Urban rail transit model construction and scheduling application platform system

Abstract

The invention provides a construction and scheduling application platform system of an urban rail transit model. Comprising a storage management server, a model construction server and a model scheduling application server, the storage management server is used for storing and managing basic data, an AI algorithm and a model, transmitting the data, the algorithm and the model to the model construction server, receiving a system algorithm transmitted by the model construction server, and updating, managing and storing the system algorithm; the model construction server performs model construction and solution according to the instruction of the model scheduling application server, supports the demand of the model scheduling application server, receives the feedback of the model scheduling application server and performs model online updating; and the model scheduling application server is used for realizing calling and evaluation of each unit domain model. The system integrates a multi-field service application function, intelligent generation and dynamic adjustment of a solution scheme are realized, and the automation and intelligence level of service processing is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban rail transit management, and in particular to a construction and dispatching application platform system for an urban rail transit model. Background Art

[0002] The business applications in the urban rail transit industry are characterized by complexity, diversity, refinement, and integration. Traditional business application systems have low automation and intelligence in terms of identifying business areas and intelligent construction and solving problems. Therefore, there is an urgent need to build an intelligent construction and application system for the urban rail transit industry that integrates multi-field business application functions, realizes intelligent generation and dynamic adjustment of solutions, and improves the level of automation and intelligence in business processing. Summary of the invention

[0003] The embodiments of the present invention provide a system for constructing an urban rail transit model and a scheduling application platform to realize automatic and intelligent solutions to transportation organization, passenger service, asset maintenance and cross-domain business needs.

[0004] In order to achieve the above object, the present invention adopts the following technical scheme.

[0005] A construction and dispatching application platform system for an urban rail transit model, comprising: a storage management server, a model construction server and a model dispatching application server;

[0006] The storage management server is used to store and manage basic data, AI algorithms and models, transmit data, algorithms and models to the model construction server, receive the system algorithm transmitted by the model construction server, and update, manage and store the system algorithm;

[0007] The model construction server is used to construct and solve the model according to the instructions of the model scheduling application server, support the needs of the model scheduling application server, receive feedback from the model scheduling application server and perform online model updates;

[0008] The model scheduling application server is used to realize the calling and evaluation of each unit domain model, including a scene analysis unit, a transportation organization application unit, a passenger service application unit, an asset maintenance application unit, a cross-domain collaborative application unit, and a model evaluation unit.

[0009] Preferably, the storage management server includes: a data storage management unit, an AI algorithm storage management unit and a model storage management unit;

[0010] The data storage management unit is used to store and manage multi-standard, multi-source and heterogeneous data of urban rail transit, including passenger flow data units, driving data units, service data units, equipment data units and asset data units;

[0011] The AI ​​algorithm storage management unit is used to receive the system algorithm transmitted by the model construction server, and the system algorithm includes: prediction and detection algorithm, optimization model and solution algorithm, and performs system algorithm storage and update, including data statistics algorithm unit, decision support algorithm unit, prediction and detection algorithm unit, optimization solution algorithm unit and exception handling algorithm unit;

[0012] The model storage management unit is used to store, manage, optimize and update various models, including service units, organization units and asset units; the service management unit includes a travel service model and a personalized recommendation model, the organization unit includes a new line opening passenger flow prediction model and a transfer passenger flow prediction model, and the asset unit includes a maintenance status perception model and an autonomous diagnosis model.

[0013] Preferably, the model construction server comprises: a prediction detection model construction unit, an optimization model construction and solution unit and a model online update unit;

[0014] The prediction detection model construction unit is used for applications related to prediction detection in the fields of transportation organization, passenger service and asset maintenance, receiving demand targets and evaluation indicators generated by the model scheduling application server according to actual scenario requirements, calling the data storage management unit and the AI ​​algorithm storage management unit to generate training samples and a combination algorithm set, training the combination algorithm, generating the optimal result through comparison, and then constructing a prediction detection algorithm, and transmitting the prediction detection algorithm to the storage management server;

[0015] The optimization model construction and solution unit is used for optimization and solution-related applications in the fields of transportation organization, passenger service and asset maintenance, receiving demand targets, constraints and decision variables generated by the model scheduling application server through scenario analysis, calling the data storage management unit and the AI ​​algorithm storage management unit to generate test data and a combination algorithm set, testing the combination algorithm, generating the optimal result through comparison, and then constructing an optimization model and a solution algorithm, and transmitting the optimization model and the solution algorithm to the storage management server;

[0016] The model online update unit is used to receive feedback from the optimization model construction and solution unit through data and scenario updates, continuously calibrate the model training results, perform algorithm training and parameter optimization according to new data, adjust model objectives and constraints, and dynamically select and apply the optimal model, algorithm and parameters, including: a prediction and detection model update unit, an optimization model update unit and a parameter update unit.

[0017] Preferably, the model scheduling application server model includes a scenario analysis unit, a transport organization application unit, a passenger service application unit, an asset maintenance application unit, a cross-domain collaboration application unit and a model evaluation unit;

[0018] The scenario analysis unit is used to analyze the business domain, state and element-related data of the application according to the demand scenario, construct the demand target, constraint conditions and decision variables, and transmit the demand target, constraint conditions and decision variables to the model construction server;

[0019] The transport organization application unit is used to call the model generation server to generate models for transport organization applications according to application scenarios, and output the models to external practical applications. The supported application scenarios include traffic demand forecasting, capacity configuration and evaluation optimization, passenger organization process simulation, and digital plans and emergency resource scheduling;

[0020] The passenger service application unit is used to call the model generation server to generate a model for passenger service applications according to application scenarios, and output the model to external actual applications. The supported application scenarios include passenger profiling, demand management, recommendation guidance, itinerary prediction and travel services;

[0021] The asset maintenance application unit is used to call the model generation server to generate models for asset maintenance applications according to application scenarios, and output the models to external practical applications. The supported application scenarios include intelligent analysis and early warning, intelligent maintenance operations, and intelligent asset linkage;

[0022] The cross-domain collaborative application unit is used to collaboratively call the transport organization application unit, the passenger service application unit, and the asset maintenance application unit according to the application scenarios that require collaboration in various fields during the operation of urban rail transit, and output the model to external practical applications. The supported application scenarios include: collaborative emergency response and collaborative passenger flow management;

[0023] The model evaluation unit is used to conduct a comprehensive evaluation of the model application effect and performance, and to evaluate, adjust and update the system application through evaluation indicators and evaluation models.

[0024] Preferably, the processing process of the system for implementing passenger services includes the following processing steps:

[0025] Step S1: The model scheduling application server analyzes the passenger service scenario to obtain the business application field, service time and space, and passenger service requirements of the passenger service;

[0026] Step S101: Generate demand targets and evaluation indicators for passenger services through passenger service scenario analysis. The passenger service demand targets include travel planning services, personalized recommendation services, and passenger portrait services. Travel planning services include travel origin-destination OD selection, travel obstacle avoidance, travel mode selection, and convenient services. Personalized recommendation services include travel time recommendations and travel route recommendations. Passenger portrait services include travel portraits and ticket analysis. Evaluation indicators include passenger satisfaction, prediction accuracy, section congestion, train full load rate, departure interval, and route congestion indicators.

[0027] Step S102: calling a specific passenger application unit in the model scheduling application server according to the demand target of the passenger service scenario analysis, calling a specific evaluation unit in the model scheduling application server according to the evaluation index of the passenger service scenario analysis, and evaluating the passenger application through the evaluation unit;

[0028] Step S103: the model scheduling application server transmits the parsing instruction to the model construction server for model construction, and receives feedback from the model construction server for server update optimization;

[0029] Step S2: The model construction server receives the application unit and model instruction transmitted by the model scheduling application server, calls the model construction server through the received specific passenger application unit, trains / solves the training samples / test data through the combination algorithm, generates results, and constructs the optimal model through comparison;

[0030] Step S201: the model construction server calls data from the storage management server and generates training data / test samples through a sampling algorithm;

[0031] Step S202: the storage management server stores system algorithms, which include recommendation algorithms, prediction algorithms, portrait generation algorithms, feature extraction algorithms, and path utility calculation algorithms;

[0032] Step S203: Generate a passenger service model by combination, the passenger service model including a travel demand prediction model, a travel time prediction model, a personalized recommendation model and a portrait label generation model;

[0033] Step S3: The storage management server receives the instruction transmission data, algorithm and model transmitted by the model construction server, and updates, optimizes, stores and manages them according to the feedback from the model construction server.

[0034] It can be seen from the technical solutions provided by the above-mentioned embodiments of the present invention that the present invention solves the current situation that traditional application systems have low automation and intelligence in identifying business fields and intelligent construction solutions, constructs an intelligent construction and application system for the urban rail transit industry, integrates multi-field business application functions, realizes intelligent generation and dynamic adjustment of solutions, and improves the automation and intelligence level of business processing.

[0035] Additional aspects and advantages of the present invention will be given in part in the following description, which will become obvious from the following description, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0037] Figure 1 The overall logical architecture diagram of a system for constructing an urban rail transit model and a scheduling application platform provided by an embodiment of the present invention;

[0038] Figure 2 An architectural diagram of a storage management server provided in an embodiment of the present invention;

[0039] Figure 3 An architecture diagram of a model construction server provided in an embodiment of the present invention;

[0040] Figure 4 A processing flow chart of a prediction detection model construction unit in a model construction server provided in an embodiment of the present invention;

[0041] Figure 5 A processing flow chart of an optimization model construction and solution unit in a model construction server provided in an embodiment of the present invention;

[0042] Figure 6 An architecture diagram of a model scheduling application server provided in an embodiment of the present invention;

[0043] Figure 7 The present invention provides an implementation case of passenger service of an urban rail transit model intelligent construction and dispatching application platform system. DETAILED DESCRIPTION

[0044] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0045] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or coupling. The term "and / or" used herein includes any unit and all combinations of one or more associated listed items.

[0046] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0047] To facilitate understanding of the embodiments of the present invention, several specific embodiments will be further explained below with reference to the accompanying drawings, and each embodiment does not constitute a limitation on the embodiments of the present invention.

[0048] The overall logical architecture of a system for constructing and dispatching an urban rail transit model provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the system includes the following processing units: a storage management server, a model construction server and a model scheduling application server. The model construction server is connected to the storage management server and the model scheduling application server respectively.

[0049] Storage management server, used to store and manage basic data, AI algorithms and models, transmit data, algorithms and models to the model construction server, receive feedback from the model scheduling application server and update, manage and store them;

[0050] Build a model construction server to construct and solve the model according to the application server instructions, including building a prediction detection model construction unit, an optimization model construction and solution unit, and a model online update unit, to support the model scheduling application server requirements and to update the model online according to feedback;

[0051] Build a model scheduling application server to realize the calling and evaluation of domain models of each unit, including scenario analysis unit, transport organization application unit, passenger service application unit, asset maintenance application unit, cross-domain collaborative application unit and model evaluation unit.

[0052] An architecture diagram of a storage management server provided by an embodiment of the present invention is as follows: Figure 2 As shown, the following units are included: a data storage management unit, an AI algorithm storage management unit and a model storage management unit.

[0053] Data storage management unit, used to store and manage multi-standard, multi-source and heterogeneous data of urban rail transit, including passenger flow data unit, driving data unit, service data unit, equipment data unit and asset data unit, etc.;

[0054] Among them, passenger flow data units include AFC (Automatic Fare Collection) data, passenger flow statistics, etc., driving data units include train operation data, arrival and departure data, etc., service data units include passenger complaint data, passenger assistance data, etc., equipment data units include station environment monitoring data, security inspection data, etc., asset data units include maintenance business data, production management data, etc.;

[0055] AI algorithm storage management unit, which is used to integrate the core algorithms of the system and is responsible for the storage and update of the system algorithms, including data statistics algorithm unit, decision support algorithm unit, prediction and detection algorithm unit, optimization and solution algorithm unit and exception handling algorithm unit;

[0056] Among them, the data statistics algorithm unit includes algorithms such as data analysis and data preprocessing, the decision support algorithm unit includes algorithms such as linear programming and reinforcement learning, the prediction and detection algorithm unit includes algorithms such as convolutional neural network and convolutional neural network, the optimization and solution algorithm unit includes heuristic algorithms and operations planning optimization algorithms, and the exception handling algorithm unit includes isolation forest and PCA (Principal Component Analysis), etc.

[0057] Model storage management unit, used to store, manage, optimize and update various models, including service units, organization units and asset units;

[0058] Among them, the service management units include travel service models, personalized recommendation models, etc.; the organizational units include new line opening passenger flow prediction models, transfer passenger flow prediction models, etc.; the asset units include maintenance status perception models, autonomous diagnosis models, etc.

[0059] The architecture diagram of a model construction server provided by an embodiment of the present invention is as follows: Figure 3 As shown, the following units are included: a prediction and detection model construction unit, an optimization model construction and solution unit, and a model online update unit.

[0060] The prediction detection model construction unit is used to train, generate, and compare prediction detection-related applications in the fields of transportation organization, passenger service, and asset maintenance according to actual scenario requirements, combined with demand goals and evaluation indicators, to construct appropriate prediction detection algorithms;

[0061] The prediction detection model construction unit receives the demand objectives and evaluation indicators transmitted by the model scheduling application server scenario analysis, calls the data storage management unit and the AI ​​algorithm storage management unit, generates training samples and a combination algorithm set, trains the combination algorithm, generates the optimal result through comparison, and then constructs a prediction detection algorithm;

[0062] The optimization model construction and solution unit is used for optimization and solution-related applications in the fields of transportation organization, passenger service, and asset maintenance. It trains, generates, and compares the actual scenario requirements in combination with demand goals, constraints, and decision variables to construct appropriate optimization models and solution algorithms.

[0063] The optimization model construction and solution unit receives the demand objectives, constraints, and decision variables transmitted by the model scheduling application server scenario analysis, calls the data storage management unit and the AI ​​algorithm storage management unit, generates test data and a combination algorithm set, tests the combination algorithm, generates the optimal result through comparison, and then constructs the optimization model and solution algorithm;

[0064] The model online update unit is used to update through data and scenarios, receive feedback from the model construction and solution unit, continuously calibrate the model training results, perform algorithm training and parameter optimization according to new data, adjust model objectives and constraints, and dynamically select and apply the optimal model, algorithm and parameters, including: prediction and detection model update unit, optimization model update unit and parameter update unit.

[0065] Figure 4 A processing flow chart of a prediction detection model construction unit in a model construction server, referring to Figure 4The prediction detection model construction unit receives the demand objectives and evaluation indicators transmitted by the model scheduling application server, calls the data storage management unit and the AI ​​algorithm storage management unit in the storage management server, and then generates a combination of training samples and prediction detection algorithms, generates the optimal result through comparison, and then constructs the prediction detection algorithm.

[0066] Figure 5 This is a processing flow chart of an optimization model construction and solution unit in a model construction server, refer to Figure 5 The optimization model construction and solution unit receives the demand objectives, constraints and decision variables transmitted by the model scheduling application server, calls the data storage management unit and the AI ​​algorithm storage management unit in the storage management server, and then generates a combination of test data and optimization algorithms, generates the optimal result through comparison, and then constructs the optimization model and solution algorithm.

[0067] The architecture diagram of a model scheduling application server provided by an embodiment of the present invention is as follows: Figure 6 As shown, the following units are included: scenario analysis unit, transportation organization application unit, passenger service application unit, asset maintenance application unit, cross-domain collaboration application unit and model evaluation unit.

[0068] The scenario analysis unit is used to analyze the business domain, status, elements and other related data of the application according to the demand scenario, construct the demand goals, evaluation indicators, etc., and perform model construction matching, transmit the analysis instructions and make analysis adjustments according to the application feedback;

[0069] The transport organization application unit is used to call the model generation server according to the application scenario for transport organization applications, generate models, and output the models to external practical applications. Common application scenarios supported include traffic demand forecasting, capacity configuration and evaluation optimization, passenger organization process simulation, digital plans and emergency resource scheduling, etc.

[0070] Among them, the traffic demand forecast includes passenger flow forecast for new line opening, passenger flow forecast for special events, medium- and long-term passenger flow forecast, passenger flow forecast for major events or typical days, daily short-term passenger flow forecast, etc.; capacity allocation and evaluation optimization include network capacity allocation, data statistical analysis, network train operation plan adjustment, energy-saving optimization, capacity allocation evaluation visualization, etc.; passenger organization process simulation includes real-time passenger flow simulation, full-time domain passenger flow simulation, full-time domain simulation evaluation, full-time domain simulation visualization, etc.; digital plans and emergency resource scheduling include emergency duty, emergency guarantee, emergency resource management, decision support, emergency plan management, etc.;

[0071] The passenger service application unit is used to call the model generation server according to the application scenario for passenger service applications, generate models, and output the models to external practical applications. Common application scenarios supported include passenger profiling, demand management, recommendation guidance, itinerary prediction, and travel services;

[0072] Among them, passenger portraits include passenger travel portraits, passenger job-residence migration, passenger relationship analysis, passenger job-residence analysis, passenger cluster analysis, etc. Demand management includes message push services, entertainment services, customized advertising, customized fares and passenger social networking, etc. Recommendation guidance includes optimal route recommendations, personalized services, real-time station conditions, travel planning suggestions, delays and adjustments, etc. Itinerary predictions include travel time predictions, itinerary destination predictions, itinerary congestion predictions, etc. Travel services include quick entry, targeted information release, smart station services and one-click special projects, etc.

[0073] The asset maintenance application unit is used to call the model generation server according to the application scenario for asset maintenance applications, generate models, and output the models to external actual applications. Common application scenarios supported include intelligent analysis and early warning, intelligent maintenance operations, and intelligent asset linkage;

[0074] Among them, smart analysis includes operation monitoring of key vehicle components, power supply monitoring, vehicle system detection, online monitoring of platform doors, and engineering detection. Smart maintenance operations include equipment maintenance management, construction scheduling management, track health management, and online battery monitoring. Smart asset linkage includes material storage management and asset management.

[0075] The cross-domain collaborative application unit is used to collaboratively call the transport organization application unit, passenger service application unit, and asset maintenance application unit according to application scenarios that require collaboration in various fields during the operation of urban rail transit, and output the model to external actual applications. Common application scenarios supported include: collaborative emergency response, collaborative passenger flow management, etc.

[0076] The model evaluation unit is responsible for conducting comprehensive evaluations of the model application effects and performance, and evaluating, adjusting, and updating system applications through evaluation indicators and evaluation models.

[0077] Among them, the evaluation indicators include driving management index library, equipment health index library, passenger service index library, safety energy consumption index library and passenger flow efficiency index library, and the evaluation models include hierarchical analysis method, fuzzy comprehensive evaluation method, entropy method, TOPSIS method and data envelopment analysis.

[0078] The passenger service implementation case of the construction of an urban rail transit model and the dispatching application platform system provided by the embodiment of the present invention is as follows: Figure 7 As shown, the processing steps include the following:

[0079] Step S1: The model scheduling application server analyzes the passenger service scenario to obtain the business application field, service time and space, and passenger service requirements of the passenger service.

[0080] Step S101: Generate passenger service demand targets and evaluation indicators through passenger service scenario analysis. Among them, passenger service demand targets include travel planning services, personalized recommendation services, and passenger portrait services. Travel planning services include travel OD (Origin-Destination) selection, travel obstacle avoidance, travel mode selection, and convenient services. Personalized recommendation services include travel time recommendations and travel route recommendations. Passenger portrait services include travel portraits and ticket card analysis. Evaluation indicators include passenger satisfaction, prediction accuracy, section congestion, train full load rate, departure interval, and route congestion.

[0081] Step S102: According to the demand target of the passenger service scenario analysis, the specific passenger application unit in the model scheduling application server is called, including application units such as travel planning engine, portrait information integration, and personalized travel plan recommendation. According to the evaluation index of the passenger service scenario analysis, the specific evaluation unit in the model scheduling application server is called, including evaluation methods such as hierarchical analysis method and fuzzy comprehensive evaluation. The passenger application is evaluated by the evaluation unit.

[0082] Step S103: The model scheduling application server transmits the parsing instruction to the model construction server for model construction, and receives feedback from the model construction server for server update optimization.

[0083] Step S2: The model construction server receives the application unit and model instructions transmitted by the model scheduling application server, calls the model construction server through the received specific passenger application unit, trains / solves the training samples / test data through the combination algorithm, generates results and constructs the optimal model through comparison.

[0084] Step S201: The model construction server calls data from the storage management server, including POI (Point of Interest), AFC (Automatic Fare Collection), AVL (Automatic Vehicle Location), weather data, road network data, equipment monitoring data, etc., and then generates training data / test samples through sampling and other algorithms.

[0085] Step S202: Call the AI ​​algorithm storage management unit, including recommendation algorithm, prediction algorithm, portrait generation algorithm, feature extraction algorithm and path utility calculation algorithm, etc.

[0086] Step S203: Generate a passenger service model through combination, including a travel demand prediction model, a travel time prediction model, a personalized recommendation model, a portrait label generation model, etc.

[0087] Step S3: The storage management server receives instructions from the model construction server to transmit data, algorithms and models, and updates, optimizes, stores and manages them according to the feedback from the model construction server.

[0088] In summary, the urban rail transit model construction and dispatching application platform system provided by the embodiment of the present invention provides an automatic and intelligent dispatching application system for scenarios such as passenger service, transportation organization, asset maintenance and cross-domain collaborative applications. It automatically calls the application server through scenario analysis, and predicts and detects the model or optimizes the model construction through the model construction server, and uses the storage management server for support. The present invention improves the intelligence and automation of the urban rail transit model through efficient system scheduling and management, and provides more convenient and intelligent services for individuals, enterprises, governments and other users. The present invention has important value and implementation significance for the intelligent construction and application of urban rail transit models.

[0089] The system of the present invention automatically calls the application server through scene analysis, and predicts and detects the model or optimizes the model construction through the model construction server, and uses the storage management server for support. The present invention integrates transportation organization, passenger service, asset maintenance and cross-domain business application functions, analyzes user needs and automatically builds models, realizes efficient system scheduling and management, improves the intelligence and automation of urban rail transit model construction and application, and provides more convenient and intelligent services for individuals, enterprises, governments and other users. Therefore, the present invention has important value and implementation significance for the intelligent construction and application of urban rail transit models.

[0090] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of an embodiment, and the units or processes in the accompanying drawings are not necessarily necessary for implementing the present invention.

[0091] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention or certain parts of the embodiments.

[0092] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0093] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A system for constructing and dispatching an urban rail transit model, characterized in that: include: Storage management server, model construction server and model scheduling application server; The storage management server is used to store and manage basic data, AI algorithms and models, transmit data, algorithms and models to the model construction server, receive the system algorithm transmitted by the model construction server, and update, manage and store the system algorithm; The model construction server is used to construct and solve the model according to the instructions of the model scheduling application server, support the needs of the model scheduling application server, receive feedback from the model scheduling application server and perform online model updates; The model scheduling application server is used to realize the calling and evaluation of each unit domain model, including a scene analysis unit, a transportation organization application unit, a passenger service application unit, an asset maintenance application unit, a cross-domain collaborative application unit, and a model evaluation unit.

2. The system according to claim 1, characterized in that The storage management server includes: a data storage management unit, an AI algorithm storage management unit and a model storage management unit; The data storage management unit is used to store and manage multi-standard, multi-source and heterogeneous data of urban rail transit, including passenger flow data units, driving data units, service data units, equipment data units and asset data units; The AI ​​algorithm storage management unit is used to receive the system algorithm transmitted by the model construction server, and the system algorithm includes: prediction and detection algorithm, optimization model and solution algorithm, and performs system algorithm storage and update, including data statistics algorithm unit, decision support algorithm unit, prediction and detection algorithm unit, optimization solution algorithm unit and exception handling algorithm unit; The model storage management unit is used to store, manage, optimize and update various models, including service units, organization units and asset units; the service management unit includes a travel service model and a personalized recommendation model, the organization unit includes a new line opening passenger flow prediction model and a transfer passenger flow prediction model, and the asset unit includes a maintenance status perception model and an autonomous diagnosis model.

3. The system according to claim 1, characterized in that The model construction server includes: a prediction detection model construction unit, an optimization model construction and solution unit and a model online update unit; The prediction detection model construction unit is used for applications related to prediction detection in the fields of transportation organization, passenger service and asset maintenance, receiving demand targets and evaluation indicators generated by the model scheduling application server according to actual scenario requirements, calling the data storage management unit and the AI ​​algorithm storage management unit to generate training samples and a combination algorithm set, training the combination algorithm, generating the optimal result through comparison, and then constructing a prediction detection algorithm, and transmitting the prediction detection algorithm to the storage management server; The optimization model construction and solution unit is used for optimization and solution-related applications in the fields of transportation organization, passenger service and asset maintenance, receiving demand targets, constraints and decision variables generated by the model scheduling application server through scenario analysis, calling the data storage management unit and the AI ​​algorithm storage management unit to generate test data and a combination algorithm set, testing the combination algorithm, generating the best result through comparison, and then constructing an optimization model and a solution algorithm, and transmitting the optimization model and the solution algorithm to the storage management server; The model online update unit is used to receive feedback from the optimization model construction and solution unit through data and scenario updates, continuously calibrate the model training results, perform algorithm training and parameter optimization according to new data, adjust model objectives and constraints, and dynamically select and apply the optimal model, algorithm and parameters, including: a prediction and detection model update unit, an optimization model update unit and a parameter update unit.

4. The system according to claim 1, characterized in that The model scheduling application server model includes a scenario analysis unit, a transport organization application unit, a passenger service application unit, an asset maintenance application unit, a cross-domain collaboration application unit and a model evaluation unit; The scenario analysis unit is used to analyze the business domain, state and element-related data of the application according to the demand scenario, construct the demand target, constraint conditions and decision variables, and transmit the demand target, constraint conditions and decision variables to the model construction server; The transport organization application unit is used to call the model generation server to generate models for transport organization applications according to application scenarios, and output the models to external practical applications. The supported application scenarios include traffic demand forecasting, capacity configuration and evaluation optimization, passenger organization process simulation, and digital plans and emergency resource scheduling; The passenger service application unit is used to call the model generation server to generate a model for passenger service applications according to application scenarios, and output the model to external actual applications. The supported application scenarios include passenger profiling, demand management, recommendation guidance, itinerary prediction and travel services; The asset maintenance application unit is used to call the model generation server to generate models for asset maintenance applications according to application scenarios, and output the models to external practical applications. The supported application scenarios include intelligent analysis and early warning, intelligent maintenance operations, and intelligent asset linkage; The cross-domain collaborative application unit is used to collaboratively call the transport organization application unit, the passenger service application unit, and the asset maintenance application unit according to the application scenarios that require collaboration in various fields during the operation of urban rail transit, and output the model to external practical applications. The supported application scenarios include: collaborative emergency response and collaborative passenger flow management; The model evaluation unit is used to conduct a comprehensive evaluation of the model application effect and performance, and to evaluate, adjust and update the system application through evaluation indicators and evaluation models.

5. The system according to any one of claims 1 to 4, characterized in that: The processing process of the system for implementing passenger services includes the following processing steps: Step S1: The model scheduling application server analyzes the passenger service scenario to obtain the business application field, service time and space, and passenger service requirements of the passenger service; Step S101: Generate demand targets and evaluation indicators for passenger services through passenger service scenario analysis. The passenger service demand targets include travel planning services, personalized recommendation services, and passenger portrait services. Travel planning services include travel origin-destination OD selection, travel obstacle avoidance, travel mode selection, and convenient services. Personalized recommendation services include travel time recommendations and travel route recommendations. Passenger portrait services include travel portraits and ticket analysis. Evaluation indicators include passenger satisfaction, prediction accuracy, section congestion, train full load rate, departure interval, and route congestion indicators. Step S102: calling a specific passenger application unit in the model scheduling application server according to the demand target of the passenger service scenario analysis, calling a specific evaluation unit in the model scheduling application server according to the evaluation index of the passenger service scenario analysis, and evaluating the passenger application through the evaluation unit; Step S103: the model scheduling application server transmits the parsing instruction to the model construction server for model construction, and receives feedback from the model construction server for server update optimization; Step S2: The model construction server receives the application unit and model instruction transmitted by the model scheduling application server, calls the model construction server through the received specific passenger application unit, trains / solves the training samples / test data through the combination algorithm, generates results, and constructs the optimal model through comparison; Step S201: the model construction server calls data from the storage management server and generates training data / test samples through a sampling algorithm; Step S202: the storage management server stores system algorithms, which include recommendation algorithms, prediction algorithms, portrait generation algorithms, feature extraction algorithms, and path utility calculation algorithms; Step S203: Generate a passenger service model by combination, the passenger service model including a travel demand prediction model, a travel time prediction model, a personalized recommendation model and a portrait label generation model; Step S3: The storage management server receives the instruction transmission data, algorithm and model transmitted by the model construction server, and updates, optimizes, stores and manages them according to the feedback from the model construction server.

Images