A method and system for intelligent scheduling of goods in multiple temperature zones in cold chain logistics

Abstract

This invention relates to the field of logistics scheduling technology, specifically disclosing a method and system for intelligent scheduling of goods in multi-temperature zones for cold chain logistics. The method involves acquiring multi-source dynamic datasets; constructing a virtual mapping body that includes the coupling relationship between the thermodynamic decay characteristics of each temperature zone and vehicle kinematic constraints; using the virtual mapping body to simulate future road network congestion events in advance, generating cold energy decay curves and remaining cold energy budget values ​​under different congestion scenarios; employing a multi-objective combined optimization algorithm based on the remaining cold energy budget value and the congestion scenario to pre-calculate a set of candidate scheduling strategies, including route detour schemes and temperature zone pre-cooling intervention schemes; when actual congestion occurs, matching and selecting the optimal strategy for execution, and dynamically correcting the virtual mapping body based on real-time feedback data. This invention realizes a shift from passive response after congestion occurs to proactive pre-congestion planning, effectively avoiding the scrapping of goods due to insufficient cold energy estimation, and improving the reliability and economy of cold chain distribution.

Description

Technical Field

[0001] This invention relates to the field of logistics scheduling technology, specifically to a method and system for intelligent scheduling of goods in multiple temperature zones in cold chain logistics. Background Technology

[0002] Cold chain logistics, a crucial link in the transportation of high-value goods such as fresh agricultural products and pharmaceutical preparations, faces the core challenge of maintaining stable temperatures across different zones within the vehicle's compartment in complex road transport environments. Currently, scheduling methods for multi-temperature zone cold chain delivery primarily rely on a combination of GPS navigation and static route planning. This involves pre-setting the route based on historical road condition data before the delivery task begins, and acquiring real-time road condition information via onboard terminals during vehicle operation. When traffic congestion is detected ahead, manual intervention from the dispatch center or simple route replanning based on the system's remaining battery power is implemented.

[0003] The existing technology has the following shortcomings: Existing cold chain scheduling methods cannot perform coupled pre-simulation of vehicle thermodynamic state and kinematic constraints before congestion events occur. As a result, when faced with sudden road conditions, the lack of accurate quantification of the dynamic relationship between the remaining cold energy decay process and driving energy consumption leads to scheduling schemes that are often based on lagging, decoupled single-dimensional data. This makes it impossible to fundamentally avoid cargo quality failures caused by misjudgments in cold energy budget. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for intelligent scheduling of goods in multiple temperature zones in cold chain logistics, so as to solve the problems mentioned above.

[0005] The objective of this invention can be achieved through the following technical solutions: A method for intelligent scheduling of goods in multiple temperature zones in cold chain logistics includes the following steps: S1: Obtain real-time status data of current transport vehicles, environmental monitoring data along the route, and road network congestion probability prediction data for future periods to form a multi-source dynamic dataset; S2: Based on a multi-source dynamic dataset, a virtual mapping of the vehicle-environment-road network during the transportation process is constructed. The virtual mapping includes the coupling relationship between the thermodynamic decay characteristics of each temperature zone and the vehicle's kinematic constraints, specifically including: Real-time temperature and cooling energy consumption data of each temperature zone of the vehicle are extracted from a multi-source dynamic dataset, and combined with the cargo heat capacity parameters of the corresponding temperature zone, heat load characteristic values ​​of each temperature zone are generated. Based on the heat load characteristic value and the ambient solar radiation intensity and ambient temperature at the vehicle's real-time location, the dynamic heat exchange rate of each temperature zone at the current moment is calculated. Based on the dynamic heat exchange rate and real-time acceleration and gradient data obtained from the vehicle's CAN bus, a coupling constraint factor characterizing the relationship between thermodynamic decay and kinematic energy consumption is calculated. Based on coupling constraint factors and road network congestion probability prediction data, a virtual mapping of transport vehicles is constructed. S3: Using a virtual mapping body, we can simulate road network congestion events that may occur within a preset time window in advance. By using a deduction algorithm that couples thermodynamic decay and kinematic change, we can generate the cold energy decay curves of goods in each temperature zone under different congestion scenarios and the corresponding remaining cold energy budget values. S4: Based on the remaining cooling capacity budget and the congestion scenario, a multi-objective combined optimization algorithm is used to pre-calculate and generate a set of candidate scheduling strategies, including path detour schemes and temperature zone pre-cooling intervention schemes; S5: When real-time monitoring detects actual road network congestion, match the current actual state with the preset scenarios in the candidate scheduling strategy set, select the optimal strategy to execute, and dynamically correct the virtual mapping body based on real-time feedback data during the execution process to update the initial conditions for subsequent simulations.

[0006] As a further aspect of the present invention: the specific calculation process of the coupling constraint factor is as follows: Real-time acceleration is integrated over time to obtain velocity fluctuation curves, and slope data is filtered to obtain road longitudinal slope change sequences. The speed fluctuation curve and the road longitudinal slope change sequence are input into the energy consumption decomposition algorithm based on the principle of work-heat conversion, and the motion energy consumption component used to overcome driving resistance is separated from it. By performing a synchronous correlation analysis on the time axis between the motion energy consumption component and the dynamic heat exchange rate, an interaction influence coefficient characterizing the inverse relationship between cooling energy consumption and driving energy consumption is generated. The interaction influence coefficient is fused with the vehicle's current load state to obtain the coupling constraint factor.

[0007] As a further aspect of the present invention: S3 specifically includes: Based on the road network congestion probability prediction data, select multiple road segments with congestion probability exceeding the threshold within a preset time window as candidate events, and set at least two congestion duration scenarios for each candidate event. For each congestion duration scenario, the coupling constraint factor output by the virtual mapping body is superimposed with the current vehicle trajectory to calculate the cumulative increase of heat load in each temperature zone during the period when the vehicle is stationary in the congested section. Substitute the cumulative increase in heat load into the temperature rise recursive algorithm based on energy conservation, and iteratively calculate the temperature values ​​of each temperature zone at each time step until a complete temperature change sequence over time is formed, i.e., the cooling capacity decay curve. Extreme value analysis is performed on the cooling capacity decay curve to extract the time when the temperature of each temperature zone reaches the preset quality critical value. The corresponding time is compared with the time when the congestion ends to calculate the remaining cooling capacity budget value of each temperature zone under the corresponding congestion scenario.

[0008] As a further aspect of the present invention: the calculation of the remaining cooling capacity budget value for each temperature zone under the corresponding congestion scenario specifically includes: The second derivative of the cooling capacity decay curve is calculated to identify the characteristic points of curvature change on the curve, and the acceleration inflection point when the rate of temperature change in each temperature zone changes from slow to fast is located. The temperature value corresponding to the acceleration inflection point is compared with the preset quality critical value. If the corresponding temperature value is lower than the critical value, the moment when the temperature first reaches the critical value is searched along the curve and taken as the critical arrival time. Calculate the time difference between the congestion end time and the critical arrival time based on the congestion end time set in the congestion duration scenario; The time difference is weighted and integrated with the maximum cooling duration supported by the vehicle's current remaining battery power to generate the remaining cooling capacity budget value for each temperature zone.

[0009] As a further aspect of the present invention: S4 specifically includes: Extract the remaining cooling capacity margin value of each temperature zone under each congestion scenario from the remaining cooling capacity budget value, and extract the corresponding congestion duration and congestion location coordinates from the congestion scenario. Compare the remaining cooling capacity margin value with the congestion duration to generate the tolerable delay threshold for each temperature zone. Using the coordinates of the congestion location as the center, and based on the tolerable delay threshold and the current movement status of the vehicles, a reachable area is delineated on the electronic map, and all passable road segments within the reachable area are enumerated to generate multiple candidate detour routes. For each candidate detour route, the additional travel time corresponding to the route is calculated based on the length of the candidate detour route and the expected travel speed. The additional travel time is then substituted into the cooling capacity decay curve to deduce the precooling intensity and precooling start and end times required to maintain the temperature of each temperature zone, thus forming a temperature zone precooling intervention plan uniquely associated with each route. Multiple candidate detour routes and their associated pre-cooling intervention schemes in temperature zones are used as decision variables. An optimal selection strategy based on congestion distance is adopted to select a set of strategies that are not superior in terms of both cooling consumption and travel time, which are then used as the candidate scheduling strategy set.

[0010] As a further aspect of the present invention: the derivation of the precooling intensity and start and end times required to maintain the temperature of each temperature zone specifically includes: Based on the length of the candidate detour route and the expected travel speed, the entire route is discretized into a continuous sequence of time segments, with each time segment corresponding to an expected location and expected travel time. The expected driving time corresponding to each segment in the time segment sequence is accumulated to obtain the new driving time. The new driving time is then superimposed on the cooling capacity decay curve as the time axis increment to form an extended predicted temperature change trajectory. On the extended predicted temperature change trajectory, starting from the endpoint time, the process is reversed time segment by time segment to calculate the target temperature value that must be reached at the beginning of each time segment to ensure that the endpoint temperature does not exceed the limit. The target temperature value at the beginning of each time segment is compared with the actual ambient temperature of the corresponding time segment. The required cooling capacity per unit time for each time segment is obtained by calculating the difference. Then, the corresponding pre-cooling intensity and the start and end times of pre-cooling are deduced based on the performance parameters of the refrigeration unit.

[0011] As a further aspect of the present invention: S5 specifically includes: When real-time monitoring detects actual road network congestion, the actual location and duration of the congestion are extracted. The actual location and duration of the congestion are compared with the similarity of each preset congestion scenario in the candidate scheduling strategy set. The strategy corresponding to the preset scenario with the highest similarity is selected as the strategy to be executed. The strategy to be executed is executed, and during the execution process, the actual temperature change data of each temperature zone and the actual energy consumption data of the vehicle are continuously collected. The actual temperature change data is compared with the temperature change trajectory predicted in the strategy to be executed point by point to generate a temperature deviation sequence. The temperature deviation sequence was correlated with the actual environmental wind speed and solar radiation intensity data collected during the same period. The disturbance component caused by the sudden change in the environment and the system component caused by the deviation of the vehicle's own thermodynamic characteristics were separated from the temperature deviation sequence. The thermodynamic decay characteristic parameters in the virtual mapping volume are corrected according to the system components, and the corrected virtual mapping volume is used as the initial condition for the next simulation.

[0012] As a further aspect of the present invention: the step of correcting the thermodynamic decay characteristic parameters in the virtual mapping body according to the system components, and using the corrected virtual mapping body as the initial condition for the next deduction, specifically includes: Calculate the gradient direction of the system component in the parameter space of the virtual mapping volume; Based on the gradient direction, select the combination of parameters that are strongly correlated with the system components from the thermodynamic decay characteristic parameters of the virtual mapping volume, and determine the adjustment step size of each parameter to be corrected. Each parameter in the combination of parameters to be corrected is iteratively updated along the gradient direction according to the corresponding adjustment step size, so that the deviation between the simulated temperature value output by the updated virtual mapping volume and the actual observed value converges to within the preset threshold. The updated virtual mapping is stored as the initial condition for the next simulation.

[0013] A multi-temperature zone intelligent cargo scheduling system for cold chain logistics includes: The data acquisition module acquires real-time status data of the current transport vehicles, environmental monitoring data along the route, and road network congestion probability prediction data for future periods, forming a multi-source dynamic dataset; The virtual mapping construction module, based on a multi-source dynamic dataset, constructs a vehicle-environment-road network virtual mapping body for transport vehicles during their operation. The virtual mapping body includes the coupling relationship between the thermodynamic decay characteristics of each temperature zone and the vehicle's kinematic constraints. The advanced simulation module uses a virtual mapping body to simulate road network congestion events that may occur within a preset time window in the future. Through a deduction algorithm that couples thermodynamic decay and kinematic change, it generates the cold energy decay curves of goods in each temperature zone under different congestion scenarios and the corresponding remaining cold energy budget values. The strategy generation module, based on the remaining cooling capacity budget and the congestion scenario, uses a multi-objective combined optimization algorithm to pre-calculate and generate a set of candidate scheduling strategies, including path detour schemes and temperature zone pre-cooling intervention schemes. The execution correction module, when real-time monitoring shows that actual road network congestion occurs, matches the current actual state with preset scenarios in the candidate scheduling strategy set, selects the optimal strategy to execute, and dynamically corrects the virtual mapping body based on real-time feedback data during the execution process to update the initial conditions for subsequent simulations.

[0014] The beneficial effects of this invention are: (1) This invention constructs a virtual mapping body that includes the coupling relationship between thermodynamic decay and kinematic constraints, and uses it to simulate future road network congestion events in advance. It can pre-calculate the cooling capacity decay curve and the remaining cooling capacity budget value under different scenarios before the actual occurrence of congestion. Compared with the existing technology, which only responds passively after congestion occurs, this invention realizes the transformation from "post-event processing" to "pre-event simulation", which enables the scheduling system to have the ability to accurately predict the cooling capacity budget. It can increase the cooling capacity reserve in advance through pre-cooling intervention, effectively avoid the scrapping of goods due to insufficient cooling capacity estimation, and significantly reduce the in-transit loss rate of high-value goods such as fresh produce and medicine.

[0015] (2) During strategy execution, this invention dynamically corrects the thermodynamic decay characteristic parameters in the virtual mapping body based on real-time collected temperature deviation data, and uses the corrected virtual mapping body as the initial condition for subsequent deductions. This adaptive online learning mechanism enables the virtual mapping body to continuously approximate the actual thermophysical characteristics of the vehicle, eliminating prediction biases caused by factors such as vehicle aging, cargo loading differences, and sudden environmental changes. As the number of runs increases, the accuracy of the advanced simulation continuously improves, thereby making the subsequently generated candidate scheduling strategy set more realistic, realizing continuous optimization and self-evolution of scheduling decisions. Attached Figure Description

[0016] The invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system block diagram of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 As shown, this invention is an intelligent scheduling method for multi-temperature zone goods in cold chain logistics, comprising the following steps: S1: Obtain real-time status data of current transport vehicles, environmental monitoring data along the route, and road network congestion probability prediction data for future periods to form a multi-source dynamic dataset; S2: Based on a multi-source dynamic dataset, a virtual mapping body of vehicle-environment-road network is constructed for the transportation vehicle during its operation. The virtual mapping body includes the coupling relationship between the thermodynamic decay characteristics of each temperature zone and the vehicle kinematic constraints. S3: Using a virtual mapping body, we can simulate road network congestion events that may occur within a preset time window in advance. By using a deduction algorithm that couples thermodynamic decay and kinematic change, we can generate the cold energy decay curves of goods in each temperature zone under different congestion scenarios and the corresponding remaining cold energy budget values. S4: Based on the remaining cooling capacity budget and the congestion scenario, a multi-objective combined optimization algorithm is used to pre-calculate and generate a set of candidate scheduling strategies, including path detour schemes and temperature zone pre-cooling intervention schemes; S5: When real-time monitoring detects actual road network congestion, match the current actual state with the preset scenarios in the candidate scheduling strategy set, select the optimal strategy to execute, and dynamically correct the virtual mapping body based on real-time feedback data during the execution process to update the initial conditions for subsequent simulations.

[0020] In S1, real-time status data of current transport vehicles, environmental monitoring data along the route, and road network congestion probability prediction data for future periods are acquired to form a multi-source dynamic dataset, specifically including: During the vehicle startup and delivery task execution process, the on-board diagnostic system interface deployed on the transport vehicle is first used to read the data stream on the vehicle's CAN bus in real time at a preset sampling frequency, and extract the vehicle's current operating status data from it. The vehicle's current operating status data includes at least the independent temperature values ​​of each temperature zone, the real-time power consumption of the refrigeration unit, the vehicle's instantaneous speed, the vehicle's current position coordinates, and the on / off status of the access control doors of each temperature zone in the compartment.

[0021] At the same time, the vehicle-mounted communication terminal sends query requests containing the vehicle's current location and preset driving route to multiple meteorological service interfaces, and receives the returned environmental monitoring data along the route. The environmental monitoring data along the route includes at least the time-series predicted values ​​of ambient temperature, ambient humidity, solar radiation intensity, and wind speed and direction for each road section.

[0022] In addition, the vehicle-mounted communication terminal sends a query request containing the vehicle's preset driving route and current time to the traffic information service interface, and receives the road network congestion probability prediction data within the future preset time window. The road network congestion probability prediction data includes at least the congestion probability value of each road segment on the preset driving route, as well as the corresponding predicted congestion start time and predicted congestion duration.

[0023] Finally, the collected vehicle current operating status data, roadside environmental monitoring data, and road network congestion probability prediction data are aligned and cleaned according to a unified time axis, outliers and duplicates are removed, and the cleaned data is packaged into a data set with correlation to form the multi-source dynamic dataset.

[0024] In S2, based on a multi-source dynamic dataset, a virtual mapping of the vehicle-environment-road network during the transportation process is constructed. This virtual mapping includes the coupling relationship between the thermodynamic decay characteristics of each temperature zone and the vehicle's kinematic constraints, specifically: First, the real-time temperature values ​​of each temperature zone of the vehicle and the real-time energy consumption values ​​of the corresponding refrigeration units are extracted from the multi-source dynamic dataset. Simultaneously, the cargo heat capacity parameters for each temperature zone are obtained from pre-recorded data stored in the vehicle's onboard memory. These cargo heat capacity parameters are obtained by multiplying the total mass of the cargo loaded in the corresponding temperature zone by its specific heat capacity. The real-time temperature value, real-time energy consumption value, and cargo heat capacity parameters for each temperature zone are then input into a heat load characteristic value calculation step. This calculation step converts the real-time energy consumption value into a cooling capacity per unit time and performs correlation analysis with the cargo heat capacity parameters and the real-time temperature change rate, outputting a heat load characteristic value representing the heat that needs to be removed from that temperature zone at the current moment.

[0025] Secondly, the ambient solar radiation intensity and ambient temperature values ​​at the vehicle's current location are extracted from the multi-source dynamic dataset. The obtained heat load characteristic values, along with the ambient solar radiation intensity and ambient temperature values, are input into a dynamic heat exchange rate calculation step. This calculation step first calculates the solar radiation heat absorbed through the vehicle's side panels per unit time based on the ambient solar radiation intensity, then calculates the convective heat transfer heat conducted through the side panels based on the difference between the ambient temperature and the interior temperature of the vehicle. The solar radiation heat and convective heat transfer heat are added together and compared and fused with the heat load characteristic values ​​to output a dynamic heat exchange rate characterizing the speed of heat exchange between the vehicle's side panels and the external environment at the current moment.

[0026] Next, real-time acceleration and gradient values ​​are extracted from the vehicle's CAN bus data over a continuous time series. The real-time acceleration values ​​are integrated sequentially over time to obtain a speed fluctuation curve reflecting vehicle speed fluctuations. The real-time gradient values ​​are processed using a moving average filtering method to remove instantaneous fluctuations caused by uneven road surfaces, resulting in a road longitudinal slope variation sequence reflecting the trend of road longitudinal inclination. The speed fluctuation curve and the road longitudinal slope variation sequence are input into an energy consumption decomposition step based on the principle of work-heat conversion. This step first calculates the total driving power required for the vehicle to overcome rolling resistance, air resistance, and gradient resistance based on the speed and gradient values ​​at each moment. This total driving power is then integrated over time to obtain the kinetic energy consumption component used to overcome driving resistance. The kinetic energy consumption component and the dynamic heat exchange rate are aligned along the same time axis, and a cross-correlation analysis method is used to calculate the temporal correlation between them, generating an interaction coefficient. This coefficient characterizes the quantitative relationship of mutual constraint and attrition between the energy consumption of the cooling system and the energy consumption of driving the vehicle during operation. Finally, the current load state value of the vehicle is extracted from the multi-source dynamic dataset, and the interaction influence coefficient is used to perform multivariate regression calculation with the current load state value of the vehicle to output a coupling constraint factor. The coupling constraint factor is used to characterize the correlation strength between thermodynamic decay characteristics and vehicle kinematic constraints.

[0027] Finally, the coupling constraint factor calculated above and the road network congestion probability prediction data from the multi-source dynamic dataset are input into a virtual mapping body construction step. This step uses the coupling constraint factor as an association weight, and superimposes the congestion probability values ​​of each road segment in the road network congestion probability prediction data with this association weight to generate a thermal-kinematic coupling field distributed on the road network topology. This thermal-kinematic coupling field, along with the vehicle's current position coordinates, the heat load characteristic values ​​of each temperature zone, and the dynamic heat exchange rate, forms a virtual mapping body with a data association structure. This virtual mapping body reflects the dynamic mapping relationship between the vehicle's thermodynamic state in each temperature zone and the kinematic constraints of the road network ahead under the current driving condition.

[0028] In S3, a virtual mapping body is used to simulate potential road network congestion events within a preset time window. Through a deduction algorithm that couples thermodynamic decay with kinematic changes, it generates cooling capacity decay curves for goods in different temperature zones under various congestion scenarios, along with corresponding remaining cooling capacity budget values. Specifically, this includes: First, road network congestion probability prediction data is extracted from the multi-source dynamic dataset. This data includes the congestion probability value of each road segment along the vehicle's preset travel path, as well as the corresponding predicted congestion start time and predicted congestion duration. A congestion probability threshold of 75% is set. All road segments with a congestion probability value exceeding 75% are selected as candidate event road segments. For each candidate event road segment, at least two congestion duration scenarios are defined based on its predicted congestion duration. The duration of the first scenario is 0.8 times the predicted congestion duration, and the duration of the second scenario is 1.2 times the predicted congestion duration, to cover the uncertainty range of congestion duration.

[0029] Secondly, for each set congestion duration scenario, a coupling constraint factor is extracted from the virtual mapping volume, and the expected arrival time of the vehicle at the candidate event segment is extracted from the vehicle's current trajectory. The coupling constraint factor is multiplied by the vehicle's dwell time during congestion to calculate the heat load accumulation rate per unit time. This heat load accumulation rate is then multiplied again by the dwell time to obtain the heat load accumulation increment for each temperature zone during the vehicle's dwell time in the congested segment. This heat load accumulation increment represents the additional heat that needs to be removed from each temperature zone due to changes in cooling efficiency and the continuous effects of the external environment when the vehicle is stationary.

[0030] Next, the calculated cumulative heat load increment is input into a temperature rise recursive step based on the principle of energy conservation. This recursive step starts at the moment the vehicle enters the congested section and iterates using a preset time step (60 seconds) as the unit of measurement, calculating the temperature values ​​of each temperature zone at the end of each time step. Specifically, for each time step, the temperature rise caused by heat accumulation is calculated based on the starting temperature value and the cumulative heat load increment within that step. Then, the starting temperature value and the temperature rise value are added together to obtain the temperature value at the end of that step. The temperature value at the end of the previous time step is used as the starting temperature value for the next time step, and the above calculation process is repeated until the congestion end time is set according to the congestion duration scenario. This forms a temperature-time change sequence consisting of continuous time points and their corresponding temperature zone values, which is recorded as a cooling capacity decay curve.

[0031] Finally, the remaining cooling capacity budget value is calculated based on the cooling capacity decay curve, specifically including the following steps: The first step is to calculate the second derivative of the temperature value sequence on the cooling capacity decay curve. The second derivative reflects the rate of change of the slope at each point on the curve, i.e., the degree of accelerated temperature increase. The zero-crossing point where the second derivative changes from less than zero to greater than zero is identified, and the time corresponding to this zero-crossing point is positioned as the acceleration inflection point. The acceleration inflection point characterizes the critical time point when the rate of temperature change begins to change from a slow increase to a rapid increase.

[0032] The second step involves comparing the temperature value corresponding to the acceleration inflection point with a preset quality threshold. The preset quality threshold refers to the highest temperature value allowed for goods in each temperature zone to maintain their quality. For example, for a temperature zone of -18 degrees Celsius, the preset quality threshold is -15 degrees Celsius; for a temperature zone of 0 to 4 degrees Celsius, the preset quality threshold is 6 degrees Celsius. If the temperature value corresponding to the acceleration inflection point is lower than the preset quality threshold for that temperature zone, then starting from the acceleration inflection point, the search proceeds point by point along the cooling capacity decay curve in the direction of increasing time. The moment when the temperature value first reaches the preset quality threshold is recorded as the critical arrival time.

[0033] The third step is to obtain the congestion end time set in the congestion duration scenario, and calculate the time difference between the congestion end time and the critical arrival time. The time difference represents the remaining tolerable time before the temperature of the temperature zone reaches the quality critical value at the end of the congestion.

[0034] The fourth step involves extracting the vehicle's current remaining battery power from the vehicle status data and converting this remaining battery power into the maximum sustainable cooling duration based on the refrigeration unit's rated power. The calculated time difference is then weighted and fused with the maximum sustainable cooling duration. This weighting and fusion follows a minimum value principle: if the time difference is less than the maximum sustainable cooling duration, the time difference is used as the remaining cooling capacity budget value for each temperature zone; if the maximum sustainable cooling duration is less than the time difference, the maximum sustainable cooling duration is used as the remaining cooling capacity budget value for each temperature zone. This remaining cooling capacity budget value represents the maximum permissible congestion duration during which goods in each temperature zone can maintain quality within acceptable limits under the current congestion scenario.

[0035] In S4, based on the remaining cooling capacity budget and congestion scenario, a multi-objective combined optimization algorithm is used to pre-calculate and generate a candidate scheduling strategy set that includes path detour schemes and temperature zone pre-cooling intervention schemes, specifically including: First, extract the remaining cooling capacity budget value for each temperature zone generated in step S3, corresponding to the remaining cooling capacity budget value for each congestion duration scenario. This remaining cooling capacity budget value is the remaining cooling capacity budget value for that scenario. Simultaneously, extract the corresponding congestion duration and congestion location coordinates from the congestion duration scenario. For each congestion duration scenario, compare the remaining cooling capacity budget value for each temperature zone with the congestion duration. The comparison steps are as follows: for each temperature zone, subtract the congestion duration from the remaining cooling capacity budget value to obtain a difference. If the difference is positive, it indicates that there is still remaining cooling capacity in that temperature zone after the congestion ends, and the tolerable additional delay time is this difference. If the difference is negative, it indicates that the temperature zone has reached the quality critical value before the congestion ends, and the tolerable additional delay time is zero. Take the minimum tolerable additional delay time among all temperature zones as the tolerable delay threshold for the entire vehicle in that scenario. This threshold represents the maximum additional time a vehicle can spend after bypassing congested sections, provided that the quality of goods in all temperature zones does not exceed the standard.

[0036] Secondly, using the projection point of the congestion location coordinates on the map as the center and the product of the tolerable delay threshold and the vehicle's current speed as the radius, a circular area is delineated on the electronic map. This area is denoted as the reachable area. The reachable area represents the set of all geographical locations that the vehicle can reach within the tolerable delay threshold time. Subsequently, by calling the road network data of the electronic map, all possible combinations of road segments within the reachable area that can bypass the congested road segment from the vehicle's current location and rejoin the original path are enumerated. Each combination of road segments constitutes a candidate detour route, and the start point, end point, intermediate nodes, and path length of each candidate detour route are recorded.

[0037] Next, for each enumerated candidate detour route, the corresponding additional travel time is calculated. The method for calculating the additional travel time is as follows: obtain the path length of the candidate detour route from the electronic map, match the expected travel speed according to the road level of the route, divide the path length by the expected travel speed to obtain the expected travel time of the route; subtract the original expected travel time from the current position to the merging point after bypassing the congestion from the expected travel time to obtain the additional travel time. Subsequently, substitute the additional travel time into the cooling capacity decay curve generated in step S3, and perform a reverse recursive step to determine the pre-cooling intervention scheme required to maintain the temperature of each temperature zone, specifically including the following sub-steps: The first sub-step involves discretizing the candidate detour path into a continuous sequence of time segments at 100-meter intervals or at each intersection. Each time segment in the sequence corresponds to a target location, and the target travel time for that segment is calculated based on its length and the target travel speed.

[0038] The second sub-step involves summing the expected travel times of all segments in the time segment sequence to obtain the total additional travel time for the candidate detour route. This total additional travel time is then used as a time axis increment and superimposed onto the cooling capacity decay curve generated in step S3. Specifically, the starting time of the original cooling capacity decay curve remains unchanged, and the temperature value corresponding to each moment on the curve is shifted backward by the total additional travel time, thus forming a predicted temperature change trajectory stretched and extended along the time axis. This trajectory reflects the expected temperature changes over time in each temperature zone if the vehicle chooses this detour route.

[0039] The third sub-step involves performing a backward recursive calculation, starting from the endpoint time and proceeding in the reverse order of discretization, time-segment by time segment, along the extended predicted temperature change trajectory. The purpose of this backward recursive calculation is to determine the target temperature value that must be reached at the beginning of each time segment to ensure that the temperature in each temperature zone does not exceed a preset quality threshold when the vehicle reaches the endpoint time. The specific calculation process for the backward recursion uses the following mathematical formula: ; In mathematical calculation formulas, T K-1 This represents the target temperature value at the starting point of the (K-1)th time segment, in degrees Celsius. T K T represents the temperature value at the starting point of the Kth time segment. In the reverse recursion process, T K Given a known value, T is used in the first recursion. K Take the preset quality threshold value at the endpoint; This indicates the rated cooling power of the refrigeration unit per unit time, in kilowatts. This value is preset based on the nameplate parameters of the vehicle's refrigeration unit. This represents the expected travel time for the k-th time segment, in hours. Q K This represents the heat load per unit time caused by the intrusion of external environmental heat in the k-th time segment, in kilowatts. This value is determined based on the dynamic heat exchange rate calculated in step S2 and the ambient temperature and solar radiation intensity corresponding to the k-th time segment. This indicates the total heat capacity of the goods within the specified temperature range, expressed in kilowatt-hours per degree Celsius. This value is calculated based on the heat capacity parameters of the goods and their total mass.

[0040] Using the above formula, starting from the preset quality threshold at the end time, the target temperature value T at the beginning of each time segment is calculated sequentially backward. K-1This process continues until the current moment is reached, thus obtaining a target temperature change curve from the current moment to the endpoint.

[0041] The fourth sub-step involves comparing the target temperature value at the start of each time segment obtained through reverse recursion with the actual ambient temperature value at the start of that time segment. If the target temperature value is lower than the actual ambient temperature value T... K-1 This indicates that refrigeration needs to be activated within that time segment to maintain the temperature. Calculate the difference between the two values ​​and multiply it by the corresponding heat capacity of the goods for that time segment. Then divide by the expected travel time of that time segment. The required cooling capacity per unit time for that time segment is obtained. Based on the performance parameters of the refrigeration unit under different operating conditions, the cooling capacity per unit time is mapped to the corresponding percentage of cooling power output, which is the pre-cooling intensity. All continuous time segments with a pre-cooling intensity greater than zero are merged, and their start time is the starting point of the pre-cooling start-end time, and their end time is the ending point of the pre-cooling start-end time. Thus, a temperature zone pre-cooling intervention scheme uniquely associated with the current candidate detour path is formed. This scheme includes at least the time variation curve of the pre-cooling intensity and the times of pre-cooling start and stop.

[0042] Finally, the multiple candidate detour paths generated in the previous steps, along with the associated pre-cooling intervention schemes for each path, are collectively used as decision variables and input into an optimization step based on congestion distance ranking. This optimization step first constructs a two-dimensional decision space with cooling consumption and travel time as two optimization dimensions. Cooling consumption is obtained by integrating the pre-cooling intensity of each time segment in the pre-cooling intervention scheme over time; travel time is the additional travel time of the candidate detour paths. Subsequently, the coordinate position of each decision variable (i.e., each candidate detour path and its associated schemes) in the two-dimensional decision space is calculated. Using a non-dominated ranking method, all decision variables are divided into multiple frontier levels, where the decision variables in the first frontier level are the combinations of strategies that are not dominant in either cooling consumption or travel time, i.e., the Pareto optimal solution set. The output of all strategy combinations in the Pareto optimal solution set is used as a candidate scheduling strategy set for real-time matching and selection in subsequent steps when congestion occurs. The non-dominated ranking means that for any two strategies, if the cooling energy consumption and travel time of one strategy are both less than or equal to the other strategy, and at least one of them is strictly less, then the former dominates the latter. Strategies not dominated by any strategy are assigned to the first frontier. The congestion distance calculation means that for strategies within the same frontier, the distances between adjacent strategies in the dimensions of cooling energy consumption and travel time are summed; a larger distance indicates greater solution diversity.

[0043] In S5, when real-time monitoring detects actual road network congestion, the current actual state is matched with preset scenarios in the candidate scheduling strategy set. The optimal strategy is selected and executed, and the virtual mapping is dynamically corrected based on real-time feedback data during execution to update the initial conditions for subsequent simulations. Specifically, this includes: First, the vehicle navigation terminal monitors road conditions in real time along the vehicle's preset driving route. When actual congestion is detected ahead, the starting coordinates and estimated duration of the congestion are immediately obtained from the traffic information service interface. These coordinates and duration are used as a state vector to be matched. From the candidate scheduling strategy set generated in step S4, a scene feature vector corresponding to each preset congestion scenario is extracted. This scene feature vector includes at least the congestion location coordinates and congestion duration set for that scenario. The Euclidean distance between the state vector to be matched and each scene feature vector is calculated. This Euclidean distance is obtained by summing the squares of the differences between the corresponding components of the two vectors and then taking the square root. The preset congestion scenario with the smallest Euclidean distance is selected, and the candidate scheduling strategy associated with this scenario is used as the strategy to be executed. This strategy includes a candidate detour path and a pre-cooling intervention scheme associated with that path.

[0044] Secondly, the candidate detour paths and temperature zone pre-cooling intervention schemes in the strategy to be executed are sent to the driver via the in-vehicle human-machine interface. Simultaneously, the start and end times and pre-cooling intensity parameters in the pre-cooling intervention scheme are sent to the refrigeration unit controllers of each temperature zone, which automatically execute the pre-cooling operation. While the vehicle travels along the candidate detour paths, the actual temperature values ​​of each temperature zone and the actual energy consumption values ​​of the refrigeration units are continuously collected at a sampling frequency of once per second. Simultaneously, the predicted temperature change trajectory in the strategy to be executed is interpolated along the same time axis to correspond to the actual collection time. The actual temperature value at each moment is subtracted from the temperature value at the corresponding moment on the predicted temperature change trajectory to obtain the temperature deviation value at that moment. All temperature deviation values ​​at all moments are arranged in chronological order to form a temperature deviation sequence.

[0045] Furthermore, while executing the strategy to be executed, the actual ambient wind speed and actual solar radiation intensity values ​​at the vehicle's current location are continuously collected from the meteorological service interface, forming an ambient wind speed sequence and a solar radiation intensity sequence synchronized with the actual temperature collection time. The temperature deviation sequence is then subjected to correlation analysis with the ambient wind speed sequence and the solar radiation intensity sequence, respectively. The specific steps of the correlation analysis are as follows: calculate the covariance of the temperature deviation sequence and the ambient wind speed sequence, and then divide by the product of their standard deviations to obtain the correlation coefficient between temperature deviation and ambient wind speed; similarly, calculate the correlation coefficient between temperature deviation and solar radiation intensity. The portion of the temperature deviation sequence that is strongly correlated with wind speed (correlation coefficient absolute value greater than 0.6) and has a consistent trend is marked as a disturbance component caused by abrupt environmental changes; similarly, the portion of the temperature deviation sequence that is strongly correlated with solar radiation intensity and has a consistent trend is also marked as a disturbance component. The remaining portion after subtracting the disturbance components from the temperature deviation sequence is the system component caused by the deviation of the vehicle's own thermodynamic characteristics. The system component reflects the inherent deviation between the preset thermodynamic decay characteristics in the virtual mapping and the actual physical characteristics of the vehicle.

[0046] Finally, the virtual mapping volume is dynamically corrected based on the system components, specifically including the following sub-steps: The first sub-step involves calculating the gradient direction of the system component within the parameter space of the virtual mapping body. The parameter space comprises multiple thermodynamic decay characteristic parameters within the virtual mapping body, including at least the overall heat transfer coefficient of the vehicle insulation layer, the actual energy efficiency ratio of the refrigeration unit, and the cold air leakage rate for each temperature zone. The specific steps for calculating the gradient direction are as follows: For each thermodynamic decay characteristic parameter, a small increment is applied to the current value of the parameter. The incremented parameter is substituted into the virtual mapping body to recalculate the simulated temperature value, obtaining the change in the simulated temperature value. This change in simulated temperature value is divided by the small increment to obtain the partial derivative of the parameter. The partial derivatives of all parameters are combined into a vector, which represents the gradient direction of the system component within the parameter space. The gradient direction indicates the trend and relative magnitude of the adjustment required for each parameter to eliminate the system component.

[0047] The second sub-step involves selecting the top three parameters with the highest absolute values ​​from all thermodynamic decay characteristic parameters based on the absolute values ​​of the partial derivatives along the gradient direction. These parameters are then used as the combination of parameters to be corrected that are strongly correlated with the system components. For each parameter to be corrected, the absolute value of its partial derivative is divided by the sum of the absolute values ​​of all partial derivatives to obtain a normalization coefficient. This normalization coefficient is then multiplied by a preset maximum adjustment step size (which is 5% of the parameter's current value) to obtain the parameter's adjustment step size. The sign of the adjustment step size is determined by the sign of the partial derivative. If the partial derivative is positive, the adjustment step size is negative, indicating that adjustment should be made in the direction of decreasing the parameter value; if the partial derivative is negative, the adjustment step size is positive, indicating that adjustment should be made in the direction of increasing the parameter value.

[0048] The third sub-step involves updating each parameter in the parameter combination to be corrected along the gradient direction according to its corresponding adjustment step size, resulting in a set of updated parameter values. These updated parameter values ​​are then substituted into the virtual mapping volume to recalculate the simulated temperature value. The simulated temperature value is then compared with the actual temperature value collected at the next moment, and the deviation between the two is calculated. If the absolute value of this deviation is less than a preset correction threshold (0.2 degrees Celsius), the iteration stops. If the deviation is still greater than or equal to the correction threshold, the first to third sub-steps are repeated, starting from the previously updated parameter values, until the deviation between the simulated and actual temperature values ​​converges to within the correction threshold.

[0049] The fourth sub-step involves storing the virtual mapping body that has been updated with parameters and meets the deviation convergence condition as the new virtual mapping body for the current vehicle, and using it as the initial condition for the next execution of steps S2 to S5.

[0050] Please see Figure 2 As shown, a multi-temperature zone intelligent cargo scheduling system for cold chain logistics includes: The data acquisition module acquires real-time status data of the current transport vehicles, environmental monitoring data along the route, and road network congestion probability prediction data for future periods, forming a multi-source dynamic dataset; The virtual mapping construction module, based on a multi-source dynamic dataset, constructs a vehicle-environment-road network virtual mapping body for transport vehicles during their operation. The virtual mapping body includes the coupling relationship between the thermodynamic decay characteristics of each temperature zone and the vehicle's kinematic constraints. The advanced simulation module uses a virtual mapping body to simulate road network congestion events that may occur within a preset time window in the future. Through a deduction algorithm that couples thermodynamic decay and kinematic change, it generates the cold energy decay curves of goods in each temperature zone under different congestion scenarios and the corresponding remaining cold energy budget values. The strategy generation module, based on the remaining cooling capacity budget and the congestion scenario, uses a multi-objective combined optimization algorithm to pre-calculate and generate a set of candidate scheduling strategies, including path detour schemes and temperature zone pre-cooling intervention schemes. The execution correction module, when real-time monitoring shows that actual road network congestion occurs, matches the current actual state with preset scenarios in the candidate scheduling strategy set, selects the optimal strategy to execute, and dynamically corrects the virtual mapping body based on real-time feedback data during the execution process to update the initial conditions for subsequent simulations.

[0051] The working principle of this invention is as follows: First, a multi-source dynamic dataset is formed by collecting real-time vehicle status, along-route environmental monitoring, and future road network congestion probability prediction data. Then, based on this dataset, a virtual mapping body is constructed that includes the coupling relationship between the thermodynamic decay characteristics of each temperature zone and vehicle kinematic constraints. This virtual mapping body is used to simulate potential future road network congestion events in advance. A deduction algorithm that couples thermodynamic decay and kinematic changes generates the cold energy decay curves of goods in each temperature zone and the corresponding remaining cold energy budget values ​​under different congestion scenarios. Subsequently, based on the remaining cold energy budget values ​​and the congestion scenarios, a multi-objective combined optimization algorithm is used to pre-calculate and generate a candidate scheduling strategy set that includes route detour schemes and temperature zone pre-cooling intervention schemes. Finally, when actual road network congestion is detected in real time, the current actual state is matched with the preset scenario in the candidate scheduling strategy set, and the optimal strategy is selected for execution. At the same time, the virtual mapping body is dynamically corrected based on real-time feedback data during the execution process to update the initial conditions for subsequent deductions.

[0052] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A method for intelligent scheduling of goods in multiple temperature zones in cold chain logistics, characterized in that, Includes the following steps: S1: Obtain real-time status data of current transport vehicles, environmental monitoring data along the route, and road network congestion probability prediction data for future periods to form a multi-source dynamic dataset; S2: Based on a multi-source dynamic dataset, a virtual mapping of the vehicle-environment-road network during the transportation process is constructed. The virtual mapping includes the coupling relationship between the thermodynamic decay characteristics of each temperature zone and the vehicle's kinematic constraints, specifically including: Real-time temperature and cooling energy consumption data of each temperature zone of the vehicle are extracted from a multi-source dynamic dataset, and combined with the cargo heat capacity parameters of the corresponding temperature zone, heat load characteristic values ​​of each temperature zone are generated. Based on the heat load characteristic value and the ambient solar radiation intensity and ambient temperature at the vehicle's real-time location, the dynamic heat exchange rate of each temperature zone at the current moment is calculated. Based on the dynamic heat exchange rate and real-time acceleration and gradient data obtained from the vehicle's CAN bus, a coupling constraint factor characterizing the relationship between thermodynamic decay and kinematic energy consumption is calculated. Based on coupling constraint factors and road network congestion probability prediction data, a virtual mapping of transport vehicles is constructed. S3: Using a virtual mapping body, we can simulate road network congestion events that may occur within a preset time window in advance. By using a deduction algorithm that couples thermodynamic decay and kinematic change, we can generate the cold energy decay curves of goods in each temperature zone under different congestion scenarios and the corresponding remaining cold energy budget values. S4: Based on the remaining cooling capacity budget and the congestion scenario, a multi-objective combined optimization algorithm is used to pre-calculate and generate a set of candidate scheduling strategies, including path detour schemes and temperature zone pre-cooling intervention schemes; S5: When real-time monitoring detects actual road network congestion, match the current actual state with the preset scenarios in the candidate scheduling strategy set, select the optimal strategy to execute, and dynamically correct the virtual mapping body based on real-time feedback data during the execution process to update the initial conditions for subsequent simulations.

2. The intelligent scheduling method for multi-temperature zone goods in cold chain logistics according to claim 1, characterized in that, The specific calculation process for the coupling constraint factor is as follows: Real-time acceleration is integrated over time to obtain velocity fluctuation curves, and slope data is filtered to obtain road longitudinal slope change sequences. The speed fluctuation curve and the road longitudinal slope change sequence are input into the energy consumption decomposition algorithm based on the principle of work-heat conversion, and the motion energy consumption component used to overcome driving resistance is separated from it. By performing a synchronous correlation analysis on the time axis between the motion energy consumption component and the dynamic heat exchange rate, an interaction influence coefficient characterizing the inverse relationship between cooling energy consumption and driving energy consumption is generated. The interaction influence coefficient is calculated by fusing it with the vehicle's current load status to obtain the coupling constraint factor.

3. The intelligent scheduling method for multi-temperature zone goods in cold chain logistics according to claim 1, characterized in that, S3 specifically includes: Based on the road network congestion probability prediction data, select multiple road segments with congestion probability exceeding the threshold within a preset time window as candidate events, and set at least two congestion duration scenarios for each candidate event. For each congestion duration scenario, the coupling constraint factor output by the virtual mapping body is superimposed with the current vehicle trajectory to calculate the cumulative increase of heat load in each temperature zone during the period when the vehicle is stationary in the congested section. Substitute the cumulative increase in heat load into the temperature rise recursive algorithm based on energy conservation, and iteratively calculate the temperature values ​​of each temperature zone at each time step until a complete temperature change sequence over time is formed, i.e., the cooling capacity decay curve. Extreme value analysis is performed on the cooling capacity decay curve to extract the time when the temperature of each temperature zone reaches the preset quality critical value. The corresponding time is compared with the time when the congestion ends to calculate the remaining cooling capacity budget value of each temperature zone under the corresponding congestion scenario.

4. The intelligent scheduling method for multi-temperature zone goods in cold chain logistics according to claim 3, characterized in that, The calculation of the remaining cooling capacity budget value for each temperature zone under the corresponding congestion scenario specifically includes: The second derivative of the cooling capacity decay curve is calculated to identify the characteristic points of curvature change on the curve, and the acceleration inflection point when the rate of temperature change in each temperature zone changes from slow to fast is located. The temperature value corresponding to the acceleration inflection point is compared with the preset quality critical value. If the corresponding temperature value is lower than the critical value, the moment when the temperature first reaches the critical value is searched along the curve and taken as the critical arrival time. Calculate the time difference between the congestion end time and the critical arrival time based on the congestion end time set in the congestion duration scenario; The time difference is weighted and integrated with the maximum cooling duration supported by the vehicle's current remaining battery power to generate the remaining cooling capacity budget value for each temperature zone.

5. The intelligent scheduling method for multi-temperature zone goods in cold chain logistics according to claim 1, characterized in that, S4 specifically includes: Extract the remaining cooling capacity margin value of each temperature zone under each congestion scenario from the remaining cooling capacity budget value, and extract the corresponding congestion duration and congestion location coordinates from the congestion scenario. Compare the remaining cooling capacity margin value with the congestion duration to generate the tolerable delay threshold for each temperature zone. Using the coordinates of the congestion location as the center, and based on the tolerable delay threshold and the current movement status of the vehicles, a reachable area is delineated on the electronic map, and all passable road segments within the reachable area are enumerated to generate multiple candidate detour routes. For each candidate detour route, the additional travel time corresponding to the route is calculated based on the length of the candidate detour route and the expected travel speed. The additional travel time is then substituted into the cooling capacity decay curve to deduce the precooling intensity and precooling start and end times required to maintain the temperature of each temperature zone, thus forming a temperature zone precooling intervention plan uniquely associated with each route. Multiple candidate detour routes and their associated pre-cooling intervention schemes in temperature zones are used as decision variables. An optimal selection strategy based on congestion distance is adopted to select a set of strategies that are not superior in terms of both cooling consumption and travel time, which are then used as the candidate scheduling strategy set.

6. The intelligent scheduling method for multi-temperature zone goods in cold chain logistics according to claim 5, characterized in that, The derivation of the precooling intensity and start and end times required to maintain the temperature in each temperature zone specifically includes: Based on the length of the candidate detour route and the expected travel speed, the entire route is discretized into a continuous sequence of time segments, with each time segment corresponding to an expected location and expected travel time. The expected driving time corresponding to each segment in the time segment sequence is accumulated to obtain the new driving time. The new driving time is then superimposed on the cooling capacity decay curve as the time axis increment to form an extended predicted temperature change trajectory. On the extended predicted temperature change trajectory, starting from the endpoint time, the process is reversed time segment by time segment to calculate the target temperature value that must be reached at the beginning of each time segment to ensure that the endpoint temperature does not exceed the limit. The target temperature value at the beginning of each time segment is compared with the actual ambient temperature of the corresponding time segment. The required cooling capacity per unit time for each time segment is obtained by calculating the difference. Then, the corresponding pre-cooling intensity and the start and end times of pre-cooling are deduced based on the performance parameters of the refrigeration unit.

7. The intelligent scheduling method for multi-temperature zone goods in cold chain logistics according to claim 1, characterized in that, S5 specifically includes: When real-time monitoring detects actual road network congestion, the actual location and duration of the congestion are extracted. The actual location and duration of the congestion are compared with the similarity of each preset congestion scenario in the candidate scheduling strategy set. The strategy corresponding to the preset scenario with the highest similarity is selected as the strategy to be executed. The strategy to be executed is executed, and during the execution process, the actual temperature change data of each temperature zone and the actual energy consumption data of the vehicle are continuously collected. The actual temperature change data is compared with the temperature change trajectory predicted in the strategy to be executed point by point to generate a temperature deviation sequence. The temperature deviation sequence was correlated with the actual environmental wind speed and solar radiation intensity data collected during the same period. The disturbance component caused by the sudden change in the environment and the system component caused by the deviation of the vehicle's own thermodynamic characteristics were separated from the temperature deviation sequence. The thermodynamic decay characteristic parameters in the virtual mapping volume are corrected according to the system components, and the corrected virtual mapping volume is used as the initial condition for the next simulation.

8. The intelligent scheduling method for multi-temperature zone goods in cold chain logistics according to claim 7, characterized in that, The step of correcting the thermodynamic decay characteristic parameters in the virtual mapping volume based on the system components, and using the corrected virtual mapping volume as the initial condition for the next deduction, specifically includes: Calculate the gradient direction of the system component in the parameter space of the virtual mapping volume; Based on the gradient direction, select the combination of parameters that are strongly correlated with the system components from the thermodynamic decay characteristic parameters of the virtual mapping volume, and determine the adjustment step size of each parameter to be corrected. Each parameter in the combination of parameters to be corrected is iteratively updated along the gradient direction according to the corresponding adjustment step size, so that the deviation between the simulated temperature value output by the updated virtual mapping volume and the actual observed value converges to within the preset threshold. The updated virtual mapping is stored as the initial condition for the next simulation.

9. A multi-temperature zone intelligent dispatching system for cold chain logistics, characterized in that, A method for implementing a multi-temperature zone intelligent scheduling method for cold chain logistics as described in any one of claims 1-8 includes: The data acquisition module acquires real-time status data of the current transport vehicles, environmental monitoring data along the route, and road network congestion probability prediction data for future periods, forming a multi-source dynamic dataset; The virtual mapping construction module, based on a multi-source dynamic dataset, constructs a vehicle-environment-road network virtual mapping body for transport vehicles during their operation. The virtual mapping body includes the coupling relationship between the thermodynamic decay characteristics of each temperature zone and the vehicle's kinematic constraints. The advanced simulation module uses a virtual mapping body to simulate road network congestion events that may occur within a preset time window in the future. Through a deduction algorithm that couples thermodynamic decay and kinematic change, it generates the cold energy decay curves of goods in each temperature zone under different congestion scenarios and the corresponding remaining cold energy budget values. The strategy generation module, based on the remaining cooling capacity budget and the congestion scenario, uses a multi-objective combined optimization algorithm to pre-calculate and generate a set of candidate scheduling strategies, including path detour schemes and temperature zone pre-cooling intervention schemes. The execution correction module, when real-time monitoring shows that actual road network congestion occurs, matches the current actual state with preset scenarios in the candidate scheduling strategy set, selects the optimal strategy to execute, and dynamically corrects the virtual mapping body based on real-time feedback data during the execution process to update the initial conditions for subsequent simulations.

Images