Urban rail operation management method and system based on data analysis

Abstract

The invention belongs to the technical field of urban rail operation management, and discloses an urban rail operation management method and system based on data analysis. The method comprises the following steps: setting a data code standard, and collecting a plurality of historical urban rail operation full life cycle data; constructing an urban rail operation data analysis model and an urban rail operation management model by using an artificial intelligence algorithm; according to the data code standard, collecting real-time urban rail operation full life cycle data; performing data processing on the real-time urban rail operation full life cycle data; performing data analysis by using the urban rail operation data analysis model to obtain a real-time urban rail operation data analysis result; and according to the real-time urban rail operation data analysis result, using the urban rail operation management model to generate a real-time urban rail operation management strategy, and performing urban rail operation management on the current urban rail operation project. The problems that in the prior art, a data management system is disordered, data analysis accuracy is insufficient, and the intelligent degree of urban rail operation management is low are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban rail transit operation management, and particularly relates to an urban rail transit operation management method and system based on data analysis. Background Art

[0002] With the rapid development of urbanization, urban rail transit has become increasingly important in public transportation. Urban rail transit operation management, that is, the operation management of urban rail transit, refers to a series of activities for planning, organizing, commanding, coordinating, and controlling the daily operation activities of urban rail transit systems such as subways and light rails. It ensures the safe, efficient, and orderly operation of urban rail transit systems to meet the urban public transportation demand. Urban rail transit operation management is a complex systems engineering that requires multi-department collaboration and the application of modern management techniques and methods to achieve the efficient operation and sustainable development of urban rail transit.

[0003] The existing technologies have the following defects:

[0004] 1) Confused data management system: The existing technologies do not clearly and meticulously divide the stages of the entire life cycle of urban rail transit operation, resulting in confusion and improper handling in data collection, analysis, and strategy application at different stages. Data management often lacks a unified data code standard, and the diversity of data formats and sources makes data processing work complex and error-prone.

[0005] 2) Insufficient accuracy of data analysis: Existing data analysis models are usually based on traditional statistical methods or simple machine learning algorithms, unable to fully explore the deep information and rules in the data. The models have limited processing capabilities for complex and non-linear relationships, resulting in insufficient accuracy of data analysis and difficulty in accurately predicting and coping with various emergencies in urban rail transit operation.

[0006] 3) Low level of intelligence in urban rail transit operation management: The existing methods for generating urban rail transit operation management strategies often rely on manual experience, lacking intelligent and automated decision support. The strategy generation process lacks real-time and dynamic adjustment capabilities and is difficult to adapt to the rapidly changing urban rail transit operation environment. Summary of the Invention

[0007] In order to solve the problems of the confused data management system, insufficient accuracy of data analysis, and low level of intelligence in urban rail transit operation management existing in the existing technologies, the purpose of the present invention is to provide an urban rail transit operation management method and system based on data analysis.

[0008] The technical solution adopted by the present invention is as follows:

[0009] An urban rail transit operation management method based on data analysis, comprising the following steps:

[0010] Set corresponding data code standards for the entire life cycle of urban rail transit operation, and collect a number of historical urban rail transit operation full-life cycle data with historical data codes according to the data code standards;

[0011] Based on a number of historical urban rail transit operation full-life cycle data with historical data codes, use artificial intelligence algorithms to construct an urban rail transit operation data analysis model and an urban rail transit operation management model;

[0012] According to the data code standards, collect real-time urban rail transit operation full-life cycle data with real-time data codes for the current urban rail transit operation project;

[0013] Perform data processing on the real-time urban rail transit operation full-life cycle data to obtain the data-processed real-time urban rail transit operation full-life cycle data with real-time data codes;

[0014] Use the urban rail transit operation data analysis model to perform data analysis on the data-processed real-time urban rail transit operation full-life cycle data with real-time data codes to obtain real-time urban rail transit operation data analysis results;

[0015] According to the real-time urban rail transit operation data analysis results, use the urban rail transit operation management model to generate real-time urban rail transit operation management strategies and conduct urban rail transit operation management on the current urban rail transit operation project.

[0016] Furthermore, the urban rail transit operation stage in the entire life cycle of urban rail transit operation includes the operation preparation stage, the formal operation stage, the operation maintenance stage, the operation optimization stage, and the operation end stage;

[0017] The historical urban rail transit operation full-life cycle data includes historical urban rail transit operation stage data for at least one urban rail transit operation stage;

[0018] The historical data codes include historical data code sub-segments for each urban rail transit operation stage, and each historical data code sub-segment corresponds to a historical urban rail transit operation stage data;

[0019] The real-time urban rail transit operation full-life cycle data includes real-time urban rail transit operation stage data for at least one urban rail transit operation stage;

[0020] The real-time data codes include real-time data code sub-segments for each urban rail transit operation stage, and each real-time data code sub-segment corresponds to a real-time urban rail transit operation stage data.

[0021] Furthermore, based on a number of historical urban rail transit operation full-life cycle data with historical data codes, using artificial intelligence algorithms to construct an urban rail transit operation data analysis model and an urban rail transit operation management model includes the following steps:

[0022] Preprocess a number of historical urban rail transit operation full - life - cycle data with historical data codes to obtain a number of preprocessed historical urban rail transit operation full - life - cycle data;

[0023] Perform data processing on a number of preprocessed historical urban rail transit operation full - life - cycle data to obtain a number of data - processed historical urban rail transit operation full - life - cycle data and a number of key data indicators for each historical urban rail transit operation stage data;

[0024] According to a number of data - processed historical urban rail transit operation full - life - cycle data with historical data codes, use deep - learning algorithms to construct an urban rail transit operation data analysis model and obtain a number of historical urban rail transit operation data analysis results;

[0025] According to a number of historical urban rail transit operation data analysis results, use reinforcement - learning algorithms to construct an urban rail transit operation management model.

[0026] Furthermore, performing data processing on a number of preprocessed historical urban rail transit operation full - life - cycle data to obtain a number of data - processed historical urban rail transit operation full - life - cycle data and a number of key data indicators for each historical urban rail transit operation stage data includes the following steps:

[0027] Perform data clustering on a number of preprocessed historical urban rail transit operation full - life - cycle data to obtain a number of clustering centers and corresponding clustering clusters;

[0028] Perform principal - component analysis on the data - cleaned historical urban rail transit operation full - life - cycle data corresponding to the clustering centers to obtain a number of key data indicators for each historical urban rail transit operation stage data in each clustering center;

[0029] According to a number of key data indicators of the clustering centers, perform data dimensionality reduction on all data - cleaned historical urban rail transit operation full - life - cycle data to obtain a number of data - processed historical urban rail transit operation full - life - cycle data.

[0030] Furthermore, the urban rail transit operation data analysis model is constructed based on the Double LSTM - Attention - MLP algorithm, and the urban rail transit operation data analysis model includes a first feature extraction module and a second feature extraction module both constructed based on the LSTM algorithm, an attention weight module constructed based on the Attention mechanism, and an urban rail transit operation data analysis module constructed based on the MLP algorithm.

[0031] Furthermore, the urban rail transit operation management model is constructed based on the MOPPO algorithm, and the urban rail transit operation management model is provided with a set of objective functions, an experience replay pool, an Actor network, a Critic network, and an agent.

[0032] Further, according to the analysis results of several historical urban rail transit operation data, use the reinforcement learning algorithm to construct an urban rail transit operation management model, including the following steps:

[0033] Take the problem of generating urban rail transit operation management strategies as the simulation environment, use the MOPPO algorithm to construct an initial urban rail transit operation management model, and set a set of objective functions, an experience replay pool, an Actor network, a Critic network, and an agent for the initial urban rail transit operation management model;

[0034] Analyze the analysis results of historical urban rail transit operation data to obtain several analysis states of historical urban rail transit operation data. Define the state space of the agent according to the several analysis states of historical urban rail transit operation data, and define the action space of the agent according to several preset urban rail transit operation management actions;

[0035] Based on any objective function in the set of objective functions, pre-train the initial urban rail transit operation management model according to the analysis results of several historical urban rail transit operation data to obtain a pre-trained urban rail transit operation management model, and generate several historical urban rail transit operation management strategies and corresponding historical urban rail transit operation management experiences;

[0036] Use the Critic network of the pre-trained urban rail transit operation management model to obtain the rewards of several historical urban rail transit operation management strategies, and obtain an optimized Critic network;

[0037] Optimize the Actor network of the pre-trained urban rail transit operation management model according to several rewards to obtain an optimized Actor network;

[0038] Traverse all the objective functions in the set of objective functions, repeat the above steps to obtain the final urban rail transit operation management model with an optimized Actor network and an optimized Critic network, and store several historical urban rail transit operation management experiences in the experience replay pool.

[0039] Further, perform data processing on the real-time urban rail transit full life cycle data to obtain the data-processed real-time urban rail transit full life cycle data with real-time data codes, including the following steps:

[0040] Perform standardization processing on the real-time urban rail transit full life cycle data to obtain the standardized real-time urban rail transit full life cycle data;

[0041] Obtain the Euclidean distances between the standardized real-time urban rail transit full life cycle data and several clustering centers, and use the clustering center with the closest Euclidean distance as the target clustering center;

[0042] Based on several target key data indicators of the target clustering center, perform data dimensionality reduction on the real-time urban rail operation full life cycle data after standardization processing to obtain the real-time urban rail operation full life cycle data after data processing with real-time data codes set.

[0043] Furthermore, according to the real-time urban rail operation data analysis results, use the urban rail operation management model to generate real-time urban rail operation management strategies, and perform urban rail operation management on the current urban rail operation project, including the following steps:

[0044] According to the real-time urban rail operation data analysis results, select the most suitable objective function from the set of objective functions of the urban rail operation management model;

[0045] Analyze the real-time urban rail operation data analysis results to obtain several real-time urban rail operation data analysis states, and update the state space of the intelligent agent according to the several real-time urban rail operation data analysis states to obtain the updated state space;

[0046] Extract several historical urban rail operation management experiences from the experience replay pool, and update the action space of the intelligent agent according to several preset urban rail operation management actions of the several historical urban rail operation management experiences to obtain the updated action space;

[0047] Based on the most suitable objective function, use the intelligent agent of the urban rail operation management model to control the Actor network, and in the updated action space, generate the probability distribution of all possible urban rail operation management actions corresponding to each real-time urban rail operation data analysis state in the updated state space;

[0048] Take the possible urban rail operation management action with the highest probability distribution in the updated action space as the executed urban rail operation management action for the real-time urban rail operation data analysis state, and integrate the executed urban rail operation management actions of all real-time urban rail operation data analysis states to obtain the real-time urban rail operation management strategy;

[0049] Send the real-time urban rail operation management strategy to the data server of the current urban rail operation project, and perform urban rail operation management on the current urban rail operation project according to the real-time urban rail operation management strategy.

[0050] An urban rail operation management system based on data analysis for implementing the urban rail operation management method. The system includes a historical data collection unit, an artificial intelligence model construction unit, a real-time data collection unit, a real-time data processing unit, a real-time data analysis unit, and an urban rail operation management unit connected in sequence.

[0051] The beneficial effects of the present invention are:

[0052] A method and system for urban rail transit operation management based on data analysis provided by the present invention provides a standardized data management system. By setting a unified data code standard for the entire life cycle of urban rail transit operation, the standardization of data collection, storage, processing, and analysis is achieved, reducing the situations of data confusion and error handling. The clear and detailed stage division ensures the accurate correspondence of data in each stage of the life cycle, improving the efficiency and accuracy of data management. An advanced artificial intelligence algorithm is used to construct an urban rail transit operation data analysis model, which can deeply excavate the deep information and laws in the data, greatly improving the ability to handle complex and non-linear relationships. The accurate data analysis results provide more reliable prediction and decision-making support for urban rail transit operation, effectively coping with various emergencies. The constructed urban rail transit operation management model realizes intelligent and automated decision-making support, reducing the dependence on manual experience. The strategy generation process has real-time and dynamic adjustment capabilities, can quickly adapt to the changes in the urban rail transit operation environment, improve the management efficiency and response speed, and improve the intelligent level of urban rail transit operation management.

[0053] Other beneficial effects of the present invention will be further described in the specific implementation manners. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a flowchart of the method for urban rail transit operation management based on data analysis in the present invention.

[0055] Figure 2 is a structural block diagram of the system for urban rail transit operation management based on data analysis in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments.

[0057] Embodiment 1:

[0058] As Figure 1 shown, this embodiment provides a method for urban rail transit operation management based on data analysis, including the following steps:

[0059] S1: Set corresponding data code standards for the entire life cycle of urban rail transit operation, and collect a number of historical urban rail transit operation full-life cycle data with historical data codes according to the data code standards;

[0060] The urban rail transit operation stages in the entire life cycle of urban rail transit operation include the operation preparation stage, the formal operation stage, the operation maintenance stage, the operation optimization stage, and the operation end stage;

[0061] The historical urban rail transit operation full-life cycle data includes historical urban rail transit operation stage data of at least one urban rail transit operation stage;

[0062] Historical operation preparation stage data includes trial operation data (train operation time, speed, acceleration, signal system response time, accuracy, and operation status of station facilities), personnel training data (training records, assessment results, and employee skill certification status), and operation plan data (train operation diagram, timetable, passenger flow prediction, and transport capacity planning);

[0063] Historical formal operation stage data includes train operation organization data (actual train operation time, speed, interval, train punctuality rate, operation diagram fulfillment rate, and real-time signal system data), passenger service data (passenger flow, occupancy rate, ticket revenue, passenger complaints, and usage of station service facilities), and safety management data (accident records, fault statistics, safety inspections, hidden danger investigation records, emergency drills, and emergency response data);

[0064] Historical operation and maintenance stage data includes daily maintenance data (equipment inspection records, maintenance logs, work records of maintenance personnel, maintenance costs, and material consumption), regular maintenance data (maintenance plans, implementation status, equipment performance test results, maintenance costs, and time statistics), and fault handling data (fault reports, handling time, fault cause analysis, recurrence situation, and fault impact assessment);

[0065] Historical operation optimization stage data includes service improvement data (results of passenger satisfaction surveys, implementation effects of service improvement measures, and evaluation of new technology application effects), efficiency improvement data (operation efficiency indicators, evaluation of technology improvement and management optimization effects, cost control, and benefit analysis), and market analysis data (passenger flow change trends, market demand analysis, operation conditions of competitors, and market share);

[0066] Historical operation end stage data includes operation handover or termination data (handover agreements, termination reasons, operation asset evaluation, and disposal records), and asset disposal data (asset sale and transfer situations, disposal income, and cost analysis):

[0067] Historical data codes include historical data code sub-segments for each urban rail operation stage, and each historical data code sub-segment corresponds to historical urban rail operation stage data;

[0068] S2: According to a number of historical urban rail operation full life cycle data set with historical data codes, use artificial intelligence algorithms to construct an urban rail operation data analysis model and an urban rail operation management model, including the following steps:

[0069] S2-1: Preprocess a number of historical urban rail operation full life cycle data set with historical data codes to obtain a number of preprocessed historical urban rail operation full life cycle data;

[0070] The preprocessing includes data cleaning and standardization processes carried out in sequence. Data cleaning eliminates the contamination of the dataset by duplicate and incorrect data, improving data quality. The standardization process includes magnitude normalization and data format processing to obtain data that the model can recognize, providing data support for subsequent model training;

[0071] S2-2: Perform data processing on a number of historical urban rail transit operation full life cycle data after preprocessing to obtain a number of key data indicators for the historical urban rail transit operation full life cycle data and each historical urban rail transit operation stage data after data processing, including the following steps:

[0072] S2-2-1: Use the Fuzzy C-mean (FCM) clustering algorithm to perform data clustering on a number of historical urban rail transit operation full life cycle data after preprocessing to obtain a number of cluster centers and corresponding cluster clusters, including the following steps:

[0073] S2-2-1-1: Select the clustering parameters of the FCM clustering algorithm, including the fuzzy factor m and the total number of cluster centers;

[0074] S2-2-1-2: According to the clustering parameters, use the FCM clustering algorithm to cluster a number of historical urban rail transit operation full life cycle data after preprocessing to obtain a number of initial cluster centers;

[0075] S2-2-1-3: According to the Euclidean distance between each historical urban rail transit operation full life cycle data after preprocessing and a number of initial cluster centers, set the corresponding fuzzy membership degree for each initial cluster center;

[0076] The formula is:

[0077] d ij =||x i -z j || 2

[0078] In the formula, d ij is the Euclidean distance between the i-th historical urban rail transit operation full life cycle data after preprocessing and the j-th cluster center; x i is the i-th historical urban rail transit operation full life cycle data after preprocessing; z j is the j-th cluster center; i is the data indicator; j is the cluster center indicator;

[0079] S2-2-1-4: Update the cluster centers according to the fuzzy membership degree to obtain a corresponding number of updated cluster centers;

[0080] The formula for updating the fuzzy membership degree is:

[0081]

[0082] where x i is the historical urban rail transit operation full - life - cycle data after the i - th pre - processing; i is the data index; j and k are both cluster - center indices; c is the total number of cluster centers; d ij and d ik are the distances from the historical urban rail transit operation full - life - cycle data after the i - th pre - processing to the j - th and k - th cluster centers; u ij is the updated fuzzy membership degree that the historical urban rail transit operation full - life - cycle data after the i - th pre - processing belongs to the j - th cluster center;

[0083] The formula for updating the cluster center is:

[0084]

[0085] where z' j is the updated j - th cluster center; m is the fuzzy factor; i is the data index; n' is the total number of data; j is the cluster - center index; x i is the historical urban rail transit operation full - life - cycle data after the i - th pre - processing; u ij is the fuzzy membership degree that the historical urban rail transit operation full - life - cycle data after the i - th pre - processing belongs to the j - th cluster center;

[0086] S2 - 2 - 1 - 5: Use the Lagrange multiplier method to calculate the merging function, and obtain the merging - function value and the merging - function change value. The formula is:

[0087]

[0088] where J t and J t-1 are the merging - function values of the Lagrange multiplier method for the t - th and (t - 1) - th iterations; ΔJ t is the corresponding change value; λ' i is the i - th characteristic parameter; t is the iteration - number index; u ij is the membership degree that the historical urban rail transit operation full - life - cycle data after the i - th pre - processing belongs to the j - th cluster center; m is the fuzzy factor; i is the data index; n' is the total number of data; j is the cluster - center index; c is the total number of cluster centers; d ij is the Euclidean distance from the historical urban rail transit operation full - life - cycle data after the i - th pre - processing to the j - th cluster center;

[0089] S2 - 2 - 1 - 6: Repeat the above steps. If the merging - function value is greater than the function threshold, or the merging - function change value is greater than the change - value threshold, then continue to update the cluster center; otherwise, take the current cluster center as the final cluster center;

[0090] S2-2-1-7: Divide the preprocessed historical urban rail transit operation full life cycle data into the nearest final clustering center in terms of Euclidean distance to obtain the corresponding clustering clusters;

[0091] S2-2-2: Use the Principal Component Analysis (PCA) algorithm to perform principal component analysis on the preprocessed historical urban rail transit operation full life cycle data corresponding to the clustering center, and obtain several key data indicators for the data of each historical urban rail transit operation stage in each clustering center;

[0092] S2-2-3: According to the several key data indicators of the clustering center, perform data dimensionality reduction on all preprocessed historical urban rail transit operation full life cycle data to obtain several processed historical urban rail transit operation full life cycle data;

[0093] S2-3: According to several processed historical urban rail transit operation full life cycle data set with historical data codes, use a deep learning algorithm to construct an urban rail transit operation data analysis model and obtain several historical urban rail transit operation data analysis results;

[0094] The urban rail transit operation data analysis model is constructed based on the Double Long Short-Term Memory (LSTM)-Attention-Multilayer Perceptron (MLP) algorithm, and the urban rail transit operation data analysis model includes a first feature extraction module and a second feature extraction module both constructed based on the LSTM algorithm, an attention weight module constructed based on the Attention mechanism, and an urban rail transit operation data analysis module constructed based on the MLP algorithm;

[0095] S2-4: According to several historical urban rail transit operation data analysis results, use a reinforcement learning algorithm to construct an urban rail transit operation management model;

[0096] The urban rail transit operation management model is constructed based on the Multi-Objective Proximal Policy Optimization (MOPPO) algorithm, and the urban rail transit operation management model is set with an objective function set, an experience replay pool, an Actor network, a Critic network, and an agent;

[0097] The Actor network is responsible for outputting the probability distribution of the actions that should be taken in a given state. The goal is to learn an optimal policy, that is, to maximize the long-term cumulative reward. In a continuous action space, the Actor network usually outputs a mean, as well as optional variance parameters, to describe the probability distribution of the actions. The Critic network is responsible for evaluating the value of a given state, that is, predicting the expected return that can be obtained starting from this state and following the current policy. It usually outputs a scalar value representing the value of the state or the state-action value. The experience replay pool is used to store historical experiences for reuse during the training process. The set of objective functions includes functions that define multiple urban rail operation management objectives, including minimizing the urban rail operation cost objective, maximizing the urban rail operation management efficiency, minimizing the urban rail operation management response time, and maximizing the urban rail operation revenue, etc.;

[0098] According to the analysis results of several historical urban rail operation data, use the reinforcement learning algorithm to construct an urban rail operation management model, including the following steps:

[0099] S2-4-1: Regard the urban rail operation management strategy generation problem as a simulation environment, use the MOPPO algorithm to construct an initial urban rail operation management model, and set a set of objective functions, an experience replay pool, an Actor network, a Critic network, and an agent for the initial urban rail operation management model;

[0100] S2-4-2: Analyze the analysis results of historical urban rail operation data to obtain several historical urban rail operation data analysis states. According to several historical urban rail operation data analysis states, define the state space of the agent, and define the action space of the agent according to several preset urban rail operation management actions;

[0101] S2-4-3: Based on any objective function in the set of objective functions, according to the analysis results of several historical urban rail operation data, pre-train the initial urban rail operation management model to obtain a pre-trained urban rail operation management model, and generate several historical urban rail operation management strategies and corresponding historical urban rail operation management experiences;

[0102] S2-4-4: Use the Critic network of the pre-trained urban rail operation management model to obtain the rewards of several historical urban rail operation management strategies, and obtain an optimized Critic network;

[0103] S2-4-5: Optimize the Actor network of the pre-trained urban rail operation management model according to several rewards to obtain an optimized Actor network;

[0104] S2-4-6: Traverse all the objective functions in the objective function set, repeat the above steps to obtain the final urban rail operation management model with an optimized Actor network and an optimized Critic network, and store a number of historical urban rail operation management experiences in the experience replay pool;

[0105] S3: According to the data code standard, collect the real-time urban rail operation full life cycle data with real-time data codes set for the current urban rail operation project, including the following steps:

[0106] S3-1: Collect the real-time urban rail operation stage data of each urban rail operation stage of the current urban rail operation project, and set the corresponding real-time data code sub-segments for each real-time urban rail operation stage data according to the data code standard;

[0107] The real-time urban rail operation full life cycle data includes the real-time urban rail operation stage data of at least one urban rail operation stage;

[0108] S3-2: Integrate all the real-time data code sub-segments to obtain the real-time data code of the real-time urban rail operation full life cycle data;

[0109] The real-time data code includes the real-time data code sub-segments of each urban rail operation stage, and each real-time data code sub-segment corresponds to a real-time urban rail operation stage data;

[0110] S3-3: Integrate all the real-time urban rail operation stage data to obtain the real-time urban rail operation full life cycle data with real-time data codes set for the current urban rail operation project;

[0111] S4: Perform data processing on the real-time urban rail operation full life cycle data to obtain the data-processed real-time urban rail operation full life cycle data with real-time data codes set, including the following steps:

[0112] S4-1: Perform standardization processing on the real-time urban rail operation full life cycle data to obtain the standardized real-time urban rail operation full life cycle data;

[0113] S4-2: Obtain the Euclidean distances between the standardized real-time urban rail operation full life cycle data and several clustering centers, and use the clustering center with the closest Euclidean distance as the target clustering center;

[0114] S4-3: Perform data dimensionality reduction on the standardized real-time urban rail operation full life cycle data according to several target key data indicators of the target clustering center to obtain the data-processed real-time urban rail operation full life cycle data with real-time data codes set;

[0115] S5: Use the urban rail operation data analysis model to perform data analysis on the real-time urban rail operation full life cycle data after data processing with real-time data codes, and obtain the real-time urban rail operation data analysis results, including the following steps:

[0116] Use the first feature extraction module of the urban rail operation data analysis model to extract the first real-time feature of the real-time urban rail operation full life cycle data after data processing;

[0117] Use the second feature extraction module of the urban rail operation data analysis model to extract the second real-time feature of the real-time data codes;

[0118] According to the preset attention weight value, use the attention weight module of the urban rail operation data analysis model to perform weighted fusion on the first real-time feature and the second real-time feature to obtain the real-time weighted fusion feature;

[0119] According to the real-time weighted fusion feature, use the urban rail operation data analysis module of the urban rail operation data analysis model to perform data analysis and obtain the real-time urban rail operation data analysis results;

[0120] S6: According to the real-time urban rail operation data analysis results, use the urban rail operation management model to generate real-time urban rail operation management strategies and perform urban rail operation management on the current urban rail operation project, including the following steps:

[0121] S6-1: According to the real-time urban rail operation data analysis results, select the most appropriate objective function from the set of objective functions of the urban rail operation management model;

[0122] S6-2: Analyze the real-time urban rail operation data analysis results to obtain several real-time urban rail operation data analysis states, and update the state space of the intelligent agent according to the several real-time urban rail operation data analysis states to obtain the updated state space;

[0123] S6-3: Extract several historical urban rail operation management experiences from the experience replay pool, and update the action space of the intelligent agent according to several preset urban rail operation management actions of the several historical urban rail operation management experiences to obtain the updated action space;

[0124] S6-4: Based on the most appropriate objective function, use the intelligent agent of the urban rail operation management model to control the Actor network, and generate the probability distribution of all possible urban rail operation management actions corresponding to each real-time urban rail operation data analysis state in the updated action space in the updated state space;

[0125] S6-6: Take the urban rail transit operation management action with the highest probability distribution in the updated action space as the urban rail transit operation management action for execution in the real-time urban rail transit operation data analysis state, and integrate all the urban rail transit operation management actions for execution in the real-time urban rail transit operation data analysis state to obtain a real-time urban rail transit operation management strategy;

[0126] S6-7: Send the real-time urban rail transit operation management strategy to the data server of the current urban rail transit operation project, and perform urban rail transit operation management on the current urban rail transit operation project according to the real-time urban rail transit operation management strategy.

[0127] Embodiment 2:

[0128] As Figure 2 shown, this embodiment provides an urban rail transit operation management system based on data analysis for implementing the urban rail transit operation management method. The system includes a historical data collection unit, an artificial intelligence model construction unit, a real-time data collection unit, a real-time data processing unit, a real-time data analysis unit, and an urban rail transit operation management unit that are connected in sequence;

[0129] The historical data collection unit is used to set corresponding data code standards for the entire life cycle of urban rail transit operation, and collect a number of historical urban rail transit operation full-life cycle data with historical data codes according to the data code standards;

[0130] The artificial intelligence model construction unit is used to construct an urban rail transit operation data analysis model and an urban rail transit operation management model using artificial intelligence algorithms based on a number of historical urban rail transit operation full-life cycle data with historical data codes;

[0131] The real-time data collection unit is used to collect real-time urban rail transit operation full-life cycle data with real-time data codes of the current urban rail transit operation project according to the data code standards;

[0132] The real-time data processing unit is used to process the real-time urban rail transit operation full-life cycle data to obtain data-processed real-time urban rail transit operation full-life cycle data with real-time data codes;

[0133] The real-time data analysis unit is used to perform data analysis on the data-processed real-time urban rail transit operation full-life cycle data with real-time data codes using the urban rail transit operation data analysis model to obtain real-time urban rail transit operation data analysis results;

[0134] The urban rail transit operation management unit is used to generate a real-time urban rail transit operation management strategy using the urban rail transit operation management model based on the real-time urban rail transit operation data analysis results, and perform urban rail transit operation management on the current urban rail transit operation project.

[0135] A method and system for urban rail transit operation management based on data analysis provided by the present invention provides a standardized data management system. By setting a unified data code standard for the entire life cycle of urban rail transit operation, the standardization of data collection, storage, processing, and analysis is realized, reducing the situations of data confusion and incorrect processing. The clear and detailed stage division ensures the accurate correspondence of data in each stage of the life cycle, improving the efficiency and accuracy of data management. An urban rail transit operation data analysis model is constructed using advanced artificial intelligence algorithms, which can deeply mine the deep information and rules in the data, greatly improving the ability to process complex and non-linear relationships. The accurate data analysis results provide more reliable prediction and decision-making support for urban rail transit operation, effectively coping with various emergencies. The constructed urban rail transit operation management model realizes intelligent and automated decision-making support, reducing the dependence on manual experience. The strategy generation process has real-time and dynamic adjustment capabilities, can quickly adapt to the changes in the urban rail transit operation environment, improve the management efficiency and response speed, and improve the intelligent level of urban rail transit operation management.

[0136] The present invention is not limited to the above optional implementation manners. Any person can obtain other various forms of products under the inspiration of the present invention. The above specific implementation manners should not be understood as limiting the protection scope of the present invention. The protection scope of the present invention should be defined by the claims, and the specification can be used to interpret the claims.

Claims

1. An urban rail transit operation management method based on data analysis, characterized in that: Including the following steps: Set corresponding data code standards for the whole life cycle of urban rail transit operation, and collect a number of historical urban rail transit operation whole life cycle data with historical data codes according to the data code standards; According to a number of historical urban rail transit operation whole life cycle data with historical data codes, use artificial intelligence algorithms to construct an urban rail transit operation data analysis model and an urban rail transit operation management model; According to the data code standards, collect real-time urban rail transit operation whole life cycle data with real-time data codes for the current urban rail transit operation project; Perform data processing on the real-time urban rail transit operation whole life cycle data to obtain data-processed real-time urban rail transit operation whole life cycle data with real-time data codes; Use the urban rail transit operation data analysis model to perform data analysis on the data-processed real-time urban rail transit operation whole life cycle data with real-time data codes to obtain real-time urban rail transit operation data analysis results; According to the real-time urban rail transit operation data analysis results, use the urban rail transit operation management model to generate real-time urban rail transit operation management strategies and conduct urban rail transit operation management on the current urban rail transit operation project.

2. The urban rail transit operation management method based on data analysis according to claim 1, wherein: The urban rail transit operation stage of the urban rail transit operation whole life cycle includes an operation preparation stage, a formal operation stage, an operation maintenance stage, an operation optimization stage, and an operation end stage; The historical urban rail transit operation whole life cycle data includes historical urban rail transit operation stage data of at least one urban rail transit operation stage; The historical data code includes historical data code sub-segments for each urban rail transit operation stage, and each historical data code sub-segment corresponds to a piece of historical urban rail transit operation stage data; The real-time urban rail transit operation whole life cycle data includes real-time urban rail transit operation stage data of at least one urban rail transit operation stage; The real-time data code includes real-time data code sub-segments for each urban rail transit operation stage, and each real-time data code sub-segment corresponds to a piece of real-time urban rail transit operation stage data.

3. The urban rail transit operation management method based on data analysis according to claim 2, wherein: According to a number of historical urban rail transit operation whole life cycle data with historical data codes, use artificial intelligence algorithms to construct an urban rail transit operation data analysis model and an urban rail transit operation management model, including the following steps: Preprocess a number of historical urban rail transit operation whole life cycle data with historical data codes to obtain a number of preprocessed historical urban rail transit operation whole life cycle data; Perform data processing on a number of preprocessed historical urban rail transit operation whole life cycle data to obtain a number of data-processed historical urban rail transit operation whole life cycle data and a number of key data indicators for each historical urban rail transit operation stage data; According to a number of data-processed historical urban rail transit operation whole life cycle data with historical data codes, use deep learning algorithms to construct an urban rail transit operation data analysis model and obtain a number of historical urban rail transit operation data analysis results; According to a number of historical urban rail transit operation data analysis results, use reinforcement learning algorithms to construct an urban rail transit operation management model.

4. The urban rail transit operation management method based on data analysis according to claim 3, wherein: Perform data processing on a number of preprocessed historical urban rail transit operation whole life cycle data to obtain a number of data-processed historical urban rail transit operation whole life cycle data and a number of key data indicators for each historical urban rail transit operation stage data, including the following steps: Perform data clustering on the historical urban rail transit operation full life cycle data after several preprocessings to obtain several clustering centers and corresponding clustering clusters; Perform principal component analysis on the historical urban rail transit operation full life cycle data corresponding to the clustering centers after data cleaning to obtain several key data indicators for the data of each historical urban rail transit operation stage in each clustering center; According to the several key data indicators of the clustering centers, perform data dimensionality reduction on all the historical urban rail transit operation full life cycle data after data cleaning to obtain several processed historical urban rail transit operation full life cycle data.

5. A method for urban rail transit operation management based on data analysis according to claim 4, characterized in that: The urban rail transit operation data analysis model is constructed based on the Double LSTM-Attention-MLP algorithm, and the urban rail transit operation data analysis model includes a first feature extraction module and a second feature extraction module both constructed based on the LSTM algorithm, an attention weight module constructed based on the Attention mechanism, and an urban rail transit operation data analysis module constructed based on the MLP algorithm.

6. The urban rail transit operation management method based on data analysis according to claim 5, characterized in that: The urban rail transit operation management model is constructed based on the MOPPO algorithm, and the urban rail transit operation management model is provided with a set of objective functions, an experience replay pool, an Actor network, a Critic network, and an agent.

7. The urban rail transit operation management method based on data analysis according to claim 6, characterized in that: According to the analysis results of several historical urban rail transit operations, use the reinforcement learning algorithm to construct an urban rail transit operation management model, including the following steps: Take the urban rail transit operation management strategy generation problem as a simulation environment, use the MOPPO algorithm to construct an initial urban rail transit operation management model, and set a set of objective functions, an experience replay pool, an Actor network, a Critic network, and an agent for the initial urban rail transit operation management model; Analyze the analysis results of the historical urban rail transit operations to obtain several historical urban rail transit operation analysis states. According to the several historical urban rail transit operation analysis states, define the state space of the agent, and define the action space of the agent according to several preset urban rail transit operation management actions; Based on any one of the objective functions in the set of objective functions, according to the analysis results of several historical urban rail transit operations, pre-train the initial urban rail transit operation management model to obtain a pre-trained urban rail transit operation management model, and generate several historical urban rail transit operation management strategies and corresponding historical urban rail transit operation management experiences; Use the Critic network of the pre-trained urban rail transit operation management model to obtain the rewards of several historical urban rail transit operation management strategies and obtain an optimized Critic network; Optimize the Actor network of the pre-trained urban rail transit operation management model according to several rewards to obtain an optimized Actor network; Traverse all the objective functions in the set of objective functions, repeat the above steps to obtain the final urban rail transit operation management model with an optimized Actor network and an optimized Critic network, and store several historical urban rail transit operation management experiences in the experience replay pool.

8. The urban rail transit operation management method based on data analysis according to claim 7, characterized in that: Perform data processing on the real-time urban rail transit operation full life cycle data to obtain the processed real-time urban rail transit operation full life cycle data with real-time data codes, including the following steps: Perform standardization processing on the real-time urban rail transit operation full life cycle data to obtain the standardized real-time urban rail transit operation full life cycle data; Obtain the Euclidean distances between the real-time urban rail transit operation full life cycle data after standardization and several clustering centers, and take the clustering center with the closest Euclidean distance as the target clustering center; According to several target key data indicators of the target clustering center, perform data dimensionality reduction on the real-time urban rail transit operation full life cycle data after standardization to obtain the real-time urban rail transit operation full life cycle data after data processing with real-time data codes set.

9. The urban rail transit operation management method based on data analysis according to claim 8, characterized in that: According to the real-time urban rail transit operation data analysis results, use the urban rail transit operation management model to generate real-time urban rail transit operation management strategies and conduct urban rail transit operation management on the current urban rail transit operation project, including the following steps: According to the real-time urban rail transit operation data analysis results, select the most suitable objective function from the set of objective functions of the urban rail transit operation management model; Analyze the real-time urban rail transit operation data analysis results to obtain several real-time urban rail transit operation data analysis states, and update the state space of the intelligent agent according to the several real-time urban rail transit operation data analysis states to obtain an updated state space; Extract several historical urban rail transit operation management experiences from the experience replay pool, and update the action space of the intelligent agent according to several preset urban rail transit operation management actions of the several historical urban rail transit operation management experiences to obtain an updated action space; Based on the most suitable objective function, use the intelligent agent of the urban rail transit operation management model to control the Actor network, and in the updated action space, generate the probability distribution of all possible urban rail transit operation management actions corresponding to each real-time urban rail transit operation data analysis state in the updated state space; Take the urban rail transit operation management action with the highest probability distribution in the updated action space as the executed urban rail transit operation management action for the real-time urban rail transit operation data analysis state, and integrate the executed urban rail transit operation management actions of all real-time urban rail transit operation data analysis states to obtain the real-time urban rail transit operation management strategy; Send the real-time urban rail transit operation management strategy to the data server of the current urban rail transit operation project, and conduct urban rail transit operation management on the current urban rail transit operation project according to the real-time urban rail transit operation management strategy.

10. An urban rail transit operation management system based on data analysis, which is used to implement the urban rail transit operation management method as described in any one of claims 1-9, and is characterized in that: The described system includes a historical data acquisition unit, an artificial intelligence model construction unit, a real-time data acquisition unit, a real-time data processing unit, a real-time data analysis unit, and an urban rail transit operation management unit that are connected in sequence.

Images