Intelligent distribution scheduling method and device for egg products and medium

Abstract

The invention discloses an intelligent distribution scheduling method and device for egg products and a medium, and relates to the technical field of intelligent scheduling, and the method comprises the steps: collecting biological characteristic parameters of target egg products in real time, carrying out the feature encryption processing through an edge calculation node to form private data packets, receiving the private data packets of all suppliers through a message queue service, and sending the private data packets to a server; calling a homomorphic encryption library to execute weighted average aggregation on the encrypted feature vectors; and loading real-time road condition data, compartment temperature and humidity distribution and a vibration spectrum in a digital twin sandbox based on a global egg rheology knowledge graph and a microbial growth kinetic model, and coupling to solve a product risk value of a protein shear stress accumulation equation and a flora growth equation. By constructing a global egg product rheology knowledge graph and a microbial growth kinetic model, the time-space coupling characteristics of biomechanical characteristics and microbial activity of egg products in transportation are accurately captured, and the quality risk quantification precision and the transportation strategy reliability are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent scheduling technology, and in particular to an intelligent delivery scheduling method, equipment and medium for egg products. Background Technology

[0002] Existing cold chain logistics technologies have enabled real-time monitoring of the egg transportation environment based on the Internet of Things (IoT), and have initially achieved privacy-preserving collaboration of multi-supplier data using a federated learning framework. The application of digital twin technology in static modeling of the microclimate of transport vehicles is maturing, supporting predictions of single-factor risks or linear microbial growth based on historical data. Egg rheological studies have confirmed a strong correlation between protein shear stress accumulation and mechanical vibration, while microbial kinetic models can also characterize the nonlinear effects of temperature and humidity on the proliferation of spoilage bacteria.

[0003] However, the essence of egg spoilage risk lies in the strong coupling effect of physical and biological factors. The accumulation of protein shear stress caused by road vibration and the accelerated proliferation of microorganisms triggered by temperature and humidity fluctuations are both affected by the nonlinear interplay of multiple parameters such as real-time road conditions, temperature and humidity drift, and vibration spectrum. Existing static models and single-factor threshold monitoring technologies cannot solve such coupled equations, resulting in the inability to quantify the composite risk value. This leads to insufficient accuracy in risk prediction, delayed response of transportation strategies, and a lack of precise quantitative basis for route optimization and refrigeration adjustment commands, resulting in a persistently high rate of egg loss during transportation. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an intelligent delivery scheduling method for eggs to solve the problem of accurately predicting risks and dynamically optimizing routes and refrigeration strategies by coupling environmental parameters in real time during egg transportation.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides an intelligent delivery scheduling method for egg products, which includes real-time collection of biological characteristic parameters of target egg products and feature encryption processing through edge computing nodes to form privacy data packets;

[0008] The message queue service receives privacy data packets from various vendors and calls the homomorphic encryption library to perform weighted average aggregation on the encrypted feature vectors.

[0009] Based on a global knowledge graph of egg rheology and a microbial growth dynamics model, real-time road condition data, temperature and humidity distribution in the carriage, and vibration spectrum are loaded into a digital twin sandbox, and the product risk value of the protein shear stress accumulation equation and the microbial community growth equation is solved in a coupled manner.

[0010] If the product risk value does not exceed the dynamic threshold, execute the basic transportation status;

[0011] If the product risk value exceeds the dynamic threshold, generate a hyperoxia exposure path optimization instruction;

[0012] The high oxygen exposure path optimization command is sent to the vehicle edge computing node to dynamically correct the geofence parameters of the navigation path and adjust the cooling power curve.

[0013] In a preferred embodiment of the intelligent delivery and scheduling method for egg products described in this invention, the specific steps of the feature encryption processing are as follows:

[0014] The temperature dependence function of egg white viscoelasticity coefficient with egg age and the basal proliferation rate of Salmonella were collected in real time from the biological characteristic parameters of the target egg products.

[0015] The feature vector of the temperature-dependent function is extracted by the edge computing node, and the feature vector is subjected to homomorphic encryption operation to generate encrypted feature vector. The basic proliferation rate of Salmonella is then bound to the encrypted feature vector to form a privacy data packet.

[0016] As a preferred embodiment of the intelligent delivery scheduling method for egg products described in this invention, the specific steps for dynamically aggregating encrypted biological parameters from multiple egg product suppliers to construct a global egg product rheological knowledge graph and a microbial growth kinetic model are as follows.

[0017] The message queue service receives privacy data packets from various vendors and calls the homomorphic encryption library to perform weighted average aggregation on the encrypted feature vectors.

[0018] The basic proliferation rate of Salmonella is weighted and fused according to the sample size of the supplier to output a global egg rheology knowledge graph containing the mapping relationship of viscoelasticity-temperature-egg age;

[0019] Based on the weighted fusion of Salmonella baseline proliferation rate, a microbial growth kinetic model incorporating the relationship between community proliferation rate and temperature response was constructed.

[0020] As a preferred embodiment of the intelligent delivery scheduling method for egg products described in this invention, the coupling solution involves the following specific steps.

[0021] The road bump index is obtained from the digital twin sandbox to extract real-time road condition data, and the spatial gradient of the temperature and humidity distribution in the vehicle compartment and the energy distribution of the vibration spectrum are loaded.

[0022] The cumulative value of protein shear stress was calculated based on a global egg rheology knowledge graph, and the instantaneous growth rate of the microbial community was calculated based on a microbial growth kinetics model.

[0023] Based on the coupling relationship between the cumulative protein shear stress and the instantaneous growth rate of the bacterial community, a product risk value is output.

[0024] As a preferred embodiment of the intelligent delivery scheduling method for egg products described in this invention, the coupling solution involves the following specific steps.

[0025] The road bump index is obtained from the digital twin sandbox to extract real-time road condition data, and the spatial gradient of the temperature and humidity distribution in the vehicle compartment and the energy distribution of the vibration spectrum are loaded.

[0026] The cumulative value of protein shear stress was calculated based on a global egg rheology knowledge graph, and the instantaneous growth rate of the microbial community was calculated based on a microbial growth kinetics model.

[0027] Based on the coupling relationship between the cumulative protein shear stress and the instantaneous growth rate of the bacterial community, a product risk value is output.

[0028] In a preferred embodiment of the intelligent delivery scheduling method for egg products described in this invention, the high-oxygen exposure path optimization instruction refers to:

[0029] The oxygen enrichment benefit weight coefficient of forest road sections is calculated based on the magnitude of the product risk value exceeding the dynamic threshold, and liquid nitrogen rapid cooling trigger conditions are generated, which include the bacterial community proliferation rate threshold and rapid cooling temperature.

[0030] The oxygen enrichment benefit weight coefficient of forest road sections and the liquid nitrogen rapid cooling trigger condition are encapsulated into parsable JSON instructions, and geofence parameter correction instructions are generated based on the oxygen enrichment benefit weight coefficient of forest road sections.

[0031] Integrate JSON commands and geofence correction commands to generate high-oxygen exposure path optimization commands.

[0032] As a preferred embodiment of the intelligent delivery scheduling method for egg products described in this invention, the specific steps for issuing the high-oxygen exposure path optimization command to the vehicle-mounted edge computing node are as follows.

[0033] The high oxygen exposure path optimization instructions are transmitted to the vehicle-mounted edge computing node via an encrypted communication link, and the high oxygen exposure path optimization instructions are received and stored in the instruction cache area.

[0034] The homomorphic decryption interface of the edge computing node is used to perform decryption operation on the high oxygen exposure path optimization instruction, and the decrypted executable instruction is written into the real-time instruction processing queue of the vehicle control system.

[0035] In a preferred embodiment of the intelligent delivery scheduling method for egg products described in this invention, the dynamic correction specifically includes the following steps.

[0036] The weight coefficients of forest road sections and the liquid nitrogen rapid cooling triggering conditions in the high oxygen exposure route optimization instructions are analyzed. Based on the weight coefficients of forest road sections, the geofencing parameters are corrected, and the optimized transportation route trajectory is generated through a route planning algorithm.

[0037] Based on the liquid nitrogen rapid cooling trigger condition, the power output curve of the refrigeration equipment is dynamically adjusted according to the temperature difference-power response relationship, and the optimized transportation path trajectory and refrigeration power execution status are fed back to the digital twin sandbox.

[0038] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the intelligent delivery scheduling method for egg products as described in the first aspect of the present invention.

[0039] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the intelligent delivery scheduling method for egg products as described in the first aspect of the present invention.

[0040] The beneficial effects of this invention are as follows: By dynamically aggregating multi-source encrypted biological parameters through a federated learning engine and simultaneously constructing a global rheology-microbiology knowledge graph, the spatiotemporal coupling characteristics of the biomechanical properties and microbial activity of eggs during transportation are accurately captured, improving the accuracy of quality risk quantification and the reliability of transportation strategies. By loading real-time road condition spectrum and temperature and humidity distribution gradients into a digital twin sandbox, the product risk value of the protein shear stress accumulation equation and the microbial community growth kinetic equation is solved, enhancing the robustness of collaborative prediction of vibration mechanical damage and biological spoilage processes, as well as data privacy and security. By dynamically generating geofence parameter correction instructions and refrigeration power curve optimization logic based on the product risk value, the suppression of multi-physics field disturbances in the transportation environment and the control of biological stability are deeply correlated, achieving mechanical integrity assurance and adaptive closed-loop protection of microbial safety throughout the entire cold chain logistics of eggs. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart of an intelligent delivery scheduling method for egg products.

[0043] Figure 2 A flowchart for aggregation in federated learning.

[0044] Figure 3 This is a flowchart for risk calculation and decision-making.

[0045] Figure 4 This is a flowchart of the instruction execution. Detailed Implementation

[0046] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0047] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0048] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0049] Reference Figures 1-4 As an embodiment of the present invention, this embodiment provides an intelligent delivery scheduling method for egg products, comprising the following steps:

[0050] S1: Real-time collection of biological characteristic parameters of target eggs, and feature encryption processing through edge computing nodes to form privacy data packets.

[0051] S1.1: Real-time acquisition of biological characteristic parameters of target eggs, including the temperature dependence function of egg white viscoelasticity coefficient with egg age and the basic proliferation rate of Salmonella.

[0052] Specifically, during egg product scheduling, the current egg temperature and egg age are monitored in real time. The viscoelastic coefficient of egg white is measured using a rheometer. At the current egg temperature, shear stress (e.g., 0.1–10 Pa shear stress) is applied and the strain response is recorded to obtain the viscoelastic coefficient value of egg white. Samples at different egg ages are measured repeatedly (e.g., samples from the same batch of eggs on storage days 3 / 6 / 9 / 12) to construct a dataset of the temperature dependence function of the viscoelastic coefficient of egg white as a function of egg age.

[0053] The basal proliferation rate of Salmonella was determined using a rapid microbial detection instrument: samples were collected from the surface or contents of egg products, cultured under constant temperature conditions, and the rate of change of Salmonella colony count over time was recorded (example: record the number of Salmonella colonies every 1 hour), and the basal proliferation rate of Salmonella was calculated.

[0054] It should be noted that the temperature-dependent function dataset is composed of sample subsets of different egg products and egg ages. Each subset contains all pairs of egg product temperature and egg white viscoelastic coefficient values ​​measured at a given egg product and egg age, forming a temperature-dependent function dataset containing temperature-viscoelastic coefficient data sets corresponding to egg ages.

[0055] S1.2: Extract the feature vector of the temperature dependence function through edge computing nodes, perform homomorphic encryption operation on the feature vector to generate encrypted feature vector, and bind the Salmonella basic proliferation rate with the encrypted feature vector to form a privacy data packet.

[0056] At the edge nodes, load the temperature-dependent function dataset of the egg white viscoelastic coefficient as a function of egg age. For the temperature-viscoelastic coefficient data set corresponding to each egg age in the temperature-dependent function dataset, use the least squares method to fit the linear regression equation, with egg temperature as the independent variable and egg white viscoelastic coefficient as the dependent variable. Calculate the slope and intercept values ​​of the temperature-viscoelastic coefficient data set, and calculate the Pearson correlation coefficient between egg temperature and egg white viscoelastic coefficient in the temperature-viscoelastic coefficient data set.

[0057] Arrange the calculation results (slope value, intercept value and Pearson correlation coefficient) of all temperature-viscoelastic coefficient data sets in ascending order of egg age, and combine them to form a multidimensional feature vector (example: number of dimensions = number of egg ages × 3).

[0058] Homomorphic encryption is performed on the multidimensional feature vector using the Paillier homomorphic encryption algorithm to generate a public key and a private key. The public key is used to encrypt each numerical element in the feature vector, and the private key is used for local storage to output the encrypted feature vector.

[0059] The Salmonella basal proliferation rate value is bound to the encrypted feature vector. The Salmonella basal proliferation rate is written into the plaintext field, and the encrypted feature vector is written into the binary field. A JSON object containing these two fields is generated to form a privacy data packet.

[0060] S2: Input the privacy data packet into the aggregation service to dynamically aggregate encrypted biological parameters from multiple egg suppliers and construct a global egg rheology knowledge graph and a microbial growth kinetic model.

[0061] S2.1: Receive privacy data packets from various vendors through the message queue service, and call the homomorphic encryption library to perform weighted average aggregation on the encrypted feature vectors.

[0062] The system receives multiple privacy data packets submitted by various vendors through a message queue service and processes each vendor's privacy data packet separately.

[0063] Specifically, the privacy data packets of each supplier are parsed sequentially, the basic proliferation rate value of Salmonella in each privacy data packet is extracted, and the average value is calculated as the basic proliferation rate value of Salmonella for each supplier. At the same time, the encrypted feature vector in the privacy data packets of each supplier is extracted.

[0064] The number of egg samples provided by each supplier in this aggregation task is obtained through the aggregation service, and the weight of each supplier is calculated.

[0065] The weight of each supplier is calculated as follows:

[0066] ;

[0067] In the formula, Weight for each supplier, For suppliers The number of egg samples provided The total number of suppliers participating in federated learning. The total sample size of all suppliers. For supplier parameters, For supplier indexing, For summation index.

[0068] The homomorphic encryption library is invoked to perform a weighted average aggregation of the encrypted feature vectors. The Paillier homomorphic encryption algorithm is used to apply the corresponding scaling operation to the encrypted feature vector of each vendor. The scaled encrypted vectors are then accumulated to generate an aggregated encrypted feature vector.

[0069] S2.2: The basic proliferation rate of Salmonella is weighted and fused according to the sample size of the supplier, and the output is a global egg rheology knowledge graph containing the mapping relationship of viscoelasticity-temperature-egg age.

[0070] When weighting and fusing the basic Salmonella proliferation rate value, the basic Salmonella proliferation rate value of each supplier is weighted according to the weight of the corresponding supplier, and the weighted results of each supplier are integrated by a linear combination method to obtain the global basic Salmonella proliferation rate value.

[0071] The Paillier private key is used to decrypt the aggregated encrypted feature vector, outputting the plaintext aggregated feature vector. The plaintext aggregated feature vector is then decomposed into multiple parameter groups in ascending order of egg age. Each parameter group contains three parameters: slope value, intercept value, and Pearson correlation coefficient (example: one parameter group for each egg age).

[0072] Based on the slope and intercept values ​​in the parameter group corresponding to each egg age, a linear equation is constructed with egg temperature as the independent variable and egg white viscoelastic coefficient as the dependent variable. The linear equations corresponding to all egg ages are integrated to form a global egg rheology knowledge graph containing the mapping relationship between viscoelasticity, temperature and egg age.

[0073] S2.3: Based on the weighted fusion of Salmonella baseline proliferation rate, construct a microbial growth kinetic model that includes the relationship between bacterial community proliferation rate and temperature response.

[0074] The microbial community proliferation rate-temperature response relationship in constructing the biological growth kinetics model is expressed as follows:

[0075] ;

[0076] In the formula, For temperature The bacterial community proliferation rate under the following conditions This represents the baseline growth rate of Salmonella globally. The preset temperature response coefficient (example: 0.02). This is the preset reference temperature (example: 25℃). For temperature.

[0077] It should be noted that, The preset temperature response coefficient was determined by isothermal growth of a standard Salmonella strain (example: the proliferation rate was measured under culture conditions of 20-30℃, and a conservative value of 0.02 was taken after calculation). The preset reference temperature is calibrated based on a combination of the upper limit of safe storage temperature for eggs and the commonly used benchmark temperature for microbial testing (example: a compromise is made based on the upper limit of safe storage temperature for eggs of 7℃ and the benchmark temperature of 25℃). =25℃).

[0078] S3: Based on the global egg rheology knowledge graph and microbial growth dynamics model, real-time road condition data, temperature and humidity distribution in the carriage and vibration spectrum are loaded into the digital twin sandbox, and the product risk value of the protein shear stress accumulation equation and the microbial community growth equation is solved in a coupled manner.

[0079] S3.1: Obtain the road bump index from the digital twin sandbox to extract real-time road condition data, and load the spatial gradient of the temperature and humidity distribution in the vehicle compartment and the energy distribution of the vibration spectrum.

[0080] By using the real-time road condition data interface of the digital twin sandbox, the vehicle vertical acceleration data sequence is obtained, the most recent acceleration sample value is extracted (example: 1 sample point per second within 60 seconds), and the standard deviation of the acceleration sample value is calculated as the road bump index.

[0081] The standard deviation of the sampled values ​​is used as the road bump index, and its expression is:

[0082] ;

[0083] In the formula, The road bump index, The number of sampling points. For the first One acceleration sample value, The average acceleration, For the sampling time window (example: fixed value 60). For sampling point index;

[0084] The average temperature and humidity of the top area (e.g., 80%-100% of the height) and the bottom area of ​​the carriage are obtained through a digital twin sandbox, and the vertical temperature gradient and vertical humidity gradient are calculated.

[0085] Wherein, the vertical temperature gradient is expressed as:

[0086] ;

[0087] In the formula, For vertical temperature gradient, For the height of the carriage, The average temperature of the bottom area of ​​the carriage. The average temperature of the top area of ​​the carriage. For temperature, For gradient, This is the top position. This is the bottom position. For the carriage.

[0088] The vertical humidity gradient is expressed as:

[0089] ;

[0090] In the formula, It is a vertical humidity gradient. The average humidity in the top area of ​​the carriage. The average humidity in the top area of ​​the carriage. Humidity.

[0091] The vibration energy index is calculated based on the ratio of the energy integral value of the 5-20Hz frequency band to the total energy integral value of the 0-100Hz frequency band obtained by the digital twin sandbox.

[0092] It should be noted that the energy integral value in the 5–20Hz frequency band is the total energy integral value in the 0–100Hz frequency band. Its setting is primarily based on the physical characteristics of eggs themselves and the distribution characteristics of the vibration spectrum of typical road transport. As a biological material, eggs' inherent frequencies are mainly distributed in the low-frequency range, especially in the 5–20Hz range, where resonance is prone to occur, leading to structural damage. Meanwhile, the vibration energy generated by vehicles traveling on common road surfaces is mostly concentrated in the 0–100Hz range, representing the overall vibration intensity.

[0093] S3.2: Calculate the cumulative value of protein shear stress based on the global egg rheology knowledge graph.

[0094] Based on a global egg rheology knowledge graph, input the transportation start timestamp to calculate the current egg age and obtain the real-time temperature of the egg in the compartment.

[0095] The viscoelasticity-temperature-egg age mapping relationship of the global egg rheology knowledge graph is invoked. Based on the current egg age, the slope and intercept values ​​of the corresponding linear equation are extracted. The real-time temperature of the vehicle compartment is input to calculate the viscoelastic coefficient of the egg white under the current environment. Combined with the vibration energy index, the cumulative shear stress of the protein is calculated. The expression for calculating the cumulative shear stress of the protein is:

[0096] ;

[0097] In the formula, This represents the cumulative shear stress of the protein. The slope value represents the viscoelastic coefficient of egg white. This represents the intercept value of the viscoelastic coefficient of egg white. As a vibration energy index, frequency band For cutting, This refers to a local location.

[0098] It should be noted that the vibration energy index The frequency range of 5-20Hz is a dangerous frequency band for egg resonance.

[0099] S3.3: Calculate the instantaneous growth rate of the microbial community based on the microbial growth kinetics model.

[0100] Based on the microbial growth kinetics model, the overall average temperature and relative humidity of the carriage were obtained, and the global Salmonella basal proliferation rate was extracted.

[0101] The microbial growth kinetics model is used to calculate the instantaneous growth rate of the microbial community by incorporating the global Salmonella baseline growth rate and the overall average temperature of the carriage.

[0102] The instantaneous growth rate of the bacterial community is calculated using the following expression:

[0103] ;

[0104] In the formula, The instantaneous growth rate of the bacterial community. The average temperature of the entire carriage. Humidity influence coefficient (example: 0.02). The average relative humidity of the entire carriage. This is the average value.

[0105] It should be noted that when When, application The humidity correction threshold (85%) was determined by a combination of egg rheological yield strength and microbial growth boundary condition determination scheme. The egg rheological yield strength was determined by shear stress to determine its structural failure critical point, while the microbial growth boundary was determined by the proliferation kinetics of Salmonella under different humidity gradients to determine the explosive proliferation critical point (example: in a 70%-95% humidity gradient culture, 85% is the critical point for bacterial population explosive proliferation). The risk weighting coefficient (1.5) was set by a conservative parameter widely adopted in scientific literature and industry consensus in the field of food microbiological safety.

[0106] S3.4: Based on the coupling relationship between the cumulative protein shear stress and the instantaneous growth rate of the bacterial community, output the product risk value.

[0107] The product risk value is expressed as follows:

[0108] ;

[0109] In the formula, This represents the product risk value.

[0110] The product risk value is normalized to 0-100, and the expression is:

[0111] ;

[0112] In the formula, This is the normalized product risk value. This is the preset maximum risk threshold.

[0113] It should be noted that, The preset maximum risk threshold is determined through statistical analysis of historical transportation data (example: analyzing the risk values ​​of 100,000 historical transportation tasks recorded in the digital twin sandbox, calculating the 99th percentile value of 0.95, and rounding it to the egg safety standard value of 1.0 as the maximum threshold). Normalization uses linear scaling. > Forced =100.

[0114] S4: If the product risk value does not exceed the dynamic threshold, execute the basic transportation status.

[0115] S4.1: Based on the critical value of egg product breakage strength, a basic threshold is set, and a dynamic threshold is generated by superimposing the correction amount of vibration spectrum energy mutation rate and temperature and humidity deviation coefficient, and adjusting the dynamic threshold sensitivity according to the transportation time.

[0116] Using the preset critical experimental value of egg product rupture strength, the basic threshold is obtained by calling the egg product physical property database interface. Based on the standard value of protein shear failure strength measured by rheometer, the preset basic threshold is output (example value: 0.8).

[0117] Based on real-time data from the digital twin sandbox, the total energy value of two adjacent time windows (example: time window length is 10 seconds) in the vibration spectrum energy distribution data is obtained, and the energy mutation rate is calculated (example: the total energy value of the vibration spectrum in the previous time window is 150, the total energy value of the vibration spectrum in the current time window is 180 → mutation rate = (180-150) / 150 = 0.2).

[0118] Based on the vertical temperature gradient and vertical humidity gradient values, preset temperature gradient reference values ​​(example: preset value is 1.0℃ / m) and humidity gradient reference values ​​(example: preset value is 10.0%RH / m). By converting the absolute value of the vertical temperature gradient to the ratio of the temperature gradient reference value, and adding the absolute value of the vertical humidity gradient to the ratio of the humidity gradient reference value, the temperature and humidity deviation coefficient is calculated (example: if the vertical temperature gradient value is -1.2℃ / m and the vertical humidity gradient value is 6%RH / m, then the temperature and humidity deviation coefficient is 1.2 / 1.0+6 / 10.0=1.8).

[0119] By reading the task records in the transportation task management system, the transportation start timestamp record value is read, and the current timestamp record value is obtained by calling the real-time clock service interface.

[0120] Perform a time difference calculation on the start timestamp record value and the current timestamp record value, output the raw time difference, perform a unit conversion operation, and generate the current transportation duration;

[0121] The logarithmic gain function is used to calculate the basic sensitivity coefficient (preset value: 0.05) and the current transportation time, and the adjusted sensitivity coefficient is output.

[0122] It should be noted that the preset critical experimental value for egg product burst strength was set by using a rheometer to conduct shear tests on various egg product samples, based on the typical vibration and temperature and humidity conditions that eggs can withstand in the transportation environment, combined with their inherent physical properties.

[0123] S4.2: Based on the energy mutation rate and the preset energy mutation correction coefficient, calculate the energy risk increment through risk weight allocation operation (Example: Energy mutation rate 0.2 → Energy risk increment 0.2 × 0.1 = 0.02).

[0124] Based on the temperature and humidity deviation coefficient, the preset temperature and humidity deviation correction coefficient, and the adjusted sensitivity coefficient, the environmental risk increment is calculated through multi-level weight allocation operation (Example: temperature and humidity deviation coefficient 1.8, adjusted sensitivity 0.0477 → environmental risk increment 1.8×0.05×0.0477≈0.0043).

[0125] Based on a preset basic threshold, energy risk increments and environmental risk increments are superimposed. A dynamic threshold is synthesized through risk accumulation and normalized. The dynamic threshold is then converted to a percentage scale and the normalized dynamic threshold is output.

[0126] If the normalized product risk value does not exceed the normalized dynamic threshold, the current vehicle control parameters are used to maintain the navigation path and cooling power curve unchanged, and the basic transportation plan is executed using the control signal maintenance method.

[0127] It should be noted that the basic transportation plan is a pre-set plan for task initialization. It is formulated based on publicly available route data and safety temperature control standards, generated through a multi-objective optimization algorithm, and verified through batch risk simulation using historical data in a digital twin sandbox. This plan serves as a pre-verified reliable benchmark and is executed as long as the real-time risk does not exceed the threshold, thereby ensuring that the transportation process maintains optimal efficiency while ensuring safety.

[0128] The preset energy mutation correction coefficient is calibrated by measuring the maximum strain energy that the egg can absorb before it breaks and analyzing the energy attenuation law of vibration transmitted from the carriage to the egg; the temperature and humidity deviation correction coefficient (0.05) is calibrated based on the microbial water activity response characteristics and is generated by arithmetic averaging of the temperature gradient sensitive factor (example: 0.025 / ℃) and the humidity gradient sensitive factor (example: 0.025 / %RH).

[0129] S5: If the product risk value exceeds the dynamic threshold, generate a hyperoxia exposure path optimization instruction.

[0130] S5.1: Calculate the oxygen enrichment benefit weighting coefficient of forest road sections based on the magnitude of the product risk value exceeding the dynamic threshold, and generate liquid nitrogen rapid cooling trigger conditions including the bacterial community proliferation rate threshold and rapid cooling temperature.

[0131] The risk offset is calculated by performing a difference operation between the normalized product risk value and the normalized dynamic threshold. The risk exceedance ratio is generated by the ratio between the risk offset and the normalized dynamic threshold.

[0132] Based on the proportion of risk exceeding the threshold, a multiplication operation is performed using a preset oxygen enrichment benefit gain coefficient to generate the oxygen enrichment benefit weight coefficient for forest road sections.

[0133] Using the global Salmonella baseline proliferation rate in the microbial growth kinetics model, a risk amplification operation is performed to generate a microbial community proliferation rate threshold. A preset rapid cooling temperature value (set according to the technical specifications of liquid nitrogen refrigeration equipment) is loaded, and the output liquid nitrogen rapid cooling trigger condition includes the microbial community proliferation rate threshold and the rapid cooling temperature value.

[0134] It should be noted that the preset oxygen enrichment benefit gain coefficient was calibrated through an egg product microbial inhibition experiment (example: based on data fitting of the inhibitory effect of oxygen concentration on Salmonella in forest road sections, the gain coefficient was set to 0.8).

[0135] S5.2: Encapsulate the oxygen enrichment benefit weight coefficient of forest road sections and the liquid nitrogen rapid cooling trigger condition into a parsable JSON command, and generate geofence parameter correction command based on the oxygen enrichment benefit weight coefficient of forest road sections.

[0136] Specifically, the oxygen enrichment benefit weighting coefficient of forest road sections and the liquid nitrogen rapid cooling trigger condition are encapsulated into parsable JSON instructions. Based on the oxygen enrichment benefit weighting coefficient of forest road sections, the preset geofence priority value is loaded (example: set according to navigation path planning algorithm parameters), a linear mapping operation is performed, and the geofence parameter correction instruction is output, which includes the geofence priority adjustment value.

[0137] S5.3: Integrates JSON commands and geofence correction commands to generate high-oxygen exposure path optimization commands.

[0138] By adopting the homomorphic encrypted data channel protocol of edge computing nodes, parsable JSON commands and geofence parameter correction commands are integrated to generate high oxygen exposure path optimization command data packets that meet the encrypted transmission requirements of edge computing nodes.

[0139] The encrypted downlink channel of the edge computing node is used to perform transmission operations on the high oxygen exposure path optimization instruction data packet. The high oxygen exposure path optimization instruction is injected into the real-time instruction processing queue of the edge computing node through the write interface of the instruction queue of the vehicle edge computing node.

[0140] S6: Send the high oxygen exposure path optimization command to the vehicle edge computing node to dynamically correct the geofence parameters of the navigation path and adjust the cooling power curve.

[0141] S6.1: Transmit high oxygen exposure path optimization instructions to the vehicle edge computing node via an encrypted communication link, receive high oxygen exposure path optimization instructions and store them in the instruction cache area.

[0142] An encrypted communication link based on the TLS 1.3 protocol is used to transmit high oxygen exposure path optimization instructions to the vehicle edge computing node, and the high oxygen exposure path optimization instructions are stored in the instruction cache area after being received.

[0143] S6.2: Use the homomorphic decryption interface of the edge computing node to perform decryption operation on the high oxygen exposure path optimization instruction, and write the decrypted executable instruction into the real-time instruction processing queue of the vehicle control.

[0144] The Paillier homomorphic decryption interface of the vehicle edge computing node is used to perform decryption operations on the high oxygen exposure path optimization instructions in the instruction buffer.

[0145] The decrypted, parsable JSON commands and geofence parameter correction commands are written into the real-time command processing queue of the vehicle control system.

[0146] S6.3: Analyze the forest area road segment weight coefficient and liquid nitrogen rapid cooling trigger condition in the high oxygen exposure route optimization instruction, correct the geofencing parameters based on the forest area road segment weight coefficient, and generate the optimized transportation route trajectory through the route planning algorithm.

[0147] Extract the oxygen enrichment benefit weight coefficient, bacterial proliferation rate threshold and rapid cooling temperature of forest road sections from parsable JSON instructions, and extract the preset geofence basic priority from geofence parameter correction instructions.

[0148] Based on the extracted oxygen enrichment benefit weight coefficient of forest road sections, the dynamic priority parameters of geofence are multiplied by the extracted preset geofence basic priority to generate geofence dynamic priority parameters.

[0149] Based on the dynamic priority parameters of the geofence, the path planning database is called to write the predefined dynamic priority parameters of the forest area geofence in the database, and the updated dynamic priority parameters of the geofence are generated.

[0150] A weighted multi-objective path planning algorithm is adopted, which manages the search path process by maintaining open and closed lists, and calculates the cost of each extended path node in real time based on the updated geofence dynamic priority parameters and real-time road network data.

[0151] By iteratively selecting the path node with the lowest overall cost from the open list for expansion and using relaxation operations to update the cost estimates of adjacent nodes, a globally optimal path from the starting point to the destination is constructed, achieving the best balance between travel time and egg safety benefits, serving as the final optimized transportation path trajectory. S6.4: Based on the liquid nitrogen rapid cooling trigger condition, the power output curve of the refrigeration equipment is dynamically adjusted according to the temperature difference-power response relationship, and the optimized transportation path trajectory and refrigeration power execution status are fed back to the digital twin sandbox.

[0152] Based on the microbial growth kinetics model, the instantaneous growth rate of the bacterial community is monitored in real time. When the instantaneous growth rate of the bacterial community is greater than or equal to the bacterial community proliferation rate threshold, the MODBUS-RTU protocol is invoked to set the cooling temperature to the rapid cooling temperature. Based on the linear response relationship between cooling power and temperature difference, the cooling power execution status is calculated.

[0153] Using the data update interface of the digital twin sandbox, the optimized transportation route trajectory and cooling power execution status are fed back to the digital twin sandbox.

[0154] This embodiment also provides a computer device applicable to the intelligent delivery scheduling method for egg products, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the intelligent delivery scheduling method for egg products as proposed in the above embodiment.

[0155] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0156] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the intelligent delivery scheduling method for egg products as proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0157] In summary, this invention dynamically aggregates multi-source encrypted biological parameters using a federated learning engine and simultaneously constructs a global rheology-microbiology knowledge graph. This accurately captures the spatiotemporal coupling characteristics of the biomechanical properties and microbial activity of eggs during transportation, improving the accuracy of quality risk quantification and the reliability of transportation strategies. By loading real-time road condition spectra and temperature and humidity distribution gradients into a digital twin sandbox, the product risk value of the protein shear stress accumulation equation and the microbial community growth kinetic equation is solved, enhancing the robustness of collaborative prediction of vibration-induced mechanical damage and biological spoilage processes, as well as data privacy and security. By dynamically generating geofence parameter correction instructions and refrigeration power curve optimization logic based on the product risk value, the invention deeply correlates the suppression of multi-physics disturbances in the transportation environment with biological stability control, achieving adaptive closed-loop protection for the mechanical integrity and microbial safety of the entire egg cold chain logistics chain.

[0158] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An intelligent egg distribution scheduling method, characterized in that: The method comprises the steps of: Real-time acquisition of biological characteristic parameters of target egg products, and feature encryption processing by edge computing nodes to form privacy data packets; Receive each supplier's privacy data packet through the message queue service, and call the homomorphic encryption library to perform weighted average aggregation on the encrypted feature vector; Based on the global egg rheology knowledge graph and the microbial growth kinetics model, load the real-time road condition data, the temperature and humidity distribution in the carriage, and the vibration frequency spectrum in the digital twin sandbox, and solve the product of the protein shear stress accumulation equation and the bacterial growth equation; If the product risk value does not exceed the dynamic threshold, execute the basic transportation state; If the product risk value exceeds the dynamic threshold, generate high-oxygen exposure path optimization instructions; The high-oxygen exposure path optimization instructions are sent to the vehicle-mounted edge computing node to dynamically correct the geographic fence parameters of the navigation path and adjust the refrigeration power curve.

2. The intelligent egg distribution scheduling method of claim 1, wherein: The feature encryption processing comprises the following specific steps: Real-time acquisition of the temperature-dependent function of the egg white viscoelastic coefficient with egg age in the biological characteristic parameters of the target egg products and the basic proliferation rate of Salmonella; Extract the feature vector of the temperature-dependent function through the edge computing node, perform homomorphic encryption operation on the feature vector to generate an encrypted feature vector, bind the basic proliferation rate of Salmonella with the encrypted feature vector to form a privacy data packet.

3. The intelligent egg distribution scheduling method of claim 1, wherein: The dynamic aggregation of encrypted biological parameters of multiple egg product suppliers constructs a global egg rheology knowledge graph and a microbial growth kinetics model, which comprises the following specific steps: Receive each supplier's privacy data packet through the message queue service, and call the homomorphic encryption library to perform weighted average aggregation on the encrypted feature vector; Weighted fusion of the basic proliferation rate of Salmonella according to the sample size of the supplier, and output of the global egg rheology knowledge graph containing the viscoelasticity-temperature-egg age mapping relationship; Based on the weighted fused basic proliferation rate of Salmonella, a microbial growth kinetics model containing the bacterial proliferation rate-temperature response relationship is constructed.

4. The intelligent egg distribution scheduling method of claim 1, wherein: The coupling solution comprises the following specific steps: Obtain the road bump index of the real-time road condition data extracted by the digital twin sandbox, load the spatial gradient of the temperature and humidity distribution in the carriage, and the vibration frequency spectrum energy distribution; Based on the global egg rheology knowledge graph, calculate the protein shear stress accumulation value, and based on the microbial growth kinetics model, calculate the instantaneous growth rate of the bacterial population; Based on the coupling relationship between the protein shear stress accumulation value and the instantaneous growth rate of the bacterial population, output the product risk value.

5. The intelligent egg distribution scheduling method of claim 1, wherein: The dynamic threshold refers to the setting of a basic threshold based on the critical value of the egg breaking strength, the correction amount of the vibration frequency spectrum energy mutation rate and the temperature and humidity deviation coefficient, and the adjustment of the dynamic threshold sensitivity according to the transportation time to generate the dynamic threshold.

6. The intelligent egg distribution scheduling method of claim 1, wherein: The high-oxygen exposure path optimization instructions are as follows: Based on the amplitude of the product risk value exceeding the dynamic threshold, calculate the oxygen enrichment benefit weight coefficient of the forest road section, and generate the liquid nitrogen rapid cooling trigger condition containing the bacterial proliferation rate threshold and the rapid cooling temperature; Encapsulate the oxygen enrichment benefit weight coefficient of the forest road section and the liquid nitrogen rapid cooling trigger condition into a parseable JSON instruction, and generate a geographic fence parameter correction instruction according to the oxygen enrichment benefit weight coefficient mapping of the forest road section; Integrate the JSON instruction and the geographic fence correction instruction to generate the high-oxygen exposure path optimization instruction.

7. The intelligent egg distribution scheduling method of claim 1, wherein: The high-oxygen exposure path optimization instruction is sent to the vehicle-mounted edge computing node, and the specific steps are as follows, The high-oxygen exposure path optimization instruction is transmitted to the vehicle-mounted edge computing node through an encrypted communication link, the high-oxygen exposure path optimization instruction is received and stored in an instruction cache area; The homomorphic decryption interface of the edge computing node is used to perform decryption operation on the high-oxygen exposure path optimization instruction, and the decrypted executable instruction is written into the real-time instruction processing queue of the vehicle-mounted control system. 8.The intelligent egg distribution scheduling method of claim 1, wherein: The dynamic correction, the specific steps are as follows, The forest area road segment weight coefficient and the liquid nitrogen rapid cooling trigger condition in the high-oxygen exposure path optimization instruction are analyzed, the forest area road segment weight coefficient is used to correct the geographic fence parameters, and the optimized transportation path trajectory is generated through a path planning algorithm; Based on the liquid nitrogen rapid cooling trigger condition, the refrigeration equipment power output curve is dynamically adjusted according to the temperature difference-power response relationship, and the optimized transportation path trajectory and the refrigeration power execution state are fed back to the digital twin sandbox. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to realize the steps of the egg intelligent distribution scheduling method of any one of claims 1-8.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the egg intelligent distribution scheduling method of any one of claims 1-8.

Images