air door opening priority, respectively, to generate the refrigeration control signal and the directional ventilation signal. After the system mode switching command is output according to the system temperature mode switching threshold and the real-time temperature data, the temperature control operation is integrated to generate, so as to improve the response speed of the refrigeration system and the distribution accuracy of the ventilation flow, and reduce the control energy consumption.

[0199] The above are preferred embodiments of the present application, and do not limit the protection scope of the present application. Therefore, any equivalent changes made on the basis of the structure, shape, principle of the present application should be covered within the protection scope of the present application.

Claims

1. A method for temperature monitoring and control during cold chain commodity transportation, characterized in that, include: Real-time temperature data is acquired, and based on the real-time temperature data and the cold chain commodity category database, a thermal response digital profile of the cold chain commodity is generated, wherein the thermal response digital profile records the safety threshold gradient and the critical phase transition point. Collect real-time location and external temperature data of the cold chain goods, and predict the temperature threshold for the next period based on the heat capacity characteristic parameters reflected in the thermal response digital file; Based on the temperature threshold of the next time period and the vibration data of the compartment where the cold chain goods are located, a three-dimensional temperature field distribution map is constructed. Based on the coordinates of the temperature anomaly region in the three-dimensional temperature field distribution map, control commands are generated; Based on the control command and the preset energy consumption optimization objective function, the control weight coefficient is calculated, and then the temperature control operation is executed according to the control weight coefficient and the control command. The step of predicting the temperature threshold for the next time period based on the heat capacity characteristic parameters reflected in the thermal response digital archive includes: Based on the safety threshold gradient and the critical phase transition point, a basic temperature change response curve is generated; Based on the transportation route of the cold chain goods, real-time road condition data and meteorological data are obtained, and the thermal disturbance impact factor is calculated. Based on the basic temperature change response curve and the thermal disturbance influence factor, a dynamic heat conduction tensor is generated. Based on the characteristic values ​​of the dynamic heat conduction tensor, the heat propagation state distribution field for the next time period is simulated; Based on the instantaneous heat flux of each node reflected by the heat propagation state distribution location in the next time period, after extracting the temperature change feature vector, a temperature deviation evolution model is constructed to obtain the temperature deviation probability distribution. The confidence coefficient is obtained by adjusting the risk weights based on the temperature deviation probability distribution. By combining the temperature deviation probability distribution and the confidence coefficient to perform risk simulation and obtain the temperature risk factor matrix, the temperature threshold for the next time period is calculated based on the temperature risk factor matrix.

2. The method for temperature monitoring and control during cold chain commodity transportation according to claim 1, characterized in that, The step of generating the thermal response digital profile of the cold chain product based on the real-time temperature data and the cold chain product category database includes: Based on the real-time temperature data, the temperature change of the cold chain goods is obtained, and temperature fluctuation data is generated; Based on the temperature fluctuation data, the temperature response rate and thermal inertia index of the cold chain product are calculated. Based on the heat sensitivity level parameters in the cold chain commodity category database, and combined with the temperature response rate and the thermal inertia index, a commodity-temperature interaction response matrix is ​​constructed. Extract the feature vectors from the commodity-temperature interaction response matrix to generate a heat capacity feature map; The thermal capacity characteristic spectrum and historical temperature fluctuation data are correlated and mapped to generate the thermal response digital profile.

3. The temperature monitoring and control method for cold chain commodity transportation according to claim 1, characterized in that, The step of constructing a three-dimensional temperature field distribution map based on the temperature threshold of the next time period and the vibration data of the cold chain goods in the container includes: Modal decomposition is performed on the vibration data to extract multi-scale vibration feature components; Based on the high-frequency vibration characteristic components in the multi-scale vibration characteristic components, a vibration entropy matrix is ​​generated; After coupling the vibration entropy matrix and the temperature threshold of the next time period to construct the heat source constraint matching condition, the local heat source distribution vector is obtained by derivation through the heat source constraint matching condition. Obtain the compartment structure data of the compartment where the cold chain goods are located, and construct the three-dimensional temperature field distribution map based on the compartment structure data and the local heat source distribution vector.

4. The method for temperature monitoring and control during cold chain commodity transportation according to claim 1, characterized in that, The step of generating control commands based on the coordinates of the temperature anomaly region in the three-dimensional temperature field distribution map includes: Based on the coordinates of the temperature anomaly region, the corresponding thermodynamic gradient response value is extracted from the three-dimensional temperature field distribution map, and then the refrigerant diffusion path matrix is ​​obtained through field strength coupling. Based on the refrigerant diffusion path matrix, after solving the airflow distribution strategy, the multi-stage compressor frequency conversion parameters are derived according to the airflow distribution strategy. Based on the variable frequency parameters of the multi-stage compressor and the structural constraints of the compartment where the cold chain goods are located, the control command is generated, wherein the control command includes refrigeration control parameters and ventilation control parameters.

5. The method for temperature monitoring and control during cold chain commodity transportation according to claim 4, characterized in that, The step of calculating the control weight coefficients based on the control command and the preset energy consumption optimization objective function includes: Based on the cooling control parameters and the ventilation control parameters, the predicted energy consumption value is calculated. The energy consumption prediction value and the preset real-time electricity price parameter in the preset energy consumption optimization objective function are weighted and calculated to obtain an initial weight vector; Based on the transportation route of the cold chain goods, the initial weight vector is adjusted to obtain the control weight coefficient, wherein the control weight coefficient includes refrigeration power parameters and ventilation distribution parameters.

6. The method for temperature monitoring and control during cold chain commodity transportation according to claim 5, characterized in that, The step of performing temperature control operation based on the control weight coefficient and the control command includes: Based on a preset cooling power allocation ratio, the cooling power parameters are adjusted to generate a cooling control signal; and based on a preset damper opening priority, the ventilation distribution parameters are adjusted to generate a directional ventilation signal. Based on the preset system temperature mode switching threshold and the real-time temperature data, a system mode switching command is output; The temperature control operation is generated by combining the cooling control signal, the directional ventilation signal, and the system mode switching command.

Images