Cold-chain logistics distribution system based on refrigerator

Abstract

The invention discloses a cold-chain logistics distribution system based on a refrigerator, and relates to the field of cold-chain logistics, the cold-chain logistics distribution system comprises a data management library, a distribution scheduling module and an optimization processing module, the data management library is used for storing and managing logistics data, the distribution scheduling module is connected with the data management library and is used for receiving and processing distribution tasks from the data management library, and the optimization processing module is connected with the data management library. Scheduling personnel distribution according to the distribution task, and updating the information of the distribution personnel currently executing the distribution task to a data management library; and the optimization processing module is connected with the data management library and the distribution scheduling module, and is used for calculating an optimal distribution scheme through an intelligent optimization algorithm according to the distribution task, the urban map and the vehicle position information, generating navigation information and sending the navigation information to the distribution scheduling module. According to the method, the cold-chain logistics distribution path optimization problem is solved, the time window constraint is considered, the method can be flexibly applied to different types of cold-chain logistics distribution scenes, and continuity and stability of cold-chain logistics distribution are ensured.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of cold chain logistics, and specifically to a cold chain logistics distribution system based on refrigerators. Background Art

[0002] A freezer, also called a freezer, is a professional storage tool used to store various foods or other items that need to be frozen. It maintains an appropriate low-temperature environment inside the box through a refrigeration system. The working principle is to use the evaporation and condensation of the refrigerant to reduce the temperature of the object. The refrigerant absorbs heat when evaporating, which reduces the temperature inside the freezer; then, the refrigerant is compressed into a high-pressure liquid in the compressor, releasing the heat previously absorbed; then, the refrigerant dissipates heat through the condenser and becomes liquid again; finally, the refrigerant reduces the pressure and temperature through the throttling component and enters the evaporator cycle again. This process is repeated continuously, thereby keeping the temperature inside the freezer lower than the room temperature.

[0003] Cold chain logistics distribution refers to a systematic project that refrigerated and frozen foods are always kept in a specified low temperature environment in all links from production, storage, transportation, sales to consumption, in order to ensure food quality and reduce food loss. There are many modes of cold chain logistics distribution. The inevitable problem in each mode is to deliver the goods to the destination in a timely and punctual manner. Reasonable arrangement of delivery vehicles and planning of reasonable delivery routes are important indicators to ensure product quality. Based on the existing technology, the route optimization calculation in cold chain logistics distribution still has the following defects: 1. The intelligently generated route information data is too similar to the navigation software, usually only considering the distance between the starting point and the destination, and auxiliary consideration of time, which can easily lead to delivery vehicles having difficulty in delivering goods to the destination on time; 2. Lack of real-time optimization of routes. Urban traffic is significantly different from high-speed traffic. In urban traffic, the road conditions are complicated, and it is more difficult for delivery vehicles to pass through. The waiting time will be expected to be longer, resulting in the inability to deliver in time. Summary of the invention

[0004] Based on this, the purpose of the present invention is to provide a cold chain logistics distribution system based on refrigerators to solve the technical problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: A cold chain logistics distribution system based on a refrigerator, comprising a data management library, a distribution scheduling module, and an optimization processing module, wherein the data management library is used to store and manage logistics data, and the logistics data includes distribution tasks, city maps, distribution personnel information, navigation information, and vehicle location information; The delivery scheduling module is connected to the data management library, and is used to receive and process delivery tasks from the data management library, and schedule personnel for delivery according to the delivery tasks, and update the information of the delivery personnel currently executing the delivery tasks to the data management library; The optimization processing module is connected to the data management library and the delivery scheduling module, and is used to calculate the optimal delivery plan through an intelligent optimization algorithm based on the delivery task, the city map and the vehicle location information, and generate navigation information to send to the delivery scheduling module; The delivery scheduling module also includes a plurality of mobile phone terminals, which are used to bind with the delivery vehicles to obtain vehicle location information in real time, and to receive and display navigation information required for the current delivery task.

[0006] Preferably, the delivery scheduling module specifically adopts the following steps: Receive delivery tasks, city maps and vehicle location information sent by the data management library; Marks are formed on the city map based on the starting point location, target point location and vehicle location information of the delivery task; Through intelligent optimization algorithms, it finds the best delivery route for delivery vehicles and generates navigation information; The navigation information is sent to the delivery scheduling module and displayed in real time on the mobile phone. The delivery person can directly view it and start the delivery task.

[0007] Preferably, the intelligent optimization algorithm divides the traversable path between the starting point and the target point of the delivery task into multiple sections from position i to position j, sets two indicators, namely, the traffic efficiency and the path priority, and calculates the section with the highest transfer probability from position i to position j, and composes the optimal navigation information through multiple sections with the highest transfer probability; wherein, the calculation formula of the transfer probability is as follows: ; in, : represents the transfer probability of the kth delivery vehicle from position i to position j at time t; :express The path priority from position i to position j at time; : represents the traffic efficiency from position i to position j; and : are two parameters, representing the relative importance of path priority and traffic efficiency respectively; : represents the set of paths that the k-th delivery vehicle can currently choose; : represents the path s that can be selected for all k, that is, , calculate the sum of the products of the path priority and the traffic efficiency from i to s; this sum is used as the denominator to normalize the probability to ensure that the sum of the probabilities of all selectable paths is 1.

[0008] Preferably, the traffic efficiency includes giving priority to spacious sections, giving priority to right turns, avoiding red lights, and avoiding congestion; Prioritizing wide roads means giving priority to roads accessible to delivery vehicles, and avoiding narrow roads and prohibited roads; Prioritize right turn: when generating navigation information, give priority to right turn in detour sections to avoid waiting for the left turn indicator light; Red light avoidance is specifically calculated by measuring the waiting time of the delivery vehicle at the intersection and detour time Numerical comparison of traffic efficiency; Congestion avoidance specifically reduces the proportion of congested sections in the generated navigation information.

[0009] Preferably, the path priority is used to update the path priority according to the path found by the kth delivery vehicle, and to strengthen the priority on the high-quality path, so as to guide the subsequent k+1th delivery vehicle to select a better path. The specific formula is as follows: ; in, : The path priority from position i to position j at time; : represents the path priority decay coefficient, which controls the speed at which the path priority decays over time; : represents the path priority from position i to position j at time t; : The sum of the path priorities marked by all delivery vehicles k on the path ij; m: represents the total number of delivery vehicles; : represents the path priority marked by the kth delivery vehicle on the path ij; : Attenuation coefficient Controls the decay of path priority, Indicates the proportion of path priority reserved; : The remaining amount of the path priority after attenuation at time t.

[0010] Preferably, the path priority is in a state of continuous superposition, so the attenuation coefficient must be set , to avoid the local path being marked as the optimal solution due to the accumulation of path priorities; The path priority is inversely proportional to the path length. The shorter the path, the more priority it will be marked with, so that the shortest path is more likely to be selected by the delivery vehicle.

[0011] Preferably, the two indicators of traffic efficiency and path priority are important factors in determining the transfer probability. The traffic efficiency and path priority are also complementary. When only one of the traffic efficiency and path priority exists, the navigation information finally generated will not be the optimal route. pass and The two coefficients assign weights to both traffic efficiency and path priority. Specifically, if When , the delivery vehicle is more inclined to follow the path of the historical delivery vehicle; when When the delivery vehicle is in a hurry, it is more inclined to choose the route with the shortest time.

[0012] Preferably, the delivery task also includes the latest acceptance time, the type of goods to be delivered, and the volume of goods to be delivered; The latest acceptance time limits the time point at which the delivery vehicle arrives at the target location at the latest. Therefore, a time constraint is set to limit the generated navigation information. The specific formula is as follows: ; in, : represents the time constraint for the delivery vehicle to perform the delivery task; : Indicates the time when the delivery vehicle actually arrives at the starting point; : Indicates the latest acceptance time.

[0013] Preferably, the city map updates the city traffic information of the delivery task execution in real time through the network, including the length of the road section, the location of the intersection, the time of the traffic light, the road condition information, the prohibition indication, and the congestion condition.

[0014] In summary, the present invention mainly has the following beneficial effects: The present invention optimizes the complex conditions of urban roads. By optimizing the road section selection, red light avoidance, necessary detours, congestion handling and other conditions of the distribution route, each generated distribution route can be within the time constraint. By adjusting the parameters of the path priority of the traffic efficiency, a new feasible path can be quickly found to ensure the continuity and stability of cold chain logistics distribution. The optimization of the delivery path is the core issue. The system has the characteristics of distributed computing, decentralized control, and indirect communication between distributed individuals, and performs well in solving complex path planning and optimization problems. In the cold chain logistics distribution system, the intelligent algorithm can gradually find the optimal or nearly optimal delivery path by simulating the traffic efficiency and path priority of the delivery vehicles during movement; At the same time, this system is combined with the existing real-time city map. On the one hand, it reduces the computing burden of the system itself. On the other hand, it can be flexibly applied to different types of cold chain logistics distribution scenarios, such as multiple distribution centers, multiple vehicles, and distribution problems with time window constraints. By reasonably setting the key parameters of the algorithm and optimizing the algorithm based on the specific needs of the cold chain logistics distribution system, it can significantly improve distribution efficiency, reduce distribution costs, and improve customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is the overall structural framework diagram of the system of the present invention; Figure 2 This is a framework diagram of the specific processing steps of the intelligent algorithm of the present invention.

[0016] Description of the accompanying drawings: 10. Data management library; 20. Distribution scheduling module; 30. Optimization processing module; 21. Mobile phone terminal. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0018] like Figure 1 and Figure 2 As shown, a cold chain logistics distribution system based on a refrigerator includes a data management library 10, a distribution scheduling module 20, and an optimization processing module 30. The data management library 10 is used to store and manage logistics data. The logistics data includes distribution tasks, city maps, distribution personnel information, navigation information, vehicle location information, refrigerator status information, access control system, etc. The distribution scheduling module 20 is connected to the data management library 10 and is used to receive and process distribution tasks from the data management library 10. The distribution tasks include the latest acceptance time, the type of distribution goods, the volume of distribution goods, the starting point location, and the target point location; The delivery personnel information includes the duty shift information, the list of duty personnel on the day, the corresponding vehicle number of the delivery person, etc., so that the delivery scheduling module 20 can arrange the delivery tasks according to the latest duty list; The distribution personnel are dispatched according to the distribution tasks, and the distribution personnel information currently executing the distribution tasks is updated to the data management library 10; the data management library 10 is also used to generate distribution reports and reports to check and eliminate the deficiencies of the subsequent intelligent algorithms and the loopholes of the distribution system, so that each distribution task can be processed by the system on time and in a timely manner; The optimization processing module 30 is connected to the data management library 10 and the distribution scheduling module 20, and is used to calculate the optimal distribution plan through an intelligent optimization algorithm according to the distribution task, the city map and the vehicle location information, and generate navigation information to send to the distribution scheduling module 20; The delivery scheduling module 20 also includes a plurality of mobile phone terminals 21, which are used to bind with the delivery vehicles to obtain the vehicle location information in real time, and receive and display the navigation information required for the current delivery task; the mobile phone terminal 21 is placed on the delivery vehicle with the corresponding number, and the mobile phone terminal 21 is connected to the delivery scheduling module 20 through the network. The system dispatches the delivery task to the delivery vehicle that is closest to the starting point and is in a resting state according to the information of the order task. After the delivery personnel receive the delivery task, they confirm it through the mobile phone terminal 21, and the delivery task starts immediately; The distribution scheduling module 20 specifically adopts the following steps: Receive the delivery tasks, city maps and vehicle location information sent by the data management library 10; Marks are formed on the city map based on the starting point location, target point location and vehicle location information of the delivery task; Through intelligent optimization algorithms, it finds the best delivery route for delivery vehicles and generates navigation information; The navigation information is sent to the delivery scheduling module 20 and displayed in real time on the mobile phone 21. The delivery person can directly view it and start to execute the delivery task.

[0019] The intelligent optimization algorithm divides the traversable path between the starting point and the target point of the delivery task into multiple sections from position i to position j, sets two indicators, namely, traffic efficiency and path priority, and calculates the section with the highest transfer probability from position i to position j. The optimal navigation information is formed through multiple sections with the highest transfer probability. The calculation formula of the transfer probability is as follows: ; in, : represents the transfer probability of the kth delivery vehicle from position i to position j at time t; :express The path priority from position i to position j at time; : represents the traffic efficiency from position i to position j; and : are two parameters, representing the relative importance of path priority and traffic efficiency respectively; : represents the set of paths that the k-th delivery vehicle can currently choose; : represents the path s that can be selected for all k, that is, , calculate the sum of the products of the path priority and the traffic efficiency from i to s; this sum is used as the denominator to normalize the probability to ensure that the sum of the probabilities of all selectable paths is 1.

[0020] The traffic efficiency includes giving priority to wide sections, giving priority to right turns, avoiding red lights, and avoiding congestion; Prioritizing wide roads means giving priority to roads accessible to delivery vehicles, and avoiding narrow roads and prohibited roads; Prioritize right turn: when generating navigation information, give priority to right turn in detour sections to avoid waiting for the left turn indicator light; Red light avoidance is specifically calculated by measuring the waiting time of the delivery vehicle at the intersection and detour time Numerical comparison of traffic efficiency; Congestion avoidance specifically reduces the proportion of congested sections in the generated navigation information.

[0021] The path priority is used to update the path priority according to the path found by the kth delivery vehicle, strengthen the priority on the high-quality path, and guide the subsequent k+1th delivery vehicle to choose a better path. The specific formula is as follows: ; in, : The path priority from position i to position j at time; : represents the path priority decay coefficient, which controls the speed at which the path priority decays over time; : represents the path priority from position i to position j at time t; : The sum of the path priorities marked by all delivery vehicles k on the path ij; m: represents the total number of delivery vehicles; : represents the path priority marked by the kth delivery vehicle on the path ij; : Attenuation coefficient Controls the decay of path priority, Indicates the proportion of path priority reserved; : The remaining amount of the path priority after attenuation at time t.

[0022] Path priorities are constantly increasing, so a decay coefficient must be set , to avoid the local path being marked as the optimal solution due to the accumulation of path priorities; The path priority is inversely proportional to the path length. The shorter the path, the more priority it will be marked with, so that the shortest path is more likely to be selected by the delivery vehicle.

[0023] The two indicators of traffic efficiency and path priority are important factors in determining the transfer probability. Traffic efficiency and path priority are also complementary. When only one of the traffic efficiency and path priority exists, the navigation information finally generated will not be the optimal route. pass and The two coefficients assign weights to both traffic efficiency and path priority. Specifically, if When , the delivery vehicle is more inclined to follow the path of the historical delivery vehicle; when When the delivery vehicle is in a hurry, it is more inclined to choose the route with the shortest time.

[0024] The latest acceptance time limits the time point at which the delivery vehicle arrives at the destination at the latest. Therefore, a time constraint is set to restrict the generated navigation information. The formula is as follows: ; in, : represents the time constraint for the delivery vehicle to perform the delivery task; : Indicates the time when the delivery vehicle actually arrives at the starting point; : Indicates the latest acceptance time.

[0025] The city map updates the city traffic information of the delivery task in real time through the network, including the length of the road section, the location of the intersection, the time of the traffic light, the road condition information, the prohibition signs, and the congestion situation.

[0026] It should be noted that in this embodiment, the whole process of executing the delivery task is simulated by the algorithm using the city map of N cities. When a customer places an order from the system, the system processes it and generates a delivery task, which includes the amount, type of goods, volume of goods, starting location of the goods, destination location of delivery, and delivery time required before the system delivers the delivery task to any delivery person; The system can provide basic urban road section information to the distribution scheduling module 20 intelligent algorithm through the city map updated in real time through the network, and the length of the road section and the location of the intersection. The starting position and the target location to be delivered are obtained from the city map. The delivery vehicle is equipped with a mobile phone terminal 21, which is bound to the delivery vehicle to obtain the vehicle location information in real time. At this time, the city map displays the starting position, the target location to be delivered, and the three coordinate points of the delivery vehicle. The goal of the intelligent algorithm is to find the optimal route by reaching the starting position through the coordinate points of the delivery vehicle and delivering it to the target location within the specified time.

[0027] It should be further explained that the optimal route is not the shortest route. Through the prediction and calculation of intelligent algorithms, the goods can be delivered to the target location in the shortest time by reasonably detouring and avoiding long parking and waiting while ensuring the delivery time. By dividing the traversable path between the starting point and the target point of the delivery task into multiple sections from position i to position j, setting two indicators, traffic efficiency and path priority, and calculating the section with the highest transfer probability from position i to position j, the optimal navigation information is formed by multiple sections with the highest transfer probability.

[0028] Through the systematic simulation of various possible routes, we find the optimal delivery route for the current task. In the simulation, we found that in the section from a certain position i to a certain position j, the time waiting for the red light is too long, and the waiting takes up a lot of the prescribed time. Based on this, we set the traffic efficiency as a reference indicator in the algorithm through optimization, so that the algorithm will not be overly dependent on the "short distance" information provided by the city map, and the traffic efficiency is better. In this indicator, through multiple simulations, at intersections with long red light times, the delivery vehicle directly chooses to turn left-turn-right-turn right to return to the main road, which takes less time than turning right-turn-left-turn and less time than waiting at the red light. Although this is just a special case, we set heuristic factors for the algorithm to optimize the final output navigation information, including giving priority to wide sections, giving priority to right turns, avoiding red lights, and avoiding congestion. Prioritize wide roads means giving priority to roads that delivery vehicles can pass through, and avoiding narrow roads and prohibited roads; prioritize right turns means giving priority to right turns in detours during the generation of navigation information, and avoiding waiting for the left turn indicator; avoid red light by calculating the waiting time for the delivery vehicle to arrive at the intersection and detour time Numerical comparison of traffic efficiency; congestion avoidance specifically reduces the proportion of congested sections in the generated navigation information.

[0029] After setting the reference indicator of traffic efficiency, the more simulation times the algorithm outputs, the more similar the results are. After analysis, it is found that the algorithm is overly dependent on the path priority indicator. After the path marked with path priority is used multiple times, the algorithm's weight on the path priority is biased. Based on this, the path priority is adjusted. and traffic efficiency Set parameter factors α and β, and continuously simulate and adjust the values ​​of parameter factors α and β to achieve stable results of the system algorithm output. In the algorithm, α and β represent the relative importance of path priority and traffic efficiency respectively. α is used to represent the relative importance of the signal priority marked during the movement of the delivery vehicle. The larger the α value is, the more delivery vehicles tend to choose the paths that other delivery vehicles have passed through, the stronger the cooperation between delivery vehicles is, and the faster the algorithm converges, but it is easy to fall into the local optimal solution; if the α value is too small, it will improve the random characteristics and global search ability of the algorithm, but will slow down the convergence speed of the algorithm; β is used to indicate the relative importance of heuristic information, reflecting the importance of heuristic information in the process of route selection when the delivery vehicle is moving. The larger the β value is, the more the delivery vehicle relies on heuristic information when selecting a route, the faster the convergence speed may be, and it is easier to find the global optimal solution. If the β value is too small, the delivery vehicle will make little or no use of the traffic efficiency, and the algorithm will become a pure random search, making it difficult to find the optimal solution.

[0030] Through multiple simulations, the α value is reduced to reduce the impact of the path priority in the algorithm, so that the delivery vehicle does not rely too much on the historical paths of other delivery vehicles when selecting a path, and it is assigned between 0.3 and 0.5; the β value is increased to increase the weight of the heuristic information in the algorithm, so that the delivery vehicle relies more on the heuristic information when selecting a path, and it is assigned between 0.6 and 0.8; α and β are not fixed. According to the complexity of the actual urban road sections, adjusting α and β can achieve the balance of the algorithm's exploration ability and development ability; Path priority The above path reinforcement formula is continuously updated. In the algorithm, the path priority will volatilize over time, which means that even if a path is selected multiple times, the path priority on it will not increase indefinitely. The attenuation mechanism ensures that the algorithm can avoid falling into the local optimal solution and has a certain exploration ability; path length and path priority: The path priority is usually inversely proportional to the path length, which means that shorter paths will mark more letter priorities, thereby attracting more delivery vehicles to choose. This mechanism helps the algorithm find the global optimal solution. The attenuation coefficient ρ is set and the speed at which the path priority decays over time is controlled by changing the value of ρ.

[0031] In addition, all generated navigation information needs to set a reference value in the predicted delivery time, otherwise the algorithm will only be responsible for calculating the route and neglect the control of time. Based on this, time constraints must be set. The latest acceptance time in the time constraint data comes from the delivery task generated by the system. The actual arrival time of the delivery vehicle at the starting point is based on the positioning of the mobile phone terminal 21. After the delivery scheduling module 20 in the system obtains the optimal navigation information through the algorithm, it sends the real-time data to the mobile phone 21, receives and displays the navigation information required for the current delivery task; the delivery personnel start to perform the delivery task through the navigation information. Before the delivery vehicle arrives at the target location, if an emergency situation occurs, such as a traffic accident causing congestion, the mobile phone 21 will actively apply for re-planning. At the same time, due to the long waiting time on the usual road section, the system will also recognize that there is a problem with the delivery vehicle, and actively calculate the new delivery route, and resend it to the delivery scheduling module 20.

[0032] Similarly, if there is an emergency with the delivery vehicle or delivery personnel, the system can calculate the delivery route to replace the problem vehicle based on the location of other delivery vehicles, as described above, to ensure on-time delivery as much as possible.

[0033] The working principle of the present invention is: The intelligent optimization algorithm divides the traversable path between the starting point and the target point of the delivery task into multiple sections from position i to position j, sets two indicators, namely, traffic efficiency and path priority, and calculates the section with the highest transfer probability from position i to position j. The optimal navigation information is formed by multiple sections with the highest transfer probability. It is used in combination with the real-time city map in the existing technology, that is, the real-time city map provides traffic information in real time, and the intelligent algorithm of the system performs segmented optimization through the reference route provided by the real-time city map.

[0034] By systematically simulating various possible routes, we can find the optimal delivery route for the current task, avoid relying on the "short distance" information provided by the city map, and set the traffic efficiency. With path priority Two indicators for path priority and traffic efficiency Set the parameter factors α and β, where α and β represent the relative importance of path priority and traffic efficiency respectively. The larger the α value, the more likely the delivery vehicle is to choose the path that other delivery vehicles have passed through, the stronger the collaboration between delivery vehicles, and the faster the algorithm converges, but it is easy to fall into the local optimal solution. If the α value is too small, it will improve the random characteristics and global search ability of the algorithm, but will slow down the convergence of the algorithm. β is used to indicate the relative importance of heuristic information, reflecting the importance of heuristic information in the process of route selection when the delivery vehicle is moving; the larger the β value is, the more the delivery vehicle relies on heuristic information when selecting a route, the faster the convergence speed may be, and it is easier to find the global optimal solution; if the β value is too small, the delivery vehicle will use little or no traffic efficiency, the algorithm will become a pure random search, and it will be difficult to find the optimal solution; and the values ​​of the parameter factors α and β are continuously simulated to achieve stable results of the system algorithm output.

[0035] Path priority The above path reinforcement formula is continuously updated. In the algorithm, the path priority will volatilize over time, which means that even if a path is selected multiple times, the path priority on it will not increase indefinitely. The attenuation mechanism ensures that the algorithm can avoid falling into the local optimal solution and has a certain exploration ability; path length and path priority: The path priority is usually inversely proportional to the path length, which means that shorter paths will mark more letter priorities, thereby attracting more delivery vehicles to choose. This mechanism helps the algorithm find the global optimal solution. The attenuation coefficient ρ is set and the speed at which the path priority decays over time is controlled by changing the value of ρ.

[0036] The latest acceptance time limits the time point at which the delivery vehicle arrives at the destination location, so setting a time constraint limits the generated navigation information.

[0037] The above embodiments are only for illustrating the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the present invention.

Claims

1. A cold chain logistics distribution system based on a refrigerator, comprising a data management library (10), a distribution scheduling module (20), and an optimization processing module (30), characterized in that: The data management library (10) is used to store and manage logistics data, wherein the logistics data includes delivery tasks, city maps, delivery personnel information, navigation information, and vehicle location information; The delivery scheduling module (20) is connected to the data management library (10) and is used to receive and process delivery tasks from the data management library (10), schedule personnel for delivery according to the delivery tasks, and update the information of the delivery personnel currently executing the delivery tasks to the data management library (10); The optimization processing module (30) is connected to the data management library (10) and the delivery scheduling module (20), and is used to calculate the optimal delivery plan through an intelligent optimization algorithm based on the delivery task, the city map and the vehicle location information, and generate navigation information to send to the delivery scheduling module (20); The delivery scheduling module (20) further comprises a plurality of mobile phone terminals (21), wherein the mobile phone terminals (21) are used to bind with the delivery vehicles to obtain vehicle location information in real time, and to receive and display navigation information required for executing the current delivery task.

2. According to claim 1, a cold chain logistics distribution system based on a refrigerator is characterized in that: The delivery scheduling module (20) specifically adopts the following steps:

1. Receive the delivery task, city map and vehicle location information sent by the data management library (10); Second, mark the city map based on the starting point location, target point location and vehicle location information of the delivery task; Third, through intelligent optimization algorithms, it finds the best delivery route for delivery vehicles and generates navigation information; Fourth, the navigation information is sent to the delivery dispatch module (20) and displayed in real time on the mobile phone (21), so that the delivery person can directly view it and start to execute the delivery task.

3. A cold chain logistics distribution system based on a refrigerator according to claim 2, characterized in that: The intelligent optimization algorithm divides the traversable path between the starting point and the target point of the delivery task into multiple sections from position i to position j, sets two indicators, namely, the traffic efficiency and the path priority, and calculates the section with the highest transfer probability from position i to position j. The optimal navigation information is formed by multiple sections with the highest transfer probability. The calculation formula of the transfer probability is as follows: ； in, : represents the transfer probability of the kth delivery vehicle from position i to position j at time t; :express The path priority from position i to position j at time; : represents the traffic efficiency from position i to position j; and : are two parameters, representing the relative importance of path priority and traffic efficiency respectively; : represents the set of paths that the k-th delivery vehicle can currently choose; : represents the path s that can be selected for all k, that is, , calculate the sum of the products of the path priority and the traffic efficiency from i to s; this sum is used as the denominator to normalize the probability to ensure that the sum of the probabilities of all selectable paths is 1.

4. A cold chain logistics distribution system based on refrigerators according to claim 3, characterized in that: The traffic efficiency includes giving priority to wide sections, giving priority to right turns, avoiding red lights, and avoiding congestion; Prioritizing wide roads means giving priority to roads accessible to delivery vehicles, and avoiding narrow roads and prohibited roads; Prioritize right turn: when generating navigation information, give priority to right turn in detour sections to avoid waiting for the left turn indicator light; Red light avoidance is specifically calculated by measuring the waiting time of the delivery vehicle at the intersection and detour time Numerical comparison of traffic efficiency; Congestion avoidance specifically reduces the proportion of congested sections in the generated navigation information.

5. The cold chain logistics distribution system based on refrigerators according to claim 3 is characterized in that: The path priority is used to update the path priority according to the path found by the kth delivery vehicle, and to strengthen the priority on the high-quality path, so as to guide the subsequent k+1th delivery vehicle to choose a better path. The specific formula is as follows: ； in, : The path priority from position i to position j at time; : represents the path priority decay coefficient, which controls the speed at which the path priority decays over time; : represents the path priority from position i to position j at time t; : The sum of the path priorities marked by all delivery vehicles k on the path ij; m: represents the total number of delivery vehicles; : represents the path priority marked by the kth delivery vehicle on the path ij; : Attenuation coefficient Controls the decay of path priority, Indicates the proportion of path priority reserved; : The remaining amount of the path priority after attenuation at time t.

6. A cold chain logistics distribution system based on refrigerators according to claim 5, characterized in that: The path priority is in a state of continuous superposition, so the attenuation coefficient must be set , to avoid the local path being marked as the optimal solution due to the accumulation of path priorities; The path priority is inversely proportional to the path length. The shorter the path, the more priority it will be marked with, so that the shortest path is more likely to be selected by the delivery vehicle.

7. The cold chain logistics distribution system based on refrigerators according to claim 3 is characterized in that: The two indicators of traffic efficiency and path priority are important factors in determining the transfer probability. Traffic efficiency and path priority are also complementary. When only one of the traffic efficiency and path priority exists, the navigation information finally generated will not be the optimal route. pass and The two coefficients assign weights to both traffic efficiency and path priority. Specifically, if When , the delivery vehicle is more inclined to follow the path of the historical delivery vehicle; when When the delivery vehicle is in a hurry, it is more inclined to choose the route with the shortest time.

8. The cold chain logistics distribution system based on refrigerators according to claim 1, characterized in that: The delivery task also includes the latest acceptance time, the type of goods to be delivered, and the volume of goods to be delivered; The latest acceptance time limits the time point at which the delivery vehicle arrives at the target location at the latest. Therefore, a time constraint is set to limit the generated navigation information. The specific formula is as follows: ； in, : represents the time constraint for the delivery vehicle to perform the delivery task; : Indicates the time when the delivery vehicle actually arrives at the starting point; : Indicates the latest acceptance time.

9. The cold chain logistics distribution system based on refrigerators according to claim 1, characterized in that: The city map updates the city traffic information of the delivery task execution in real time through the network, including the length of the road section, the location of the intersection, the time of the traffic light, the road condition information, the prohibition indication, and the congestion condition.

Images