Intelligent scheduling management system for new energy battery recovery

Abstract

The invention relates to the technical field of intelligent scheduling, in particular to an intelligent scheduling management system for new energy battery recovery, which comprises a real-time monitoring module, a dynamic relation graph construction module, a double-gradient decision module and a closed-loop updating module, and is characterized in that the real-time monitoring module continuously acquires the real-time position and moving path of each transport carrier and generates transport delay; capturing a multi-dimensional feature vector; the dynamic relation graph construction module models the multi-dimensional feature vectors into a relation graph comprising nodes and relation edges, wherein the relation edges among the nodes comprise weight parameters; the double-gradient decision-making module outputs a cross-regional scheduling path scheme set based on the weight parameters of the relation graph; the double-gradient decision module receives the transportation delay amount, and generates a weight adjustment amount according to a relation edge and a weight parameter of each node in the cross-regional scheduling path scheme set; and the closed-loop updating module feeds back the weight adjustment amount to the dynamic relation graph construction module to trigger dynamic updating of the weight parameter of the node relation edge.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent scheduling, and specifically to an intelligent scheduling management system for new energy battery recycling. Background Art

[0002] With the rapid development of the new energy industry, the demand for power battery recycling has surged, and the traditional scheduling management system has gradually shown significant deficiencies when dealing with complex transportation scenarios in multiple regions and multiple nodes. Traditional systems mostly rely on static path planning models and lack the ability to perceive and respond to dynamic environments (such as real-time traffic flow, sudden weather changes, and fluctuations in vehicle conditions), resulting in fixed transportation paths and low resource allocation efficiency. Especially in cross-regional scheduling, the dynamic coupling relationship between road network conditions, vehicle performance, and environmental factors has not been fully modeled, and transportation delay problems frequently occur, making it difficult to achieve global optimization.

[0003] In addition, the traditional system has limited ability to integrate multi-source heterogeneous data (such as vehicle positioning, road condition sensors, and meteorological information), and cannot construct a dynamic association map, making path decisions lack real-time and self-adaptability, further exacerbating the problems of rising transportation costs and lagging scheduling strategies. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent scheduling management system for new energy battery recycling to solve the problems raised in the above background art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: An intelligent scheduling management system for new energy battery recycling, including a real-time monitoring module, a dynamic relationship map construction module, a dual-gradient decision module, and a closed-loop update module, where:

[0006] The real-time monitoring module continuously collects the real-time positions and moving paths of each transport vehicle; generates a transportation delay amount by comparing the difference between the actual arrival time of the transport vehicle and the planned time node; and captures multi-dimensional feature vectors in real time through a multi-source heterogeneous sensor network;

[0007] The dynamic relationship map construction module models the multi-dimensional feature vectors into a relationship map including nodes and relationship edges, where the relationship edges between nodes include weight parameters;

[0008] The dual-gradient decision module includes a forward path generation unit and a reverse optimization unit, where:

[0009] The forward path generation unit calculates the transportation efficiency evaluation values of each candidate path through gradient direction analysis based on the weight parameters of the relationship map, and outputs a set of cross-regional scheduling path solutions;

[0010] The reverse optimization unit receives the transportation delay amount and generates a weight adjustment amount according to the relationship edges and weight parameters of each node in the cross-regional scheduling path scheme set.

[0011] The closed-loop update module feeds back the weight adjustment amount to the dynamic relationship graph construction module, triggering the dynamic update of the weight parameters of the node relationship edges, and is used to form the real-time adaptation ability of the scheduling strategy to the changes in the transportation environment.

[0012] As a further improvement of this technical solution, the multi-dimensional feature vector includes the working conditions of transportation vehicles, the road network passing capacity, and the environmental status.

[0013] As a further improvement of this technical solution, the nodes in the relationship graph include transportation vehicle nodes, road network nodes, and environmental nodes.

[0014] As a further improvement of this technical solution, the relationship edges in the relationship graph include the relationship between transportation vehicle nodes and road network nodes, the relationship between transportation vehicle nodes and environmental nodes, and the relationship between road network nodes and environmental nodes.

[0015] As a further improvement of this technical solution, the calculation formula of the weight parameter is as follows:

[0016] , where represents the weight parameter between node and node , and this weight parameter characterizes the dynamic association strength between node and node ; represents the physical distance between node and node ; represents the time required from node to node ; represents the traffic flow of node ; represents the weather condition of node ; represents the road condition of node ; represents a function for comprehensively calculating the weight parameter according to the above parameters , and the specific form of this function includes but is not limited to linear combination and weighted average mathematical models.

[0017] As a further improvement of this technical solution, the calculation process of the transportation efficiency evaluation value is as follows:

[0018] Use the graph search algorithm on the relationship graph to generate all candidate paths from the starting point to the ending point.

[0019] For each candidate path, calculate its transportation efficiency evaluation value, which is a comprehensive evaluation based on the relationship edge weight parameters between all nodes on the path . The calculation formula is as follows:

[0020] Let the candidate path be composed of a series of nodes , then , where is the weight parameter between node and node ; is the number of all nodes.

[0021] As a further improvement of this technical solution, the generation process of the cross-regional scheduling path set is as follows:

[0022] According to the transportation efficiency evaluation value and a preset transportation efficiency threshold, select the paths within the range of the transportation efficiency threshold as the cross-regional scheduling path set , where are the paths within the range of the transportation efficiency threshold, is the number of paths within the range of the transportation efficiency threshold.

[0023] As a further improvement of this technical solution, the reverse optimization unit is used to generate the optimal path, specifically including:

[0024] Receive the transportation delay amount data from the real-time monitoring module; and according to the cross-regional scheduling path set , select the path with the largest transportation delay amount in the paths as the optimal path .

[0025] As a further improvement of this technical solution, the generation process of the weight adjustment amount specifically includes:

[0026] For the optimal path , trace back the influence coefficient of the weight parameter corresponding to the node relationship edge in this path in reverse and generate the weight adjustment amount . The process is expressed as:

[0027] , where represents the influence coefficient of node and node in the optimal path . In addition, trace back each node on the reverse tracing path from ; represents the transportation delay amount of the optimal path ; represents the partial derivative symbol;

[0028] , where is an adjustment coefficient used to control the amplitude of the weight parameter adjustment.

[0029] As a further improvement of this technical solution, the formula for dynamically updating the weight parameter is as follows:

[0030] , where represents the weight parameter between node and node , represents the weight parameter between the updated node and node .

[0031] Compared with the prior art, the beneficial effects of the present invention are:

[0032] Through the real-time monitoring module and the dynamic relationship graph construction module, the intelligent scheduling management system for new energy battery recycling realizes the dynamic perception and structured modeling of multi-dimensional characteristics of the transportation environment, breaking through the limitation of data isolation in traditional systems; the dual-gradient decision-making module combines the forward path generation and reverse optimization mechanisms to synchronously promote path planning and weight parameter iteration in the dynamic relationship graph, significantly improving the real-time performance and global optimality of path selection; the closed-loop update module enables the system to have the dynamic adaptation ability to transportation delays and environmental changes through continuous feedback of the weight adjustment amount, effectively reducing transportation risks and resource waste; this technical framework not only solves the problem of path rigidity in traditional systems in dynamic scenarios, but also provides scalable intelligent decision-making support for the efficient scheduling of new energy battery recycling through a data-driven collaborative optimization mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is the overall module schematic diagram of the present invention;

[0034] Figure 2 is the unit schematic diagram of the dual-gradient decision-making module of the present invention.

[0035] In the figure: 100, real-time monitoring module; 200, dynamic relationship graph construction module; 300, dual-gradient decision-making module; 301, forward path generation unit; 302, reverse optimization unit; 400, closed-loop update module. DETAILED DESCRIPTION OF THE INVENTION

[0036] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0037] Next, please refer to Figure 1 , the present invention provides a technical solution: an intelligent scheduling management system for new energy battery recycling, including a real-time monitoring module 100, a dynamic relationship graph construction module 200, a dual-gradient decision module 300, and a closed-loop update module 400.

[0038] The real-time monitoring module 100 continuously collects the real-time positions and moving paths of each transport vehicle, generates a transport delay amount by comparing the difference between the actual arrival time of the transport vehicle and the planned time node, and simultaneously captures multi-dimensional feature vectors including the operating conditions of the transport vehicle, the road network passing capacity, and the environmental state through a multi-source heterogeneous sensor network. Specifically, it includes:

[0039] Each transport vehicle is equipped with a GPS positioning device, which sends the current position coordinates once per minute (or more frequently); according to the received coordinate information, continuously records and plots the complete moving path of the transport vehicle from the starting point to the end point;

[0040] Before each transport task starts, a schedule is generated for each transport vehicle based on factors such as the starting point, the end point, and the expected driving speed, specifying the expected arrival time at each key node (such as intermediate checkpoints, final destinations); when the transport vehicle arrives at each key node, record its actual arrival time; compare the actual arrival time with the expected arrival time (planned time node) in the schedule, calculate the difference between the two. If the actual arrival time is later than the expected time, the difference is positive, indicating a delay; if it is earlier than the expected time, the difference is negative, indicating an early arrival. This difference is used as the transport delay amount;

[0041] Capture multi-dimensional feature vectors including the operating conditions of the transport vehicle, the road network passing capacity, and the environmental state through a multi-source heterogeneous sensor network. The configuration of the multi-source heterogeneous sensor network specifically includes:

[0042] Operating conditions of the transport vehicle:

[0043] Temperature sensor: Measures the temperature inside and outside the transport vehicle;

[0044] Humidity sensor: Measures the humidity inside and outside the transport vehicle;

[0045] Vibration sensor: Monitors the vibration situation during transportation;

[0046] Battery status sensor: monitors the status of new energy batteries, including remaining power, charging status, etc.;

[0047] Road network traffic capacity:

[0048] Traffic flow sensor: installed at key sections to monitor the traffic flow of the current road;

[0049] Vehicle speed sensor: monitors the driving speed of transport vehicles on different sections;

[0050] Road condition camera: provides real-time road video footage for detecting congestion, accidents, etc.;

[0051] Environmental status:

[0052] Weather sensor: measures the current weather conditions, including temperature, humidity, wind speed, rainfall, etc.;

[0053] Light sensor: measures the current light intensity;

[0054] Air quality sensor: monitors the concentration of pollutants in the air.

[0055] The multi-source heterogeneous sensor network collects data at a fixed frequency (e.g., once per minute) and transmits the data to the system center in real time; integrates data from different sensors into a multi-dimensional feature vector containing the operating conditions of transport vehicles, road network traffic capacity, and environmental status.

[0056] The dynamic relationship graph construction module 200 models the multi-dimensional feature vector as a relationship graph containing nodes and relationship edges, where the nodes include transport vehicle nodes, road network nodes, and environmental nodes, and the relationship edges between nodes characterize the dynamic association strength through weight parameters, specifically including:

[0057] Each transport vehicle is represented as a node in the relationship graph, and the transport vehicle node includes the unique identifier of the transport vehicle (such as license plate number), current location, movement path, and operating condition information (such as internal temperature, humidity, vibration intensity, battery status, etc.);

[0058] Key points in the road network (such as intersections, checkpoints, destinations, etc.) are represented as road network nodes in the graph, and the road network nodes include the geographical location of the nodes, traffic flow, current vehicle speed, and road condition information (such as congestion situation, accident report, etc.);

[0059] Environmental factors (such as weather, light, air quality, etc.) are represented as environmental nodes in the graph, and the environmental nodes include: the current weather conditions (such as temperature, humidity, wind speed, rainfall, etc.), light intensity, air quality index;

[0060] The relationship edges between nodes are represented by relationship edges, which reflect the interactions and influences between different nodes; specifically including:

[0061] The relationship between the transportation vehicle node and the road network node. For example, transportation vehicle A is currently located at road network node B, and this relationship is represented by an edge.

[0062] The relationship between the transportation vehicle node and the environment node. For example, transportation vehicle A is affected by environment node C (such as the current weather conditions), and this relationship is represented by an edge.

[0063] The relationship between the road network node and the environment node. For example, the traffic capacity of road network node D is affected by environment node E (such as traffic congestion caused by bad weather), and this relationship is also represented by an edge;

[0064] Definition of weight parameter: The weight parameter on each relationship edge is used to characterize the dynamic association strength of the relationship between nodes. The weight parameter is calculated based on multiple factors, including but not limited to:

[0065] Distance: The physical distance between two nodes;

[0066] Time: The time required for a transportation vehicle to move from one node to another;

[0067] Traffic flow: The impact of the traffic flow of the road network node on the transportation vehicle;

[0068] Weather conditions: The impact of the weather conditions of the environment node on the road network node and the transportation vehicle;

[0069] Road conditions: The impact of the road conditions (such as congestion level, accident situation) of the road network node on the transportation vehicle.

[0070] The following formula is used to calculate the weight parameter:

[0071] , where represents the weight parameter between node and node , and this weight parameter characterizes the dynamic association strength between node and node ; represents the physical distance between node and node , and this distance can be the actual geographical distance or other forms of distance (such as time distance); represents the time required from node to node , and this time can be the estimated driving time or the actual driving time; Represents the traffic flow of a node , which reflects the number of vehicles passing through the node or the traffic density; Represents the weather conditions of the node , including meteorological data such as temperature, humidity, wind speed, rainfall, etc.; Represents the road conditions of the node , and this parameter reflects the road conditions of the node , such as whether it is congested, whether there is an accident, etc.; Represents a function for comprehensively calculating the weight parameter according to the above parameters , and the specific form of this function is designed according to actual needs. For example, it can be a linear combination, weighted average or other complex mathematical models.

[0072] Each node, relationship edge and weight parameter in the dynamic relationship graph construction module 200 together constitute a relationship graph.

[0073] Please refer to Figure 2 , the forward path generation unit 301 in the double gradient decision module 300 calculates the transportation efficiency evaluation value of each candidate path based on the weight parameters of the relationship graph through gradient direction analysis, and outputs a cross-regional scheduling path scheme set, specifically including:

[0074] Use a graph search algorithm (such as Dijkstra algorithm or A* algorithm) on the relationship graph to generate all candidate paths from the starting point to the ending point;

[0075] For each candidate path, calculate its transportation efficiency evaluation value, and the transportation efficiency evaluation value is based on the relationship edge weight parameters between all nodes on the path Comprehensive evaluation, The calculation formula is as follows:

[0076] Suppose the candidate path Consists of a series of nodes , then , where Is the weight parameter between node And node ; Is the number of all nodes;

[0077] According to the transportation efficiency evaluation value and preset the transportation efficiency threshold, select the paths within the range of the transportation efficiency threshold as the cross-regional scheduling path scheme set , where Is the path within the range of the transportation efficiency threshold, Is the number of paths within the range of the transportation efficiency threshold.

[0078] The reverse optimization unit 302 in the dual-gradient decision module 300 receives the transportation delay amount, and based on the relationship edges and weight parameters of each node in the cross-regional scheduling path scheme set, traces back the influence coefficient of the weight parameter in reverse order along the path execution sequence to generate a weight adjustment amount, specifically including:

[0079] Receiving the transportation delay amount data from the real-time monitoring module 100; and according to the cross-regional scheduling path scheme set , selecting the path with the largest transportation delay amount in the path as the optimal path ;

[0080] For the optimal path , trace back the influence coefficient of the weight parameter corresponding to the node relationship edge in this path in reverse order, and generate a weight adjustment amount The process is expressed as:

[0081] , where represents the influence coefficient of nodes in the optimal path and node . Additionally, Trace back each node on the path from to 0 in reverse order; represents the transportation delay amount of the optimal path ; represents the partial derivative symbol;

[0082] , where is an adjustment coefficient (also known as the learning rate or step size), used to control the adjustment amplitude of the weight parameter.

[0083] The closed-loop update module 400 feeds back the weight adjustment amount to the dynamic relationship graph construction module 200, triggering the dynamic update of the weight parameter of the node relationship edge. The specific formula is as follows:

[0084] , where represents the weight parameter between nodes and node , represents the updated weight parameter between nodes and node .

[0085] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and the descriptions in the specification are only preferred examples of the present invention, which are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. New energy battery recycling intelligent scheduling and management system, characterized by: The system comprises a real-time monitoring module (100), a dynamic relationship graph construction module (200), a dual gradient decision module (300) and a closed-loop update module (400), wherein: The real-time monitoring module (100) continuously collects the real-time position and movement path of each transport vehicle; generates the transport delay by comparing the difference between the actual arrival time of the transport vehicle and the planned time node; and captures the multi-dimensional feature vector in real time through a multi-source heterogeneous sensor network; The dynamic relationship graph construction module (200) models the multi-dimensional feature vector as a relationship graph including nodes and relationship edges, wherein the relationship edges between nodes include weight parameters; The dual gradient decision module (300) comprises a forward path generation unit (301) and a reverse optimization unit (302), wherein: The forward path generation unit (301) calculates the transport efficiency evaluation value of each candidate path through gradient direction analysis based on the weight parameters of the relationship graph, and outputs a set of cross-regional scheduling path solutions; The reverse optimization unit (302) receives the transport delay amount, and generates a weight adjustment amount according to the relationship edges and weight parameters of each node in the cross-regional scheduling path solution set; The closed-loop update module (400) feeds back the weight adjustment amount to the dynamic relationship graph construction module (200), triggering a dynamic update of the weight parameters of the node relationship edges, so as to form a real-time adaptation capability for the scheduling strategy and the transportation environment changes.

2. The intelligent dispatching and management system for recycling new energy batteries according to claim 1 is characterized in that: The multi-dimensional feature vector includes the operating conditions of the transport vehicle, the road network capacity and the environmental status.

3. The intelligent dispatching and management system for recycling new energy batteries according to claim 1 is characterized in that: The nodes in the relationship graph include transport vehicle nodes, road network nodes and environment nodes.

4. The intelligent dispatching and management system for recycling new energy batteries according to claim 1 is characterized in that: The relationship edges in the relationship graph include the relationship between the transportation vehicle node and the road network node, the relationship between the transportation vehicle node and the environment node, and the relationship between the road network node and the environment node.

5. The intelligent dispatching and management system for recycling new energy batteries according to claim 1 is characterized in that: The calculation formula of the weight parameter is as follows: ,in Representation Node and nodes The weight parameter between nodes and nodes The strength of the dynamic association between Representation Node and nodes The physical distance between Represents a slave node and nodes The time required between Representation Node Traffic flow; Representation Node weather conditions; Representation Node road conditions; Represents a function used to comprehensively calculate the weight parameters based on the above parameters , the specific form of the function includes but is not limited to linear combination and weighted average mathematical model.

6. The intelligent dispatching and management system for recycling new energy batteries according to claim 1 is characterized in that: The calculation process of the transport efficiency evaluation value is as follows: Use graph search algorithm to generate all candidate paths from the starting point to the end point on the relationship graph; For each candidate path, calculate its transport efficiency evaluation value, which is based on the relationship edge weight parameters between all nodes on the path. A comprehensive assessment of The calculation formula is as follows: Set candidate path A series of nodes Composition, then ,in Is a node and nodes The weight parameter between ; is the number of all nodes.

7. The intelligent dispatching and management system for recycling new energy batteries according to claim 1 is characterized in that: The generation process of the cross-regional scheduling path solution set is as follows: According to the transport efficiency evaluation value and the preset transport efficiency threshold, the paths that meet the transport efficiency threshold range are selected as the cross-regional scheduling path solution set. ,in is a path that meets the transport efficiency threshold. is the number of routes that meet the transportation efficiency threshold.

8. The intelligent dispatching and management system for recycling new energy batteries according to claim 1 is characterized in that: The reverse optimization unit (302) is used to generate an optimal path, specifically including: Receive the transport delay data from the real-time monitoring module (100); and , select the path with the largest transportation delay as the optimal path .

9. The intelligent dispatching and management system for recycling new energy batteries according to claim 8 is characterized in that: The generation process of the weight adjustment amount specifically includes: For the optimal path , trace back the influence coefficient of the weight parameter corresponding to the node relationship edge in the path, and generate the weight adjustment The process is expressed as: ,in Represents the optimal path Midpoint and nodes In addition, from Go back to each node on the path from 0; Represents the optimal path The amount of shipping delays; represents the sign of partial derivative; ,in It is an adjustment coefficient used to control the magnitude of the weight parameter adjustment.

10. The intelligent dispatching and management system for recycling new energy batteries according to claim 9 is characterized in that: The formula for dynamically updating the weight parameters is as follows: ,in Representation Node and nodes The weight parameter between Represents the updated node and nodes The weight parameter between .

Images