Smart city construction system based on big data

Abstract

The invention provides a smart city construction system based on big data, and relates to the technical field of smart cities, and the system comprises a traffic data collection module, a transmission and preprocessing module, a flow prediction and analysis module, an intelligent signal optimization module, a parking lot scheduling and induction module, a decision support module and a user service module. The traffic flow prediction and analysis module predicts a traffic state in a set time length in the future through a space-time diagram convolutional network in combination with a seasonal autoregression model; and the intelligent traffic signal optimization module is used for dynamically adjusting signal lamp timing based on a deep reinforcement learning algorithm and realizing intersection linkage through a regional cooperative control algorithm. According to the invention, the space-time diagram convolutional network, the deep reinforcement learning intelligent algorithm, the block chain and the distributed computing technology are fused to construct a multi-source data-driven traffic whole-flow intelligent management system, so that the whole-chain intelligence from real-time sensing and dynamic optimization to cross-platform service is realized; and the collaboration, the prediction accuracy and the service ecological openness of the urban traffic system are obviously improved.

Description

Technical Field

[0001] The present invention relates to the field of smart city technology, and in particular to a smart city construction system based on big data. Background Art

[0002] With the acceleration of global urbanization, urban populations and vehicle ownership are growing dramatically, posing significant challenges to traditional urban traffic management models. Problems such as traffic congestion, low parking utilization, and persistently high carbon emissions not only exacerbate energy consumption but also severely impact residents' quality of life and sustainable urban development. Existing urban transportation systems often suffer from limitations such as limited data collection, delayed analysis and decision-making, and insufficient coordination among subsystems. These limitations make it difficult for traffic management to adapt to dynamic, real-time demands.

[0003] In the field of traffic control, single-point signal control often uses fixed timing schemes, which cannot dynamically respond to real-time traffic demands. Regional coordinated control relies on preset timed linkage strategies, making it difficult to cope with dynamic scenarios such as sudden congestion or traffic accidents. Although some systems have incorporated traditional machine learning algorithms, they lack effective modeling of spatiotemporal correlations, resulting in insufficient prediction accuracy. This is particularly true in complex road networks, where the effectiveness of multi-intersection coordinated optimization is limited.

[0004] The development of big data, artificial intelligence, and the Internet of Things (IoT) offers new avenues for addressing these challenges. Algorithms such as spatiotemporal graph neural networks and reinforcement learning demonstrate significant advantages in traffic prediction and control, while distributed computing frameworks offer efficient solutions for processing multi-source, heterogeneous data. However, existing technologies have yet to develop a comprehensive ecosystem integrating data collection, intelligent analysis, dynamic optimization, and precise services. In particular, gaps remain in multi-module collaboration mechanisms and cross-platform service integration.

[0005] Therefore, it is necessary to provide a smart city construction system based on big data to solve the above technical problems. Summary of the Invention

[0006] The present invention provides a smart city construction system based on big data, which solves the problems raised in the above background technology.

[0007] To solve the above technical problems, the present invention provides a smart city construction system based on big data, comprising: Traffic data collection module, used to obtain real-time road traffic, parking lot status, mobile terminal travel trajectory and public transportation operation data; The data transmission and preprocessing module is used to encrypt and transmit heterogeneous data, clean it, unify its format, and align it in time and space; Traffic flow prediction and analysis module, which uses a spatiotemporal graph convolutional network combined with a seasonal autoregressive model to predict traffic conditions for a set time period in the future; Intelligent traffic signal optimization module, which dynamically adjusts traffic light timing based on deep reinforcement learning algorithms and realizes intersection linkage through regional collaborative control algorithms;

[0008] The parking lot intelligent scheduling and guidance module uses the gradient boosting tree to predict parking space demand, combines it with the improved collaborative filtering algorithm to recommend the best parking lot for car owners, and dynamically guides them through multiple channels.

[0009] Preferably, the traffic data collection module includes: Real-time collection of traffic flow parameters and environmental data through a distributed sensor network, including geomagnetic sensor arrays, microwave radar detectors, video surveillance equipment, and meteorological monitoring devices. Traffic flow parameters include vehicle volume, speed, and lane occupancy; Communicate with the parking management system through a standardized data interface to obtain parking space occupancy status, geographic location and charging standards in real time. The data interface supports RESTful API or message queue protocol; Establish a data sharing mechanism with third-party map applications to obtain anonymized user travel trajectories, real-time traffic reports, and navigation request data, and use data desensitization technology to protect user privacy; Access the public transportation system to obtain real-time vehicle location, operating status and passenger flow data collected through on-board sensors or ticketing systems.

[0010] Preferably, the data transmission and preprocessing module includes an encryption transmission unit and a data management unit; The encryption transmission unit is used to achieve secure data transmission using a hybrid communication network architecture, including 5G wireless communication, optical fiber backbone network and low-power wide area network, and uses the TLS1.3 protocol to perform end-to-end encryption on the transmitted data; The data governance unit consists of a data cleaning subunit, a format conversion subunit, and a spatiotemporal alignment subunit; The data cleaning subunit identifies and removes abnormal data points based on a density clustering algorithm. The core parameters of the algorithm include the neighborhood radius and the minimum number of samples MinPts; The format conversion subunit is used to uniformly convert the original data of heterogeneous data sources into a standardized data model, and the data model uses a hierarchical JSON architecture to represent spatiotemporal data; The spatiotemporal alignment subunit is used to map multi-source data to a preset spatiotemporal grid and use a spatiotemporal interpolation algorithm to fill in missing data.

[0011] Preferably, the traffic flow prediction and analysis module includes: Historical modeling unit, which constructs traffic flow trends based on a time series model with seasonal decomposition: The real-time prediction unit uses a fusion model of spatiotemporal graph neural network and long short-term memory network to integrate historical trends and real-time data to predict future traffic conditions; The impact analysis unit analyzes traffic influencing factors using an association rule mining algorithm, which evaluates the effectiveness of the rules by calculating support, confidence, and lift.

[0012] Preferably, the intelligent traffic signal optimization module includes: A single-point control unit dynamically adjusts traffic light timing based on a deep Q-network reinforcement learning algorithm. The algorithm's Markov decision process model includes: The state space S includes the traffic flow in each direction of the intersection, queue length, average speed, and the status of adjacent intersections; Action space A, including signal light phase switching time and green light extension / shortening strategy; The reward function is defined as a quantitative indicator of traffic efficiency improvement, including a comprehensive evaluation of the average vehicle delay time, number of stops, and queue length; The state transition probability is obtained through statistical learning of historical data.

[0013] The regional collaborative unit uses a model predictive control algorithm to achieve multi-intersection linkage optimization. The algorithm determines the control sequence by solving the following optimization problem The emergency response unit executes a preset signal control strategy for emergencies. The strategy is activated by an event trigger mechanism and manages multi-event conflicts through a priority queue.

[0014] Preferably, the parking lot intelligent scheduling and guidance module includes: Demand forecasting unit, which predicts parking space demand based on a gradient boosting tree algorithm. The algorithm integrates historical parking data, traffic flow data, and passenger flow data of surrounding points of interest, and selects key predictors through feature importance analysis; Intelligent matching unit, which uses multi-objective optimization algorithm to generate parking lot recommendation solutions; The dynamic guidance unit releases real-time parking guidance information through multiple channels such as variable information boards, navigation applications and in-vehicle terminals. The information includes parking lot location, remaining parking spaces, estimated arrival time and recommended routes.

[0015] Preferably, the present invention further includes a traffic management decision support module, including: A visualization unit uses geographic information system technology to build a traffic operation status visualization platform that supports multi-dimensional data query, spatiotemporal data analysis, and abnormal event warnings; The decision generation unit evaluates the effectiveness of traffic management strategies based on a Monte Carlo simulation model. The model simulates the response of the traffic system to different strategies through random sampling and generates a strategy evaluation report.

[0016] Preferably, the present invention further includes a user service module, including: The travel service unit provides real-time traffic query, dynamic route planning and parking reservation services. The route planning algorithm comprehensively considers real-time traffic conditions, historical congestion patterns and user preferences; The green travel unit recommends public transportation travel plans and establishes a carbon emission reduction incentive mechanism. The mechanism uses blockchain technology to record users' green travel behaviors and redeem points for rewards.

[0017] Preferably, the data preprocessing module realizes real-time data processing through a distributed computing framework. The framework adopts a master-slave architecture design, including data acquisition nodes, processing nodes and storage nodes, and each node realizes data flow through a message queue.

[0018] The big data-based smart city construction system according to claim 1 is characterized in that the parking lot intelligent scheduling and guidance module is integrated with a third-party navigation platform through an open API interface, and the interface complies with the OAuth2.0 authentication protocol and supports cross-platform data exchange and service collaboration.

[0019] Compared with related technologies, the smart city construction system based on big data provided by the present invention has the following beneficial effects: 1. This invention integrates spatiotemporal graph convolutional networks, deep reinforcement learning intelligent algorithms, blockchain, and distributed computing technologies to build a multi-source data-driven full-process intelligent management system for transportation, achieving full-chain intelligence from real-time perception and dynamic optimization to cross-platform services. It breaks through the data silos and static control limitations of traditional transportation management, and significantly improves the coordination, prediction accuracy, and service ecosystem openness of urban transportation systems.

[0020] 2. The present invention realizes the full-process processing of traffic data through multi-module collaboration, accurately predicts traffic status and intelligently optimizes signal and parking lot scheduling, significantly improving urban traffic operation efficiency, alleviating congestion and improving traffic smoothness. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a principle block diagram of each module in the big data-based smart city construction system provided by the present invention. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0023] The terms used in this disclosure are for the purpose of describing specific embodiments only and are not intended to limit the disclosure. As used in this disclosure and the appended claims, the singular forms "group," "class," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0024] It should be understood that although the terms first, second, third, etc. may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining."

[0025] Please refer to Figure 1 A smart city construction system based on big data, including: Traffic data collection module, used to obtain real-time road traffic, parking lot status, mobile terminal travel trajectory and public transportation operation data; The data transmission and preprocessing module is used to encrypt and transmit heterogeneous data, clean it, unify its format, and align it in time and space; Traffic flow prediction and analysis module, which uses a spatiotemporal graph convolutional network combined with a seasonal autoregressive model to predict traffic conditions for a set time period in the future; Intelligent traffic signal optimization module, which dynamically adjusts traffic light timing based on deep reinforcement learning algorithms and realizes intersection linkage through regional collaborative control algorithms; The parking lot intelligent scheduling and guidance module uses the gradient boosting tree to predict parking space demand, combines it with the improved collaborative filtering algorithm to recommend the best parking lot for car owners, and dynamically guides them through multiple channels.

[0026] It should be noted that the system realizes the full-process processing of traffic data through the collaboration of various modules, accurately predicts traffic conditions, intelligently optimizes signal and parking lot scheduling, and improves urban traffic operation efficiency and travel experience.

[0027] In this application, the traffic data collection module includes: Real-time collection of traffic flow parameters and environmental data through a distributed sensor network, including geomagnetic sensor arrays, microwave radar detectors, video surveillance equipment, and meteorological monitoring devices. Traffic flow parameters include vehicle volume, speed, and lane occupancy; Communicate with the parking management system through a standardized data interface to obtain parking space occupancy status, geographic location and charging standards in real time. The data interface supports RESTful API or message queue protocol; Establish a data sharing mechanism with third-party map applications to obtain anonymized user travel trajectories, real-time traffic reports, and navigation request data, and use data desensitization technology to protect user privacy; Access the public transportation system to obtain real-time vehicle location, operating status and passenger flow data collected through on-board sensors or ticketing systems.

[0028] It should be noted that the traffic data collection module uses distributed sensor networks, standardized interfaces, data sharing mechanisms and public transportation access to collect data on traffic flow, parking lots, travel trajectories and public transportation in an all-round and multi-channel manner, providing a rich and accurate data foundation for subsequent traffic analysis and management.

[0029] In this application, the data transmission and preprocessing module includes an encryption transmission unit and a data governance unit; The encryption transmission unit is used to achieve secure data transmission using a hybrid communication network architecture, including 5G wireless communication, optical fiber backbone network and low-power wide area network, and uses the TLS1.3 protocol to perform end-to-end encryption on the transmitted data; The data governance unit consists of a data cleaning subunit, a format conversion subunit, and a spatiotemporal alignment subunit; The data cleaning subunit identifies and removes abnormal data points based on the density clustering algorithm. The core parameters of the algorithm include the neighborhood radius. and the minimum number of samples MinPts; The format conversion subunit is used to uniformly convert the original data of heterogeneous data sources into a standardized data model, and the data model uses a hierarchical JSON architecture to represent spatiotemporal data; The spatiotemporal alignment subunit is used to map multi-source data to a preset spatiotemporal grid and use a spatiotemporal interpolation algorithm to fill in missing data. The calculation formula of the interpolation algorithm is: ; Where, is the value of the neighbor data point, is the spatial distance, is the time interval, is the weight adjustment parameter, i represents the index variable, which is used to identify each neighbor data point involved in the calculation. The value range is from 1 to m, where m is the number of neighbor data points.

[0030] It should be noted that the data transmission and preprocessing module uses an encrypted transmission unit to ensure data transmission security, and uses the sub-units of the data governance unit to clean, convert the format, and align the data in time and space to improve data quality and standardization, laying a reliable data foundation for subsequent traffic flow analysis, prediction and other work.

[0031] In this application, the traffic flow prediction and analysis module includes: Historical modeling unit, which constructs traffic flow trends based on a time series model with seasonal decomposition: Obtain historical traffic flow time series data , where T is the length of the time series; The time series is decomposed using the additive model or multiplicative model. The mathematical expression of the additive model is: ,in is the observation sequence at time t (actual traffic flow value), is the trend component at time t (reflecting the long-term growth or decline trend), is the residual component at time t (random noise or outliers); The mathematical expression of its multiplication model is: , which is used for scenarios where seasonal fluctuations vary with trends; The trend component is modeled using an autoregressive integrated moving average model, whose parameters are automatically optimized using the information criterion, specifically: Separate trend components The autoregressive integrated moving average model (ARIMA (p, d, q)) is used for modeling. The model expression is: , where B is the lag operator, , d is the difference order, used to make the series stable, is the autoregressive coefficient, is the sliding average coefficient, is a white noise sequence, where i is the index variable of the autoregressive part, ranging from 1 to p, used to traverse the autoregressive coefficient, and j is the index variable of the sliding average part, ranging from 1 to q, used to traverse the sliding average coefficient; Automatically select the optimal (p, d, q) through the information criterion to balance model complexity and goodness of fit; The real-time prediction unit uses a fusion model of spatiotemporal graph neural network and long short-term memory network to integrate historical trends and real-time data to predict future traffic conditions. The specific steps are as follows: The historical trend component Integrate with real-time collected traffic data into a spatiotemporal feature matrix; traffic data includes but is not limited to current traffic volume, vehicle speed, and spatial adjacent intersection status; Adopting the fusion model of spatiotemporal graph neural network and long short-term memory network: In the spatiotemporal graph modeling step, the road network is abstracted into a graph structure G=(V,E), where the node V represents an intersection or road section, and the edge E represents a spatial association. Spatial associations are based on the graph, such as the distance between adjacent intersections and travel time. Spatial dependency features are extracted through a graph convolutional network. ; Time series modeling step, using long short-term memory network to capture the dynamic dependency characteristics of the time dimension ; Feature fusion, and Splicing or weighted summation to output prediction results ; Optimize model parameters by minimizing mean square error and regularization term , train the loss function: ; Where N is the number of training samples, is the true value of the i-th sample, is the predicted value of the i-th sample, is a regularization term used to prevent overfitting, is the regularization strength parameter, which is used to balance the fitting error and model complexity; The impact analysis unit uses an association rule mining algorithm to analyze traffic influencing factors. The algorithm evaluates the effectiveness of the rules by calculating support, confidence, and lift. The specific steps are as follows: Data preprocessing: Traffic flow data and influencing factors are encoded into a binary dataset D, where the value is 1 if the influencing factor exists and 0 otherwise. Influencing factors include weather type, date type, construction events, etc. Association rule mining uses Apriori algorithm or FP-Growth algorithm to extract frequent item sets and generate Rules, such as "Rainy Day" "congestion"; Filter effective rules by the following indicators: Support , represents the frequency of occurrence of the rule in the data set; Confidence , which means the probability that a sample containing A also contains B; Lift, , indicating that the actual impact of the rule is a multiple of the random level, and the rule is effective when Lift>1.

[0032] It should be noted that the traffic flow prediction and analysis module uses the historical modeling unit to mine historical trends in traffic flow, the real-time prediction unit to integrate multi-source data to predict future states, and the impact analysis unit to mine traffic influencing factors, providing trend prediction, state prediction, and factor correlation analysis and other support for traffic management and decision-making.

[0033] In this application, the intelligent traffic signal optimization module includes a single-point control unit and a regional coordination unit; The single-point control unit dynamically adjusts traffic light timing based on a deep Q-network reinforcement learning algorithm. The Markov decision process model of the algorithm includes: State space S, including traffic flow in all directions of the intersection , queue length , average speed and adjacent intersection status , the state vector of its adjacent intersection state is represented by ; Where i represents the number of lanes in different directions, and j represents the number of adjacent intersections; Action space A, including the signal light phase switching time , Green light extension / shortening strategy , whose action vector is expressed as ; Reward Function , defined as a quantitative indicator of traffic efficiency improvement, including a comprehensive assessment of the average vehicle delay time, number of stops, and queue length. The formula is: ;in is the average delay time of all vehicles in the current cycle, is the total number of stops in the current cycle, is the average queue length of all lanes, 、 、 is the weight coefficient, represents a small constant that prevents the denominator from being zero; State transition probability , obtained through statistical learning of historical data, and approximated by Gaussian mixture model: ; where K is the number of mixed components, The weight of the kth component, and is the mean vector and covariance matrix based on the state vector s and the action vector a; Then use deep Q network training, using neural network Estimate the state action value, and its loss function is , where r represents the immediate reward, represents the discount factor, represents the target network parameters; The regional collaborative unit uses a model predictive control algorithm to achieve multi-intersection linkage optimization, and the specific steps are as follows: Abstract the traffic network as a discrete-time system: ,in is the system state vector (including traffic flow at each intersection, queue length, etc.), To control the input vector (phase and timing of traffic lights at each intersection), represents external interference (such as randomly arriving vehicles), and f represents the system dynamic function (learned from historical data); Then within the forecast time domain N, solve the optimal control sequence , minimizing the following objective function: ;in is the system output (such as average delay time, queue length, etc.), is the reference trajectory (ideal state), is the control input, Q represents the state error weight matrix (the diagonal elements represent the importance of each output variable), and R represents the control input change weight matrix (used to prevent frequent switching of traffic lights); Then set the constraints, including physical constraints, logical constraints and safety constraints: Physical constraints: the timing of each phase must be tp ≥ min time and tp ≤ max time. tp represents the timing of each phase of the traffic light in the intelligent traffic signal optimization module, min time represents the minimum time allowed for the phase timing of the traffic light, and max time represents the maximum time allowed for the phase timing of the traffic light. Logical constraint: only one phase can be green at a time; Safety constraints: Phase switching between adjacent intersections must meet the minimum interval time; Use sequential quadratic programming or interior point method to solve the above optimization problem and obtain the optimal control at the current moment , and execute; Emergency Response Unit: Executes preset signal control strategies for emergencies. These strategies are activated through event triggering mechanisms and manage multi-event conflicts through priority queues. Emergencies include traffic accidents, road construction, and other events. The preset signal control strategies for emergencies are as follows: Detect emergencies in real time through traffic sensors; classify the events and assess their severity level (L1), impact level (L2), and urgency level (L3); Preset corresponding signal control strategies for different event types and severity levels. For example, in the case of a traffic accident, priority is given to rescue vehicles and the green light time is extended. In the case of road construction, the right of way of the lane in the construction direction is reduced and the green light time in the detour direction is increased. When multiple events occur at the same time, conflicts are managed through the priority queue Q, with the following rules: Obtain information on the impact of the incident, including severity level L1, impact scope level L2, and urgency level L3; Calculate the priority of an event using the following formula: , where z1, z2, and z3 represent the weights corresponding to the severity level, impact level, and urgency level, respectively; Strategies for high-priority events are executed first; When a low-priority event conflicts with a high-priority event, the low-priority strategy is partially or completely replaced; A recovery mechanism is also provided for returning to normal control from the emergency strategy through a smooth transition function after the incident is resolved.

[0034] It should be noted that the intelligent traffic signal optimization module uses deep Q network reinforcement learning through a single-point control unit to dynamically adjust the timing of traffic lights at a single intersection. The regional coordination unit uses model predictive control to achieve multi-intersection linkage optimization. Combined with the emergency response unit, it flexibly adjusts strategies for emergencies, comprehensively improving the intelligence and efficiency of traffic signal control and ensuring the smooth operation of urban traffic.

[0035] In this application, the parking lot intelligent scheduling and guidance module includes: Demand forecasting unit, which predicts parking space demand based on a gradient boosting tree algorithm. The algorithm integrates historical parking data, traffic flow data, and passenger flow data of surrounding points of interest, and selects key predictors through feature importance analysis; The intelligent matching unit uses a multi-objective optimization algorithm to generate parking lot recommendation solutions. The optimization objective function of the algorithm is: ;in Including distance target, cost target, remaining parking space target and user preference target, is the corresponding weight coefficient, which is determined by the hierarchical analysis method. It should be noted that i is an index variable used to traverse different objective functions. , where i ranges from 1 to n, n represents the total number of objective functions and corresponds to their weight coefficients ;

[0036] The dynamic guidance unit releases real-time parking guidance information through multiple channels such as variable information boards, navigation applications and in-vehicle terminals. The information includes parking lot location, remaining parking spaces, estimated arrival time and recommended routes.

[0037] It should be noted that the parking lot intelligent scheduling and induction module accurately predicts parking space demand through the demand forecasting unit, the intelligent matching unit generates the optimal recommendation based on multiple objectives, and the dynamic induction unit publishes real-time information through multiple channels to achieve efficient utilization of parking resources and optimize the parking experience of car owners.

[0038] In this application, the present invention also includes a traffic management decision support module, including: A visualization unit uses geographic information system technology to build a traffic operation status visualization platform that supports multi-dimensional data query, spatiotemporal data analysis, and abnormal event warnings; The decision generation unit evaluates the effectiveness of traffic management strategies based on a Monte Carlo simulation model. The model simulates the response of the traffic system to different strategies through random sampling and generates a strategy evaluation report.

[0039] It should be noted that the traffic management decision support module uses a visualization unit to intuitively present the traffic operation status and warn of anomalies, and uses a decision generation unit to evaluate the effectiveness of strategies based on simulation, providing data support and analysis tools for traffic management departments to make scientific and effective decisions.

[0040] In this application, the present invention also includes a user service module, including: The travel service unit provides real-time traffic query, dynamic route planning and parking reservation services. The route planning algorithm comprehensively considers real-time traffic conditions, historical congestion patterns and user preferences; The green travel unit recommends public transportation travel plans and establishes a carbon emission reduction incentive mechanism. The mechanism uses blockchain technology to record users' green travel behaviors and redeem points for rewards.

[0041] It should be noted that the user service module provides users with convenient travel-related services through the travel service unit, and the green travel unit encourages users to choose public transportation, which not only improves the travel experience, but also helps promote green travel, meet user needs and promote sustainable urban development.

[0042] In this application, the data preprocessing module realizes real-time data processing through a distributed computing framework. The framework adopts a master-slave architecture design, including data acquisition nodes, processing nodes and storage nodes. Each node realizes data flow through a message queue.

[0043] It should be noted that the data preprocessing module uses the master-slave architecture and message queue of the distributed computing framework to realize the real-time flow and processing of data between nodes, efficiently integrate the collection, processing and storage functions, and provide timely and orderly data support for subsequent data analysis and application.

[0044] In this application, the parking lot intelligent scheduling and guidance module is integrated with a third-party navigation platform through an open API interface. The interface complies with the OAuth2.0 authentication protocol and supports cross-platform data exchange and service collaboration.

[0045] It should be noted that the parking lot intelligent scheduling and guidance module is integrated with the third-party navigation platform through an open API interface that complies with the OAuth2.0 authentication protocol, realizing cross-platform data exchange and service collaboration, expanding service channels, improving the efficiency of parking lot information dissemination and utilization, and optimizing the parking navigation experience of car owners.

[0046] In this application, it should be noted that the following expressions can be used for the reuse of symbolic variables: The mathematical symbols and variables involved in the present invention may have different meanings in different technical scenarios, but are clearly defined when they first appear.

[0047] For example, the index variable i represents the number of lanes in different directions in the intelligent traffic signal optimization module, and represents the index of neighboring data points in the spatiotemporal interpolation algorithm. The variable n represents the total number of objective functions in the multi-objective optimization algorithm and represents the number of samples in the training sample set (as defined by the loss function).

[0048] Among them, the specific meanings of all symbols are clearly defined through context or formula annotations, and technical personnel in the relevant technical field can understand their exact references based on the corresponding technical scenarios without causing confusion.

[0049] Other embodiments of the present invention will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the invention being indicated by the following claims.

[0050] It should be understood that the present invention is not limited to the exact construction described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A smart city construction system based on big data, characterized by: include: Traffic data collection module, used to obtain real-time road traffic, parking lot status, mobile terminal travel trajectory and public transportation operation data; The data transmission and preprocessing module is used to encrypt and transmit heterogeneous data, clean it, unify its format, and align it in time and space; Traffic flow prediction and analysis module, which uses a spatiotemporal graph convolutional network combined with a seasonal autoregressive model to predict traffic conditions for a set time period in the future; Intelligent traffic signal optimization module, which dynamically adjusts traffic light timing based on deep reinforcement learning algorithms and realizes intersection linkage through regional collaborative control algorithms; The parking lot intelligent scheduling and guidance module uses the gradient boosting tree to predict parking space demand, combines it with the improved collaborative filtering algorithm to recommend the best parking lot for car owners, and dynamically guides them through multiple channels.

2. A smart city construction system based on big data according to claim 1, characterized in that: The traffic data collection module includes: Real-time collection of traffic flow parameters and environmental data through a distributed sensor network, including geomagnetic sensor arrays, microwave radar detectors, video surveillance equipment, and meteorological monitoring devices. Traffic flow parameters include vehicle volume, speed, and lane occupancy; Communicate with the parking management system through a standardized data interface to obtain parking space occupancy status, geographic location and charging standards in real time. The data interface supports RESTful API or message queue protocol; Establish a data sharing mechanism with third-party map applications to obtain anonymized user travel trajectories, real-time traffic reports, and navigation request data, and use data desensitization technology to protect user privacy; Access the public transportation system to obtain real-time vehicle location, operating status and passenger flow data collected through on-board sensors or ticketing systems.

3. The smart city construction system based on big data according to claim 1 is characterized in that: The data transmission and preprocessing module includes an encryption transmission unit and a data management unit; The encryption transmission unit is used to achieve secure data transmission using a hybrid communication network architecture, including 5G wireless communication, optical fiber backbone network and low-power wide area network, and uses the TLS1.3 protocol to perform end-to-end encryption on the transmitted data; The data governance unit consists of a data cleaning subunit, a format conversion subunit, and a spatiotemporal alignment subunit; The data cleaning subunit identifies and removes abnormal data points based on a density clustering algorithm. The core parameters of the algorithm include the neighborhood radius and the minimum number of samples MinPts; The format conversion subunit is used to uniformly convert the original data of heterogeneous data sources into a standardized data model, and the data model uses a hierarchical JSON architecture to represent spatiotemporal data; The spatiotemporal alignment subunit is used to map multi-source data to a preset spatiotemporal grid and use a spatiotemporal interpolation algorithm to fill in missing data.

4. The smart city construction system based on big data according to claim 1, characterized in that: The traffic flow prediction and analysis module includes: Historical modeling unit, which constructs traffic flow trends based on a time series model with seasonal decomposition: The real-time prediction unit uses a fusion model of spatiotemporal graph neural network and long short-term memory network to integrate historical trends and real-time data to predict future traffic conditions; The impact analysis unit analyzes traffic influencing factors using an association rule mining algorithm, which evaluates the effectiveness of the rules by calculating support, confidence, and lift.

5. The smart city construction system based on big data according to claim 1 is characterized in that: The intelligent traffic signal optimization module includes: A single-point control unit dynamically adjusts traffic light timing based on a deep Q-network reinforcement learning algorithm. The algorithm's Markov decision process model includes: The state space S includes the traffic flow in each direction of the intersection, queue length, average speed, and the status of adjacent intersections; Action space A, including signal light phase switching time and green light extension / shortening strategy; The reward function is defined as a quantitative indicator of traffic efficiency improvement, including a comprehensive evaluation of the average vehicle delay time, number of stops, and queue length; The state transition probability is obtained through statistical learning of historical data; The regional collaborative unit uses a model predictive control algorithm to achieve multi-intersection linkage optimization. The algorithm determines the control sequence by solving the following optimization problem The emergency response unit executes a preset signal control strategy for emergencies. The strategy is activated by an event trigger mechanism and manages multi-event conflicts through a priority queue.

6. The smart city construction system based on big data according to claim 1, characterized in that: The parking lot intelligent scheduling and guidance module includes: Demand forecasting unit, which predicts parking space demand based on a gradient boosting tree algorithm. The algorithm integrates historical parking data, traffic flow data, and passenger flow data of surrounding points of interest, and selects key predictors through feature importance analysis; Intelligent matching unit, which uses multi-objective optimization algorithm to generate parking lot recommendation solutions; The dynamic guidance unit releases real-time parking guidance information through multiple channels such as variable information boards, navigation applications and in-vehicle terminals. The information includes parking lot location, remaining parking spaces, estimated arrival time and recommended routes.

7. The smart city construction system based on big data according to claim 1 is characterized in that: Also included are traffic management decision support modules, including: A visualization unit uses geographic information system technology to build a traffic operation status visualization platform that supports multi-dimensional data query, spatiotemporal data analysis, and abnormal event warnings; The decision generation unit evaluates the effectiveness of traffic management strategies based on a Monte Carlo simulation model. The model simulates the response of the traffic system to different strategies through random sampling and generates a strategy evaluation report.

8. The smart city construction system based on big data according to claim 1 is characterized in that: Also includes user service modules, including: The travel service unit provides real-time traffic query, dynamic route planning and parking reservation services. The route planning algorithm comprehensively considers real-time traffic conditions, historical congestion patterns and user preferences; The green travel unit recommends public transportation travel plans and establishes a carbon emission reduction incentive mechanism. The mechanism uses blockchain technology to record users' green travel behaviors and redeem points for rewards.

9. The smart city construction system based on big data according to claim 1, characterized in that: The data preprocessing module realizes real-time data processing through a distributed computing framework. The framework adopts a master-slave architecture design, including data acquisition nodes, processing nodes and storage nodes. Each node realizes data flow through a message queue.

10. The smart city construction system based on big data according to claim 1, characterized in that: The parking lot intelligent scheduling and guidance module is integrated with a third-party navigation platform through an open API interface. The interface complies with the OAuth2.0 authentication protocol and supports cross-platform data exchange and service collaboration.

Images