CN-122022653-A - Intelligent management system and method for new energy logistics network based on digital twin

CN122022653ACN 122022653 ACN122022653 ACN 122022653ACN-122022653-A

Abstract

The invention discloses a new energy logistics network intelligent management system and method based on digital twin, which relate to the technical field of logistics network management, and the system and method provided by the invention are used for sensing abnormal events in a logistics network in real time, mapping the events to a digital twin through semantic analysis, and adjusting the topological structure and operation parameters of the digital twin network; the method comprises the steps of establishing a seepage model for the influence propagation of abnormal events based on a digital twin network, analyzing the infection probability and seepage centrality of network nodes to generate an abnormal influence thermodynamic diagram, establishing a scene tree aiming at the abnormal influence thermodynamic diagram and a current scheduling plan, carrying out parallel simulation on each branch of the scene tree in the digital twin environment, selecting an optimal scheduling strategy through target evaluation, tracking an execution effect in real time, establishing an experience case library, and matching similar historical scenes through analog reasoning to realize the self-adaptive optimization of the abnormal coping strategy. Improving response speed and providing robustness of the system.

Inventors

DU SONGLIN
ZHANG LIYANG
LEI TAO
Sadamu Shadik
WANG BINGQUAN
WANG XIAOFENG
YU JIONG
DU XUSHENG

Assignees

杭州骋风而来数字科技有限公司
新疆丝路融创网络科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260414

Claims (10)

1. The intelligent management method of the new energy logistics network based on digital twin is characterized by comprising the following steps of: S1, sensing an abnormal event in a logistics network in real time, mapping the abnormal event to a digital twin body through semantic analysis, and adjusting the topological structure and the operation parameters of the digital twin network; S2, constructing a seepage model for the influence propagation of an abnormal event based on a digital twin network, analyzing the infection probability and seepage centrality of network nodes, and generating an abnormal influence thermodynamic diagram; S3, constructing a scene tree aiming at an abnormal influence thermodynamic diagram and a current scheduling plan, carrying out parallel simulation on each branch of the scene tree in a digital twin environment, and selecting an optimal scheduling strategy through target evaluation; S4, tracking the execution effect in real time, constructing an experience case library, and matching similar historical scenes through analog reasoning to realize the self-adaptive optimization of the abnormality coping strategy.
2. The intelligent management method of the new energy logistics network based on digital twinning according to claim 1, wherein in S1, the method comprises the following steps: S101, acquiring multi-source data in a logistics network in real time, wherein the multi-source data comprises abnormal event data and conventional operation data, the abnormal event data at least comprises new energy vehicle fault data, a charging pile disconnection warning flag bit, an order cancel mark and traffic accident data, and the conventional operation data at least comprises new energy vehicle position, residual electric quantity, charging pile power and order starting and ending point coordinates; mapping the abnormal event data to the attribute of the corresponding entity in the digital twin body through a predefined event semantic analysis template; And S102, adjusting the topological structure and the operation parameters of the digital twin network according to the extracted abnormal event type, position and influence parameters, wherein the reconstructed digital twin network is expressed as a graph structure and comprises a solid node set, a time-varying edge set and a node attribute matrix, the node attribute matrix comprises conventional operation data and abnormal event feature vectors, and the abnormal event feature vectors at least comprise event type codes, intensity levels, abnormal occurrence positions, influence radiuses and starting time fields.
3. The intelligent management method of the new energy logistics network based on the digital twin is characterized in that mapping the abnormal event to the attribute of the corresponding entity in the digital twin comprises the steps of endowing the new energy vehicle fault data to the fault type attribute of the virtual new energy vehicle model and recording the abnormal starting time, mapping the traffic accident data to the corresponding road section in the road network model to generate the temporary obstacle attribute, setting the state field of the charging pile with the charging pile disconnection warning flag bit of 1 to be unavailable, and setting the state field of the order with the order cancellation mark of 1 to be cancelled.
4. The intelligent management method for the new energy logistics network based on the digital twin system according to claim 3, wherein the step of adjusting the topological structure and the operation parameters of the digital twin system comprises the steps of temporarily removing corresponding charging pile nodes from a charging service network for charging pile disconnection warning, updating the adjacent relation of the charging network, locating affected road sections according to accident coordinates for traffic accident data, degrading the traffic capacity of the corresponding road sections, updating road network edge weights, marking fault states on a new energy vehicle model according to fault types and influence radiuses for new energy vehicle fault data, adding temporary nodes to represent fault occupation, updating order pool states for order cancellation, and removing transportation requirements of corresponding orders from a dispatching task list.
5. The intelligent management method of the new energy logistics network based on digital twinning according to claim 2, wherein in S2, the method comprises the following steps: S201, defining initial infection probability of each node on a digital twin network, defining probability of an infected node transmitting abnormality to adjacent susceptible nodes, defining recovery rate of the adjacent susceptible nodes as reciprocal of time required for abnormality recovery, wherein evolution of the infection probability and recovery probability of the adjacent susceptible nodes is described by a seepage model; s202, solving the seepage model to obtain the infection probability of each adjacent susceptible node at the moment, extracting the nodes with the infection probability exceeding a preset infection probability threshold value, and defining an abnormal influence area; Marking the nodes with the seepage center degree exceeding the seepage center degree threshold as key nodes, and outputting a key node list and influence region boundary information; and S203, dividing the monitoring area into uniform geographical grids, wherein the thermodynamic value of each grid is equal to the average infection probability of all nodes in the grid, and generating an abnormal influence thermodynamic diagram.
6. The intelligent management method of the new energy logistics network based on digital twinning according to claim 5, wherein in S3, the method comprises the following steps: S301, constructing a scene tree by taking an abnormal influence thermodynamic diagram and a current scheduling plan as inputs, wherein the current scheduling plan comprises a new energy vehicle path, charging task allocation and an order new energy vehicle matching relation, a root node of the scene tree represents a network state at the current moment, a plurality of branches are generated from the root node based on a predefined intervention rule base, each branch represents a scheduling intervention measure, and each branch corresponds to a group of intervention operation sets; s302, in a digital twin environment, carrying out parallel simulation on each branch of a scene tree, and deducing evolution of network states in a future time window from the current moment; after deduction is finished, recording a target evaluation index vector of each branch, wherein the target evaluation index vector comprises total delay time, total energy consumption and abnormal recovery time, the total delay time is the sum of the differences between the actual completion time and the planned completion time of all orders, the total energy consumption is the sum of electric quantity consumed by all new energy vehicles during deduction, and the abnormal recovery time is the time from the current moment until the thermal values of all grids are lower than a preset recovery threshold; S303, constructing a reference point set, wherein the reference points consist of historical optimal target values, business expected targets or ideal points, and each reference point comprises a corresponding optimal evaluation index vector; The branch with the minimum adaptability is selected as an optimal branch, the corresponding intervention operation set is an optimal scheduling strategy, and the optimal strategy is converted into a specific scheduling instruction set and is issued to an actual logistics execution system.
7. The intelligent management method of the new energy logistics network based on digital twinning of claim 6 is characterized in that the intervention rule base at least comprises marking a corresponding geographic grid as a high-impact area when the grid thermal value is larger than the grid thermal value threshold, generating a bypass branch, the intervention operation set comprises rescheduling a path for an affected new energy vehicle, generating a charge load transfer branch when a charge pile is abnormally unavailable or is in the high-impact area, acquiring a list of the affected new energy vehicle, solving an adjustment scheme of a charge distribution matrix by adopting bipartite graph matching or linear programming based on the charge pile state and the residual electricity quantity of the new energy vehicle, updating a charge pile distribution target of each new energy vehicle, generating an order redistribution branch when an order starting end point is in the high-impact area or cannot be achieved due to a new energy vehicle fault, releasing an incomplete order from an original matching relation, updating the order pool, calling a new energy vehicle order matching algorithm, and generating a new order new energy vehicle matching relation by using an assignment model considering a time window but not limited to the time window.
8. The intelligent management method of the new energy logistics network based on the digital twin is characterized in that a simulation mechanism is as follows, in each simulation step length, the topology and the node attribute of the digital twin network are updated according to a branched intervention operation set, the node infection probability is updated based on an updated network state calling seepage model, meanwhile, logistics operation indexes are calculated according to new energy vehicle paths, charging plans and order matching, the logistics operation indexes at least comprise new energy vehicle positions, residual electric quantity and order completion states, and the process is repeated until deduction finishing time.
9. The intelligent management method of the new energy logistics network based on digital twinning according to claim 7, wherein in S4, the method comprises the following steps: S401, continuously collecting real-time operation data of a logistics network in an execution process, wherein the real-time operation data comprise, but are not limited to, real-time position, speed, residual electric quantity and fault state of a new energy vehicle, working state and output power of a charging pile, actual completion condition of an order, and newly-generated abnormal events, wherein the newly-generated abnormal events comprise, but are not limited to, secondary faults, traffic accidents and road congestion; S402, combining the characteristic vector of the current abnormal event, the selected intervention strategy, the execution deviation vector and the final recovery mark into a case, and storing the case into an experience case library; the final recovery mark is a binary variable, and is 1 if the abnormality is relieved within a preset time window and a new linkage abnormality is not caused, or is 0; S403, when a new abnormal event occurs, extracting a corresponding abnormal event feature vector, calculating Euclidean distance between the new abnormal event feature vector and the abnormal event feature vector of each historical case in the knowledge base, if the minimum distance is smaller than a preset threshold value and the final recovery mark of the corresponding case is 1, directly calling an intervention strategy of the corresponding case as a candidate scheme, otherwise, constructing a scene tree to generate a new strategy.
10. The intelligent management system for the new energy logistics network based on the digital twin is applied to the intelligent management method for the new energy logistics network based on the digital twin, which is realized by any one of claims 1-9, and is characterized by comprising an anomaly adjustment module, an anomaly influence propagation analysis module, a scheduling strategy analysis module and a strategy optimizing module; the anomaly adjustment module senses an anomaly event in the logistics network in real time, maps the event to a digital twin body through semantic analysis, and adjusts the topological structure and the operation parameters of the digital twin network; The abnormal influence propagation analysis module is used for constructing a seepage model of abnormal event influence propagation based on a digital twin network, analyzing the infection probability and seepage centrality of network nodes and generating an abnormal influence thermodynamic diagram; The scheduling strategy analysis module builds a scene tree aiming at an abnormal influence thermodynamic diagram and a current scheduling plan, carries out parallel simulation on each branch of the scene tree in a digital twin environment, and selects an optimal scheduling strategy through target evaluation; The strategy optimizing module tracks the execution effect in real time, builds an experience case base, and matches similar historical scenes through analog reasoning so as to realize the self-adaptive optimization of the abnormality coping strategy.

Description

Intelligent management system and method for new energy logistics network based on digital twin Technical Field The invention relates to the technical field of logistics network management, in particular to a new energy logistics network intelligent management system and method based on digital twinning. Background In the current new energy logistics network management system, the application of the digital twin technology is mainly focused on the monitoring and simulation of the conventional running state. In the prior art, a static or quasi-static twin model is built for path planning and scheduling optimization generally depending on vehicle tracks, charging pile states and order data acquired in a fixed period. However, this type of approach has significant drawbacks in the face of emergency events. Firstly, the response of the existing system to the abnormal event has serious lag, and the conditions such as faults, accidents or order cancellation can be found by manually confirming or waiting for the next data acquisition period, so that the twin model is disjointed with the real-time state of the physical world. When a vehicle fault or a broken connection of a charging pile occurs, the system still makes a scheduling decision based on an outdated network topology, and other vehicles can possibly be introduced into a charging node or a congestion road section which is already in failure, so that the problem cannot be relieved, and the diffusion of abnormal influence is aggravated. Secondly, the traditional method lacks the capability of quantitative analysis of the abnormal event propagation mechanism, and often treats the abnormality as an isolated local event, and ignores the cascading effect existing among traffic jam, energy shortage and order delay. A failure of one node may propagate rapidly through a network connection, resulting in a larger range of network performance degradation, where it is difficult in the prior art to identify such propagation paths and impact ranges in advance. Therefore, the invention discloses a new energy logistics network intelligent management system and method based on digital twinning to solve the problems. Disclosure of Invention The invention aims to provide a new energy logistics network intelligent management system and method based on digital twinning, so as to solve the problems in the prior art. In order to achieve the purpose, the invention provides the following technical scheme that the intelligent management method of the new energy logistics network based on digital twinning comprises the following steps: S1, sensing an abnormal event in a logistics network in real time, mapping the abnormal event to a digital twin body through semantic analysis, and adjusting the topological structure and the operation parameters of the digital twin network; S2, constructing a seepage model for the influence propagation of an abnormal event based on a digital twin network, analyzing the infection probability and seepage centrality of network nodes, and generating an abnormal influence thermodynamic diagram; S3, constructing a scene tree aiming at an abnormal influence thermodynamic diagram and a current scheduling plan, carrying out parallel simulation on each branch of the scene tree in a digital twin environment, and selecting an optimal scheduling strategy through target evaluation; S4, tracking the execution effect in real time, constructing an experience case library, and matching similar historical scenes through analog reasoning to realize the self-adaptive optimization of the abnormality coping strategy. In S1, the following are included: S101, acquiring multi-source data in a logistics network in real time, wherein the multi-source data comprises abnormal event data and conventional operation data, the abnormal event data at least comprises new energy vehicle fault data, a charging pile disconnection warning flag bit, an order cancel mark and traffic accident data, and the conventional operation data at least comprises new energy vehicle position, residual electric quantity, charging pile power and order starting and ending point coordinates; Mapping the abnormal event data to attributes of corresponding entities in the digital twin through a predefined event semantic parsing template, The method comprises the steps of endowing new energy vehicle fault data with fault type attributes of a virtual new energy vehicle model, recording abnormal starting time, mapping traffic accident data to corresponding road sections in a road network model, generating temporary barrier attributes, setting a state field of a charging pile with a charging pile disconnection warning zone bit of 1 to be unavailable, setting a state field of an order with an order cancellation mark of 1 to be cancelled, and completing structural characterization of an abnormal event to be associated with a twin entity; S102, adjusting the topological structure and the operation parameters of the dig