Big data analysis method and system for intelligent parking

Abstract

The present invention relates to the field of intelligent parking scheduling, and in particular to a big data analysis method and system for intelligent parking. The method comprises the following steps: obtaining all parking lot information in the city based on a parking management platform and marking parking lot nodes; visualizing the three-dimensional distribution of parking lot nodes based on all parking lot information in the city, and constructing a three-dimensional distribution visualization network of parking lots; obtaining urban road traffic flow data based on multi-source sensors; performing real-time vehicle dynamic movement analysis on urban road traffic flow data, and performing multi-time point distribution evolution to construct a digital urban traffic trajectory portrait; obtaining historical urban parking operation logs; performing parking demand trend evolution on historical urban parking operation logs, and then performing dynamic parking demand prediction for future time periods, thereby obtaining dynamic parking demand prediction data. The present invention realizes efficient and high-utilization urban parking scheduling.

Description

Technical Field

[0001] The present invention relates to the field of intelligent parking scheduling, and in particular to a big data analysis method and system for intelligent parking. Background Art

[0002] With the acceleration of urbanization, the urban population and the number of vehicles are increasing, and the problem of urban parking has gradually become an important issue affecting urban traffic management and the quality of life of residents. The traditional parking management method relies on manual guidance and fixed parking resource allocation, which cannot adapt to the growing parking demand and the complex and changing urban traffic environment. This traditional method not only has problems such as waste of resources, unreasonable allocation of parking spaces, and difficulty for car owners to find parking spaces, but also cannot effectively respond to sudden changes in demand during peak hours and dynamic adjustment of parking resources. With the development of information technology, especially the widespread application of big data, the Internet of Things (IoT) and artificial intelligence (AI) technologies, intelligent parking scheduling methods based on data analysis have gradually become an innovative path to solve this problem.

[0003] Big data analysis provides scientific decision-making support for urban parking management by collecting and analyzing multi-dimensional data such as urban parking data, traffic flow, weather conditions, and regional activities. Using sensors and smart parking equipment, it is possible to monitor the parking space occupancy and owner behavior in real time and collect a large amount of parking data. These data provide sufficient basis for the development of parking demand prediction, parking resource optimization and scheduling, and intelligent recommendation systems. Through in-depth analysis of these data, the intelligent parking system can not only dynamically evaluate the real-time demand of each parking area, but also predict future parking demand trends, thereby effectively solving the problems of unbalanced parking resources and difficulty for car owners to find parking spaces.

[0004] However, although big data analysis provides new solutions for intelligent parking scheduling, how to efficiently process and analyze massive data, how to respond to dynamic changes in real time, and how to balance the contradiction between parking demand and resource supply are still important challenges in current technological development. In response to these problems, researching and developing more advanced intelligent parking scheduling analysis methods, especially innovative solutions based on big data analysis, has become an important direction for improving urban parking management and responding to future urban development needs. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a big data analysis method and system for intelligent parking to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a big data analysis method for smart parking, comprising the following steps:

[0007] Step S1: Based on the parking management platform, all parking lot information in the city is obtained and parking lot nodes are marked; based on the information of all parking lots in the city, the parking lot nodes are visualized in three dimensions to construct a three-dimensional distribution visualization network of parking lots;

[0008] Step S2: Acquire urban road traffic flow data based on multi-source sensors; perform real-time vehicle dynamic movement analysis on urban road traffic flow data, and perform multi-time point distribution evolution to construct a digital urban traffic trajectory portrait;

[0009] Step S3: Obtain historical city parking operation logs; perform parking demand trend evolution on the historical city parking operation logs, and then perform dynamic parking demand forecasting for future periods, thereby obtaining dynamic parking demand forecasting data;

[0010] Step S4: Calculate the optimal parking time and intelligently recommend parking requirements for the three-dimensional distribution visualization network of the parking lot based on the dynamic parking demand prediction data, thereby obtaining intelligently recommended parking lot location data;

[0011] Step S5: Perform dynamic parking path planning on the location data of the intelligently recommended parking lot based on the city's digital traffic trajectory portrait, and then make dynamic parking scheduling decisions to build a dynamic parking scheduling strategy;

[0012] Step S6: Perform multi-time point parking scheduling simulation based on the dynamic parking scheduling strategy, and perform deep parking migration optimization to build an intelligent parking collaborative scheduling model.

[0013] The present invention visualizes the three-dimensional distribution of all parking lot information in the city, intuitively displays the spatial distribution of parking lots, helps users quickly understand the distribution of parking resources, and constructs a three-dimensional distribution visualization network of parking lots, which helps optimize the utilization of parking resources, improves parking lot management efficiency, reduces parking congestion and the time spent looking for parking spaces. Through real-time vehicle dynamic movement analysis and multi-time point distribution evolution, a digital urban traffic trajectory portrait is generated, which helps to gain a deep understanding of the distribution of urban traffic flow and traffic congestion conditions. The construction of a digital urban traffic trajectory portrait provides real-time traffic information support for intelligent parking scheduling, helps optimize parking path planning and reduce traffic congestion. Through the analysis of historical urban parking operation logs and the evolution of parking demand trends, the dynamic parking demand in future time periods is accurately predicted, providing a basis for the rational allocation of urban parking resources. The generation of dynamic parking demand prediction data helps to achieve intelligent parking lot location recommendation and improve parking lot Resource utilization, reduce parking search time, calculate the optimal parking time and intelligent parking demand recommendation based on dynamic parking demand prediction data, provide users with personalized parking suggestions, reduce parking time and cost, and generate intelligent recommended parking lot location data to help optimize user parking experience, improve parking efficiency, and reduce urban parking congestion and emissions. Dynamic parking path planning and parking scheduling decisions are made based on urban digital traffic trajectory portraits to achieve intelligent parking path planning and reduce traffic congestion. Building a dynamic parking scheduling strategy helps improve urban traffic operation efficiency, optimize parking resource allocation, and improve urban traffic management. Through multi-time point parking scheduling simulation and deep parking migration optimization, an intelligent parking collaborative scheduling model is built to achieve the optimal configuration and management of parking resources. The intelligent parking collaborative scheduling model helps to improve the utilization efficiency of urban parking resources, reduce traffic congestion and emissions, and improve the intelligence level of urban traffic operation.

[0014] Preferably, step S1 comprises the following steps:

[0015] Step S11: Obtain all parking lot information in the city based on the parking management platform and mark the parking lot nodes;

[0016] Step S12: performing geospatial positioning calculation on parking lot nodes to extract the geographical location coordinates of each parking lot;

[0017] Step S13: Performing urban spatial distribution analysis on the geographical location coordinates of each parking lot to generate urban spatial distribution data of parking lot nodes;

[0018] Step S14: performing node distribution discrete fitting on the parking lot node urban spatial distribution data to construct a parking lot urban distribution network;

[0019] Step S15: Based on the information of all parking lots in the city, the three-dimensional distribution visualization of the parking lot city distribution network is performed to construct a three-dimensional parking lot distribution visualization network.

[0020] The present invention acquires information of all parking lots in the city and marks parking lot nodes, thereby establishing a comprehensive database of parking lots, which is convenient for managers and users to search for and understand parking resources. The establishment of parking lot node marking helps to improve the visual display of parking lot information, promotes the rational use of parking resources and the intelligentization of parking management, performs geospatial positioning calculations on parking lot nodes, accurately extracts the geographical location coordinates of each parking lot, and provides accurate data support for subsequent urban spatial distribution analysis. The geospatial positioning calculations help to establish a precise geographic information database of parking lots, provides accurate location data for intelligent parking scheduling, and improves the effective utilization rate of parking resources. By performing urban spatial distribution analysis on the geographical location coordinates of each parking lot, an in-depth understanding of the distribution of parking lots in urban space is obtained, which helps to optimize the layout of parking resources and generate parking lots. Node city spatial distribution data helps urban planners optimize the layout of urban parking lots, improve parking efficiency, reduce congestion and environmental pollution, and perform node distribution discrete fitting on the parking lot node city spatial distribution data to better understand the distribution pattern of parking lots and provide a more accurate data basis for building a parking lot city distribution network. Node distribution discrete fitting helps optimize the modeling process of the parking lot network, improve the understanding of the distribution of urban parking resources, and provide a more scientific basis for intelligent parking scheduling. Based on the information of all parking lots in the city, the three-dimensional distribution visualization of the parking lot city distribution network is performed to intuitively display the distribution of parking lots in the urban space and provide users with more intuitive parking information. Building a three-dimensional parking lot distribution visualization network helps optimize parking lot management and user experience, improve the utilization efficiency of urban parking resources, and reduce parking congestion and emissions.

[0021] Preferably, step S15 specifically comprises the following steps:

[0022] Calculate the vehicle entry and exit times of each parking lot in the city, and extract the vehicle entry and exit timestamps of each parking lot;

[0023] Calculate the available parking spaces in each parking lot based on the vehicle entry and exit timestamps to generate the number of available parking spaces in each parking lot;

[0024] Perform real-time status analysis of vacant parking spaces based on the number of vacant parking spaces in each parking lot, and generate real-time status data of vacant parking spaces;

[0025] Based on the real-time status data of vacant parking spaces, the city distribution network of parking lots is mapped in real time to obtain a vacant parking space status mapping distribution network;

[0026] Carry out topological correlation analysis between different parking lots in the urban distribution network of parking lots and extract the topological correlation features between parking lots;

[0027] Based on the topological association characteristics between parking lots, the three-dimensional distribution visualization of the vacant parking space status mapping distribution network is performed to construct a three-dimensional distribution visualization network of parking lots.

[0028] The present invention calculates the entry and exit times of vehicles in each parking lot one by one, and can establish vehicle flow data for each parking lot, thereby helping to evaluate the usage and busyness of the parking lot. Extracting vehicle entry and exit timestamps is helpful to realize time series analysis of parking lot data, and provides a basis for calculating and predicting the real-time parking space status. By calculating the real-time number of vacant parking spaces through vehicle entry and exit timestamps, the present invention helps managers and drivers understand the available parking spaces in the parking lot in real time, thereby improving the utilization rate of parking lot resources. Generating data on the number of vacant parking spaces helps to optimize the management and scheduling of parking lots, provide more convenient parking information query services, and alleviate parking congestion problems. Analyzing the real-time status data of vacant parking spaces helps managers monitor the usage of parking lots, adjust the allocation of parking resources in a timely manner, and improve the operating efficiency of parking lots. Real-time status analysis helps to discover parking space shortages in advance, take corresponding measures, optimize the utilization rate of parking spaces in parking lots, and improve user experience. The distributed network performs real-time status mapping to realize the dynamic display of parking space status and provide users with real-time parking information. The mapped distributed network can intuitively display the parking space situation of various parking lots in the city, help drivers quickly choose suitable parking lots, reduce the time of finding parking spaces and vehicle congestion, and analyze the topological association characteristics between different parking lots to help understand the correlation between parking lots and traffic flow, and provide more accurate data support for intelligent parking scheduling. The topological association characteristics are extracted to optimize the layout of the parking lot network, improve the allocation of urban parking resources, and improve the overall efficiency and service level of parking lots. Based on the topological association characteristics between parking lots, the vacant parking space status mapping distribution network is three-dimensionally visualized to intuitively display the vacant parking space situation of the parking lot network, providing users with more intuitive parking information. The construction of a three-dimensional distributed visualization network improves users' understanding of the vacant parking space situation in parking lots, provides drivers with a smarter and more efficient parking lot selection plan, and reduces urban traffic congestion.

[0029] Preferably, the specific steps of step S2 are:

[0030] Step S21: Acquire urban road traffic flow data based on multi-source sensors;

[0031] Step S22: removing abnormal values ​​from the urban road traffic flow data to obtain abnormal filtered traffic flow data;

[0032] Step S23: Perform real-time vehicle dynamic movement analysis on the abnormal filtered traffic flow data to generate real-time dynamic trajectory of vehicles;

[0033] Step S24: performing regional trajectory distribution analysis on the real-time dynamic trajectory of the vehicle to obtain regional trajectory distribution characteristics of the vehicle;

[0034] Step S25: performing multi-time point distribution evolution on the regional trajectory distribution characteristics of the vehicle to generate regional multi-time point trajectory distribution data;

[0035] Step S26: Conduct in-depth distribution law mining on the regional multi-time point trajectory distribution data to construct a digital traffic trajectory portrait of the city.

[0036] The present invention realizes all-round traffic monitoring by acquiring urban road traffic flow data by combining multi-source sensor data, and provides more comprehensive data support. Multi-source sensor data helps to capture the real-time status of urban road traffic, and provides accurate traffic flow information for the intelligent parking dispatching system, which helps to optimize parking lot selection and vehicle driving path planning. After removing outliers, the accuracy and reliability of data are improved, and the effectiveness of subsequent analysis is ensured. The abnormal filtered traffic flow data reduces the interference of error factors, ensures the reliability of subsequent analysis, and improves the quality of traffic data. Real-time vehicle dynamic movement analysis realizes real-time monitoring and analysis of urban road traffic, helps managers understand real-time traffic conditions, and generates real-time dynamic vehicle trajectories, which helps to realize real-time monitoring and optimization of vehicle driving paths, improves traffic operation efficiency, and provides more accurate data support for intelligent parking dynamic dispatching. Regional trajectory Distribution analysis helps understand the distribution of vehicles in different areas of the city, and provides a basis for identifying traffic congestion points and hot spots. Obtaining the regional trajectory distribution characteristics of vehicles helps optimize parking lot layout and traffic management, improve urban traffic operation efficiency, and reduce traffic congestion problems. The evolution of multi-time point distribution reveals the distribution changes of vehicles in different periods of time, which helps to analyze the spatiotemporal variation of traffic flow. Generating regional multi-time point trajectory distribution data provides more comprehensive traffic data support for the intelligent parking dynamic scheduling system, helps optimize parking resource utilization and vehicle scheduling, and in-depth distribution law mining helps understand the complexity and regularity of urban traffic, providing a deeper reference basis for traffic management and planning. Constructing a digital urban traffic trajectory portrait improves the level of intelligence in urban traffic management, provides more accurate data support for the intelligent parking dynamic scheduling system, and optimizes urban traffic operation efficiency.

[0037] Preferably, step S3 specifically comprises the following steps:

[0038] Step S31: Obtain historical city parking operation logs;

[0039] Step S32: mining the personalized parking behavior of car owners on historical urban parking operation logs to extract the personalized parking behavior characteristics of car owners;

[0040] Step S33: Performing a spatiotemporal dynamic parking demand driving analysis based on the personalized parking behavior characteristics of the car owner and the urban digital traffic trajectory portrait to generate spatiotemporal parking demand data within the city;

[0041] Step S34: performing parking demand trend evolution on the spatiotemporal parking demand data in the city to generate urban parking demand trend data;

[0042] Step S35: Perform dynamic parking demand forecasting for future time periods on the urban parking demand trend data, thereby obtaining dynamic parking demand forecasting data.

[0043] The present invention records the data of past parking activities through historical urban parking operation logs. By analyzing these data, we can gain insights into parking patterns, hot spots and peak hours, and provide a basis for future parking demand prediction. Obtaining historical data is helpful for understanding the historical trend of urban parking operations, and provides a reference for formulating parking policies and optimizing parking resource allocation. By mining the personalized parking behaviors of car owners, we can understand the parking preferences, habits and behavior patterns of different car owners, and provide support for personalized parking services. Extracting the personalized parking behavior characteristics of car owners is helpful for personalized parking lot recommendations, optimizing parking route planning, and improving parking experience and efficiency. Combining the personalized parking behaviors of car owners with urban traffic trajectory data, we can more accurately analyze the spatiotemporal characteristics of parking demand, and provide support for parking resources. It provides a basis for reasonable allocation. Through demand-driven analysis, we can better understand the dynamic changes in parking demand, improve the utilization rate of parking resources, reduce congestion and car-hunting time, and analyze the trend evolution of parking demand to help predict future changes in parking demand, and provide long-term decision-making support for parking resource planning and management. Understanding the evolution trend of parking demand will help to adjust parking strategies and optimize parking services in a timely manner, improve the effectiveness and efficiency of urban parking management, and predict parking demand in future time periods. The allocation and arrangement of parking resources can be done in advance to avoid waste of parking resources and parking problems. Dynamic parking demand forecasting data helps to optimize parking resource allocation, improve the response speed of parking services, and provide urban transportation systems with smarter and more convenient parking solutions.

[0044] Preferably, the specific steps of step S32 are:

[0045] Calculate the owner's historical parking time based on the historical city parking operation log to obtain the owner's historical parking time;

[0046] Identify the parking location distribution of car owners based on historical city parking operation logs and extract the parking location distribution data of car owners in the city;

[0047] Carry out parking location preference analysis on the parking location distribution data of car owners in the city to obtain the parking location preference characteristics of car owners;

[0048] The personalized parking behavior of car owners is mined based on their historical parking duration and parking location preference characteristics to extract the personalized parking behavior characteristics of car owners.

[0049] The present invention reveals the duration of parking behavior by understanding the historical parking time of the car owner, which is helpful to predict future parking demand and optimize the allocation of parking resources. By calculating the historical parking time, the distribution pattern of parking time is identified, which provides a basis for formulating a more intelligent parking strategy. Identifying the distribution of the parking positions of the car owners helps the city understand parking hot spots and congestion points, and provides data support for improving the layout of parking facilities and traffic planning. Extracting parking position distribution data helps to analyze the spatial characteristics of parking demand and provide guidance for optimizing parking resource allocation and improving parking efficiency. Analyzing the parking position preferences of the car owners reveals the parking preferences and habits of the car owners, and provides a basis for personalized parking recommendations and route planning. Obtaining parking position preference characteristics helps to optimize the layout and management of parking facilities, improve parking resource utilization and user satisfaction. By mining the personalized parking behavior characteristics of the car owners, personalized parking services and customized parking solutions are provided to the car owners, and the parking experience is improved. Extracting personalized parking behavior characteristics helps the intelligent parking system better understand the needs of the car owners and realize intelligent parking resource allocation and management.

[0050] Preferably, the specific steps of step S4 are:

[0051] Step S41: extracting the user's current parking demand request data;

[0052] Step S42: Calculate the user's real-time location based on the user's current parking demand request data to obtain the user's real-time location;

[0053] Step S43: Calculate the user's estimated parking time based on the user's current parking demand request data to obtain the user's estimated parking demand timestamp;

[0054] Step S44: Calculate the optimal parking time for the user's real-time location and the user's estimated parking demand timestamp based on the dynamic parking demand prediction data, thereby obtaining the user's current optimal parking time;

[0055] Step S45: Based on the user's current optimal parking time, intelligent parking demand recommendation is performed on the three-dimensional distribution visualization network of the parking lot, so as to obtain intelligent recommended parking lot location data.

[0056] The present invention extracts the user's current parking demand request data, so that the system accurately captures the user's parking demand and provides basic data support for subsequent personalized parking recommendations and route planning. Real-time acquisition of parking demand data helps the system quickly respond to user needs and improves the timeliness and accuracy of parking services. Calculating the user's real-time location helps the system accurately understand the user's current location and provides positioning data support for recommending the nearest parking lot and optimizing parking routes. Acquiring the user's real-time location helps improve the accuracy of parking recommendations and reduce the time and cost of users looking for parking lots. Estimating the user's parking time helps the system predict the user's parking time, providing a basis for recommending suitable parking lots and optimizing parking resource allocation, and obtaining the user's estimated parking demand. Requesting timestamps helps the system prepare and arrange parking resources in advance according to user needs, thereby improving the efficiency of parking services. Calculating the user's current optimal parking time helps the system recommend the best parking time for the user, avoiding parking problems during peak hours, and improving parking experience and efficiency. Based on the optimal parking time calculation, the system optimizes the user's parking experience, reduces user waiting time and congestion, and improves parking efficiency. Through intelligent parking demand recommendations, the system recommends the best parking lot location based on the user's real-time situation, improves the user's parking experience and saves parking time. Providing intelligent recommended parking lot location data helps optimize the utilization of parking resources, improve parking lot utilization and reduce congestion, and enhance the intelligence level of urban parking services.

[0057] Preferably, the specific steps of step S5 are:

[0058] Step S51: performing a real-time road traffic status assessment on the intelligently recommended parking lot location data based on the city digital traffic trajectory portrait to obtain parking lot location traffic status assessment data;

[0059] Step S52: performing congestion risk analysis on the parking lot location traffic status assessment data to generate road traffic congestion risk data;

[0060] Step S53: Performing dynamic parking path planning based on the road traffic congestion risk data, thereby generating a dynamic parking path;

[0061] Step S54: making dynamic parking scheduling decisions for the dynamic parking path and constructing a dynamic parking scheduling strategy.

[0062] The present invention helps the system understand the current traffic conditions of the road through real-time road traffic status evaluation, provides real-time traffic information support for parking lot recommendation and parking path planning, obtains parking lot location traffic status evaluation data, helps the system adjust the parking recommendation strategy according to the actual traffic conditions, improves parking efficiency and user experience, congestion risk analysis helps the system evaluate the congestion of road traffic, provides users with parking path planning and recommended parking lot locations that avoid congestion, generates road traffic congestion risk data, helps the system to warn and avoid traffic congestion in advance, improves the efficiency of parking paths and user satisfaction, dynamic parking path planning optimizes parking paths according to road traffic congestion risk data, provides users with faster and smoother parking routes, reduces parking time and congestion risks, generates dynamic parking paths, helps the system provide users with flexible parking path selection according to real-time traffic conditions, improves parking efficiency and user experience, dynamic parking scheduling decisions adjust parking strategies according to real-time road traffic conditions and user needs, realizes personalized parking services and optimal parking scheduling, and constructs dynamic parking scheduling strategies that help the system flexibly adjust the allocation and management of parking resources according to actual conditions, improves parking efficiency and user satisfaction.

[0063] Preferably, the specific steps of step S6 are:

[0064] Step S61: Performing real-time intelligent parking scheduling based on a dynamic parking scheduling strategy and monitoring real-time parking scheduling data;

[0065] Step S62: performing multi-time point parking scheduling simulation on the real-time parking scheduling data, and extracting parking scheduling simulation data at multiple time points;

[0066] Step S63: performing a city parking demand peak prediction on the parking scheduling simulation data at multiple time points to generate parking demand peak time point prediction data;

[0067] Step S64: performing coordinated dispatch of parking demand in adjacent suburbs on the parking demand peak time point prediction data to obtain a coordinated dispatch strategy for suburban parking;

[0068] Step S65: Perform deep parking migration optimization on the suburban parking collaborative scheduling strategy and the dynamic parking scheduling strategy to build an intelligent parking collaborative scheduling model.

[0069] The present invention can adjust the allocation of parking resources in real time according to the dynamic parking scheduling strategy through real-time intelligent parking scheduling, optimize the parking process, and improve parking efficiency. Monitoring real-time parking scheduling data helps the system to grasp the parking situation in real time, adjust the parking strategy in time, and provide better parking services. Multi-time point parking scheduling simulation helps the system to evaluate the effect of parking scheduling strategies at different time points and optimize the utilization of parking resources. Extracting parking scheduling simulation data at multiple time points helps the system to better understand the changing trend of parking demand and provide a reference for future parking planning. Urban parking demand peak prediction can help the system predict the peak period of parking demand, reasonably allocate parking resources, improve the parking service level, and generate parking demand peak time. Point prediction data helps the system to formulate targeted parking scheduling strategies and prepare for peak hours in advance. The coordinated scheduling of parking needs in the suburbs can coordinate the allocation of parking resources in the city fringe and suburbs, improve overall parking efficiency, and alleviate parking pressure in the city center. The coordinated scheduling strategy for suburban parking helps to achieve the rational allocation of urban parking resources, improve parking utilization in marginal areas, and optimize urban parking layout. Deep parking migration optimization combines suburban parking coordinated scheduling with dynamic parking scheduling strategies to improve the efficiency and intelligence level of the overall parking system. Building an intelligent parking coordinated scheduling model helps the system better adapt to changes in urban parking needs, provide comprehensive parking solutions, and optimize the urban parking service experience.

[0070] In this specification, a big data analysis system for smart parking is provided, which is used to execute the big data analysis method for smart parking as described above, including:

[0071] The three-dimensional distribution visualization module is used to obtain all parking lot information in the city based on the parking management platform and mark parking lot nodes; based on all parking lot information in the city, the three-dimensional distribution visualization of parking lot nodes is visualized to build a three-dimensional distribution visualization network of parking lots;

[0072] Traffic trajectory module, used to obtain urban road traffic flow data based on multi-source sensors; conduct real-time vehicle dynamic movement analysis on urban road traffic flow data, and perform multi-time point distribution evolution to build a digital urban traffic trajectory portrait;

[0073] The parking demand prediction module is used to obtain historical city parking operation logs; perform parking demand trend evolution on historical city parking operation logs, and then perform dynamic parking demand prediction for future periods, thereby obtaining dynamic parking demand prediction data;

[0074] The parking demand recommendation module is used to calculate the optimal parking time and intelligently recommend parking demands for the three-dimensional distribution visualization network of parking lots based on dynamic parking demand prediction data, thereby obtaining intelligently recommended parking lot location data;

[0075] The parking scheduling decision module is used to plan the dynamic parking path for the intelligent recommended parking lot location data based on the city's digital traffic trajectory portrait, and then make dynamic parking scheduling decisions to build a dynamic parking scheduling strategy;

[0076] The collaborative scheduling module is used to simulate multi-time point parking scheduling based on dynamic parking scheduling strategies, perform in-depth parking migration optimization, and build an intelligent parking collaborative scheduling model.

[0077] The present invention uses three-dimensional distribution visualization to intuitively present the location and distribution of all parking lots in the city, helping users to better understand the distribution of parking resources. Building a three-dimensional distribution visualization network of parking lots helps city planners and users to quickly find surrounding parking lots, improve parking retrieval efficiency, and optimize parking experience. Real-time vehicle dynamic movement analysis and multi-time point distribution evolution can help urban traffic managers better grasp urban road traffic conditions and adjust traffic management strategies in a timely manner. Building a digital urban traffic trajectory portrait helps to deeply understand the changing trend of urban traffic flow and provide data support for traffic planning and optimization. Dynamic parking demand prediction data helps the parking management platform to effectively plan parking resources, allocate parking resources in advance, and improve parking efficiency. Through the analysis of historical urban parking operation logs and the evolution of parking demand trends, the system can more accurately predict parking demand in future time periods and provide smarter parking services. , the optimal parking time calculation and intelligent parking demand recommendation can help drivers find the best parking lot and parking time, reduce the time and cost of finding parking spaces, and provide intelligent recommended parking lot location data to optimize parking lot resource utilization, alleviate parking congestion, and improve urban parking efficiency. Dynamic parking path planning and parking scheduling decisions based on urban digital traffic trajectory portraits can optimize urban parking processes, improve parking efficiency, and reduce traffic congestion. The construction of dynamic parking scheduling strategies dynamically adjusts parking resource allocation according to real-time traffic conditions and parking needs, providing more flexible and efficient parking services. Multi-time point parking scheduling simulation and deep parking migration optimization improve the overall utilization of parking resources, optimize collaborative scheduling between parking lots, and reduce urban parking pressure. The construction of an intelligent parking collaborative scheduling model helps to achieve the optimal configuration of parking resources, provide smarter and more efficient parking services, and improve the level of urban parking management. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 A schematic diagram of the steps of a big data analysis method for intelligent parking according to the present invention;

[0079] Figure 2 Detailed implementation flow chart of step S1;

[0080] Figure 3 Detailed implementation flow chart of step S2;

[0081] Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0082] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0083] The present application example provides a big data analysis method and system for smart parking. The execution subject of the big data analysis method and system for smart parking includes but is not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0084] See also Figures 1 to 4 The present invention provides a big data analysis method for smart parking, and the big data analysis method for smart parking comprises the following steps:

[0085] Step S1: Based on the parking management platform, all parking lot information in the city is obtained and parking lot nodes are marked; based on the information of all parking lots in the city, the parking lot nodes are visualized in three dimensions to construct a three-dimensional distribution visualization network of parking lots;

[0086] Step S2: Acquire urban road traffic flow data based on multi-source sensors; perform real-time vehicle dynamic movement analysis on urban road traffic flow data, and perform multi-time point distribution evolution to construct a digital urban traffic trajectory portrait;

[0087] Step S3: Obtain historical city parking operation logs; perform parking demand trend evolution on the historical city parking operation logs, and then perform dynamic parking demand forecasting for future periods, thereby obtaining dynamic parking demand forecasting data;

[0088] Step S4: Calculate the optimal parking time and intelligently recommend parking requirements for the three-dimensional distribution visualization network of the parking lot based on the dynamic parking demand prediction data, thereby obtaining intelligently recommended parking lot location data;

[0089] Step S5: Perform dynamic parking path planning on the location data of the intelligently recommended parking lot based on the city's digital traffic trajectory portrait, and then make dynamic parking scheduling decisions to build a dynamic parking scheduling strategy;

[0090] Step S6: Perform multi-time point parking scheduling simulation based on the dynamic parking scheduling strategy, and perform deep parking migration optimization to build an intelligent parking collaborative scheduling model.

[0091] The present invention visualizes the three-dimensional distribution of all parking lot information in the city, intuitively displays the spatial distribution of parking lots, helps users quickly understand the distribution of parking resources, and constructs a three-dimensional distribution visualization network of parking lots, which helps optimize the utilization of parking resources, improves parking lot management efficiency, reduces parking congestion and the time spent looking for parking spaces. Through real-time vehicle dynamic movement analysis and multi-time point distribution evolution, a digital urban traffic trajectory portrait is generated, which helps to gain a deep understanding of the distribution of urban traffic flow and traffic congestion conditions. The construction of a digital urban traffic trajectory portrait provides real-time traffic information support for intelligent parking scheduling, helps optimize parking path planning and reduce traffic congestion. Through the analysis of historical urban parking operation logs and the evolution of parking demand trends, the dynamic parking demand in future time periods is accurately predicted, providing a basis for the rational allocation of urban parking resources. The generation of dynamic parking demand prediction data helps to achieve intelligent parking lot location recommendation and improve parking lot Resource utilization, reduce parking search time, calculate the optimal parking time and intelligent parking demand recommendation based on dynamic parking demand prediction data, provide users with personalized parking suggestions, reduce parking time and cost, and generate intelligent recommended parking lot location data to help optimize user parking experience, improve parking efficiency, and reduce urban parking congestion and emissions. Dynamic parking path planning and parking scheduling decisions are made based on urban digital traffic trajectory portraits to achieve intelligent parking path planning and reduce traffic congestion. Building a dynamic parking scheduling strategy helps improve urban traffic operation efficiency, optimize parking resource allocation, and improve urban traffic management. Through multi-time point parking scheduling simulation and deep parking migration optimization, an intelligent parking collaborative scheduling model is built to achieve the optimal configuration and management of parking resources. The intelligent parking collaborative scheduling model helps to improve the utilization efficiency of urban parking resources, reduce traffic congestion and emissions, and improve the intelligence level of urban traffic operation.

[0092] In the embodiment of the present invention, refer to Figure 1 , is a schematic diagram of the steps of a big data analysis method for smart parking of the present invention. In this example, the steps of the big data analysis method for smart parking include:

[0093] Step S1: Based on the parking management platform, all parking lot information in the city is obtained and parking lot nodes are marked; based on the information of all parking lots in the city, the parking lot nodes are visualized in three dimensions to construct a three-dimensional distribution visualization network of parking lots;

[0094] In this embodiment, the interface or database of the parking management platform is determined to ensure access to relevant information of the parking lot, including the parking lot name, location (latitude and longitude), capacity, currently available parking spaces, charging standards, etc., and data is extracted from the parking management platform using API calls or SQL queries. The Python requests library (for API) or pandas library (for database) is used to obtain data, and the extracted parking lot information is cleaned to ensure data integrity and format consistency, including removing duplicates and processing missing values, etc., and ensuring that each parking lot information contains necessary fields, such as parking lot ID, name, latitude and longitude, capacity, etc. According to the extracted parking lot information, a query is generated for each parking lot. Create a node. The node should contain the unique identifier (ID), name, and location information (latitude and longitude) of the parking lot. Use data structures in Python (such as dictionaries or classes) to represent parking lot nodes for subsequent visualization. Select a suitable 3D visualization library, such as matplotlib's 3D function, Plotly, or Three.js (suitable for Web applications). Use the selected visualization tool to draw the parking lot nodes in 3D space. The location of the nodes is mapped according to the latitude and longitude of the parking lot to form a 3D distribution. To enhance the visualization effect, apply different colors or sizes to different parking lot nodes to indicate the number of available parking spaces or other characteristics of the parking lot.

[0095] Step S2: Acquire urban road traffic flow data based on multi-source sensors; perform real-time vehicle dynamic movement analysis on urban road traffic flow data, and perform multi-time point distribution evolution to construct a digital urban traffic trajectory portrait;

[0096] In this embodiment, multiple sensor types used in the city are determined, such as traffic cameras, radars, geomagnetic sensors, GPS devices, etc. These sensors can provide real-time traffic flow data, and a data integration platform is designed and built. It can receive and process data from different sensors. It is an intermediate layer responsible for data collection, storage and preliminary processing, and a suitable data collection protocol (such as MQTT, HTTP REST API or WebSocket) is determined to obtain sensor data in real time. Traffic flow data is obtained from each sensor regularly or in real time, including information such as the number, speed, and type of vehicles, to ensure the real-time and accuracy of the data. The acquired traffic flow data is cleaned to remove duplicate records, process missing values ​​and outliers, and ensure data quality. The data from different sensors are standardized into a unified format to facilitate subsequent analysis and processing. For example, the consistency of fields such as timestamps and vehicle types is ensured. Dynamic tracking algorithms (such as Kalman filters, particle filters, etc.) are applied to analyze the real-time position and speed of each vehicle, generate vehicle dynamic trajectories, and obtain real-time and accurate data. Extract key features from vehicle dynamic data, such as flow density, average speed, degree of traffic congestion, etc., for subsequent analysis; divide the data into different time periods (such as peak period and off-peak period) to analyze traffic flow changes in different time periods; select appropriate time series analysis methods (such as moving average, exponential smoothing) to analyze the evolution trend of traffic flow data; record the results of evolution analysis in the database, and generate visualization charts (such as line charts, heat maps) to intuitively understand the changes in traffic flow; integrate the dynamic trajectory data of all vehicles to form a traffic trajectory database within the city; use the integrated trajectory data to build a digital traffic trajectory portrait of the city to reflect the traffic flow and dynamic changes in different time periods and different sections; use visualization tools (such as GIS software, data visualization platform) to visualize the traffic trajectory portrait to form a dynamic traffic flow map for traffic management and decision-making; apply the generated traffic trajectory portrait and analysis results to urban traffic management, congestion prediction, traffic planning, etc. to support the construction of intelligent transportation systems.

[0097] Step S3: Obtain historical city parking operation logs; perform parking demand trend evolution on the historical city parking operation logs, and then perform dynamic parking demand forecasting for future periods, thereby obtaining dynamic parking demand forecasting data;

[0098] In this embodiment, the sources of historical parking operation logs are determined, including urban traffic management departments, intelligent parking systems, parking lot management software, etc. The historical parking operation logs are extracted from the parking management system through database query (SQL) or API call, and the extracted data is cleaned using Python's pandas library to remove duplicate records and outliers to ensure the accuracy and completeness of the data. The data is converted into a unified time format, and the parking time is calculated to determine the key characteristics of parking demand, such as the average number of daily parking times, the average parking time, and the parking space utilization rate. The historical parking data is divided into multiple time windows according to time (such as hours, days, and weeks) to facilitate the analysis of demand trends. Pandas is used to perform statistical analysis on parking data in different time windows, calculate the parking demand characteristics in each time window, draw demand trend graphs (such as line graphs and bar graphs), and visualize the changing trend of parking demand. An appropriate time series analysis model (such as ARIMA and seasonal decomposition models) is selected to analyze the evolution trend of parking demand. The selected model is applied to analyze parking demand to identify the seasonality and weekly characteristics of the trend. Generate a parking demand trend report based on the characteristics of parking demand, summarize the changes in parking demand, and identify demand peaks and troughs. According to the results of demand trend analysis, select a suitable dynamic prediction model, such as: ARIMA model: suitable for linear time series prediction, machine learning model: such as random forest, support vector machine (SVM), etc., suitable for processing complex nonlinear relationships, prepare training and test sets, including historical parking data and extracted parking demand features, ensure the time series integrity of the data set, use the training set to train the selected model, adjust the parameters to improve the prediction accuracy, evaluate the prediction performance of the model through the test set, and use indicators (such as root mean square error RMSE, mean absolute error MAE) for verification. Based on the trained model, predict the parking demand in the future time period and generate dynamic parking demand prediction data. The prediction results should include the peak values ​​of demand and demand in different time periods. Store the dynamic parking demand prediction data in the database for subsequent use, and use visualization tools (such as Matplotlib, Seaborn) to generate visualization charts of the prediction results for easy understanding and analysis.

[0099] Step S4: Calculate the optimal parking time and intelligently recommend parking requirements for the three-dimensional distribution visualization network of the parking lot based on the dynamic parking demand prediction data, thereby obtaining intelligently recommended parking lot location data;

[0100] In this embodiment, the location information of the parking lot is collected, including longitude and latitude, number of parking spaces, current parking space occupancy, etc., a three-dimensional distribution model of the parking lot is constructed, and a geographic information system (GIS) tool (such as ArcGIS, QGIS) is used to model the location and availability of the parking lot. The predicted parking demand data is divided into several time windows (for example, by hour or by day) to analyze the parking demand in different time periods, and a suitable demand analysis model (such as linear regression, time series model) is selected to calculate the optimal parking time for each parking lot in different time periods. The historical data of the parking lot and the dynamic demand forecast are combined to identify the peak and trough of demand. For each parking lot, the optimal parking time is calculated in different time windows, and the current parking space occupancy, historical usage rate and predicted demand are considered. Algorithms (such as optimization algorithms, heuristic algorithms) are used to determine the optimal parking time to ensure that the needs of car owners can be met to the greatest extent during peak demand periods. A suitable recommendation algorithm is selected, and commonly used ones include: content-based recommendation: recommendations are made based on the characteristics of the parking lot (such as location, number of parking spaces, etc.), collaborative filtering: based on the user's historical parking behavior To recommend parking lots, the optimal parking time and the three-dimensional distribution data of parking lots are combined to generate a data set containing parking lot characteristics and their optimal parking time. The recommendation algorithm is used to process the input data to generate a parking lot recommendation list for the user. Factors considered include: the distance between the user's current location and the parking lot, the number of available parking spaces in the parking lot, and the predicted peak parking demand. The recommended parking lot location data (including parking lot ID, recommendation reason, estimated parking time, etc.) is stored in the database for easy user access. The three-dimensional distribution of parking lots and the recommendation results are visualized using visualization tools (such as Plotly, Three.js, or Cesium). An interactive three-dimensional parking lot distribution map is created, and users can view the location, available parking spaces, and optimal parking time of different parking lots on the map. The recommended parking lots are marked on the map, and users can view the recommendation reasons and related information of the parking lots in real time. A friendly user interface is designed to allow users to obtain intelligently recommended parking lot information by entering their current location and demand period. A filtering function is provided to allow users to further filter based on conditions such as distance and number of parking spaces.

[0101] Step S5: Perform dynamic parking path planning on the location data of the intelligently recommended parking lot based on the city's digital traffic trajectory portrait, and then make dynamic parking scheduling decisions to build a dynamic parking scheduling strategy;

[0102] In this embodiment, the city's digital traffic trajectory portrait data and the intelligent recommended parking lot location data are extracted from the database to ensure that the data includes information such as real-time traffic conditions and available parking spaces in the parking lot. A suitable path planning algorithm is selected, such as: Dijkstra algorithm: suitable for calculating the shortest path from the starting point to all other points. A* algorithm: combined with heuristic search, the target path is found more efficiently. Traffic flow model: dynamic path planning considering real-time traffic conditions. Real-time traffic flow data and urban digital traffic trajectories are used to analyze the current road congestion and traffic flow direction, identify the best path, and calculate the optimal dynamic parking path based on the user's current location to the location of the recommended parking lot using the selected path planning algorithm. Factors considered include: path distance, real-time traffic conditions, and predicted driving time. The generated dynamic parking path data (including the distance of each path segment, the estimated driving time, etc.) is stored in the database for subsequent use. The goal of dynamic parking scheduling is determined, such as: optimizing the parking space utilization rate of the parking lot, reducing the waiting time of users, and improving the overall parking efficiency. Real-time parking space usage and user parking demand data of the parking lot are obtained from the monitoring system and sensors. A suitable scheduling model is selected, such as: heuristic algorithm :Such as genetic algorithm and ant colony algorithm, which are suitable for complex scheduling problems. Linear programming: used to optimize resource allocation to meet user needs. Based on real-time data and dynamic parking paths, implement dynamic parking scheduling decisions, calculate the available parking spaces in the current parking lot, dynamically adjust parking space allocation to respond to users' real-time needs, guide incoming vehicles to ensure that they can quickly find available parking spaces, use feedback mechanisms to monitor scheduling effects, such as user satisfaction, parking space utilization, etc., regularly adjust and optimize scheduling strategies, store dynamic parking scheduling strategies and their execution results in the database for subsequent analysis, and display real-time parking scheduling through visualization tools so that users can intuitively understand the current parking lot status and scheduling effects. Combine dynamic parking path planning and dynamic parking scheduling decisions to form a comprehensive dynamic parking scheduling strategy. Consider introducing machine learning algorithms to continuously optimize parking paths and scheduling strategies based on historical data and real-time feedback to improve the intelligence level of the system. Design a user feedback mechanism to collect users' experience and suggestions during the parking process in order to further improve services. Document the final dynamic parking scheduling strategy to ensure that team members can understand and implement the strategy. Update the strategy regularly to adapt to changes in urban traffic and parking needs.

[0103] Step S6: Perform multi-time point parking scheduling simulation based on the dynamic parking scheduling strategy, and perform deep parking migration optimization to build an intelligent parking collaborative scheduling model.

[0104] In this embodiment, the simulation goal and scope are determined, including the time period to be simulated (such as peak period, non-peak period) and the scheduling strategy, and a suitable simulation framework or tool (such as AnyLogic, MATLABSimulink or SimPy library in Python) is selected to extract the required historical parking data, real-time parking scheduling data and dynamic parking scheduling strategy from the database to ensure data integrity, including the number of parking spaces in the parking lot, historical utilization rate, user parking behavior and other information. According to the set time window, the parking scheduling simulation is run to simulate the parking demand and scheduling decision at different time points, and the parking situation at each time point, including the parking space usage, is recorded. The simulation results are displayed through visualization tools to analyze the changing trend of parking demand, conduct statistical analysis on the simulation results, evaluate the scheduling effect at different time points (for example, average parking time, parking space utilization rate, etc.), identify the parking behavior patterns during peak and off-peak hours, provide a basis for subsequent optimization, design a deep parking migration optimization model, determine the optimization goals (such as reducing the owner's search time, improving the efficiency of parking space use, etc.), and select appropriate optimization algorithms, such as: genetic algorithm: used to solve complex scheduling and resource allocation problems, particle swarm optimization: used to find the global optimal solution in multidimensional space, integrate real-time The parking lot usage data, user demand data and historical scheduling data provide input for the optimization model to ensure the timeliness and accuracy of the data to improve the reliability of the optimization results. The optimization algorithm is used to conduct in-depth analysis of parking migration and identify the best parking space scheduling and vehicle guidance strategies. By simulating scheduling strategies in different scenarios, the effects of each plan are evaluated and the optimal plan is selected to verify the optimization results. The optimized scheduling strategy is tested with actual data to evaluate its effectiveness. The optimization model is further adjusted based on feedback to ensure its adaptability and accuracy. The overall framework of the intelligent parking collaborative scheduling model is designed to clarify the model's input, processing logic and output results. Identify the key components of the model, including data collection, real-time scheduling, feedback mechanism, etc., integrate dynamic parking scheduling strategies and deep parking migration optimization results into the intelligent scheduling model, ensure that the model can respond to user needs and changes in traffic conditions in real time, and make dynamic adjustments. Introduce machine learning algorithms (such as reinforcement learning) so that the model can continuously optimize scheduling decisions based on historical data and user feedback. By training the model, it can predict user needs and automatically adjust scheduling strategies. Verify the constructed intelligent parking collaborative scheduling model, use actual data to test the model's prediction and scheduling capabilities, iteratively optimize the model based on the verification results, and continuously improve the performance and accuracy of the model.

[0105] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0106] Step S11: Obtain all parking lot information in the city based on the parking management platform and mark the parking lot nodes;

[0107] Step S12: performing geospatial positioning calculation on parking lot nodes to extract the geographical location coordinates of each parking lot;

[0108] Step S13: Performing urban spatial distribution analysis on the geographical location coordinates of each parking lot to generate urban spatial distribution data of parking lot nodes;

[0109] Step S14: performing node distribution discrete fitting on the parking lot node urban spatial distribution data to construct a parking lot urban distribution network;

[0110] Step S15: Based on the information of all parking lots in the city, the three-dimensional distribution visualization of the parking lot city distribution network is performed to construct a three-dimensional parking lot distribution visualization network.

[0111] In this embodiment, it is ensured that a connection is established with the API interface of the parking management platform, so that relevant information of all parking lots in the city can be obtained. Parking lot information, including parking lot name, address, capacity, opening hours, charging standards, etc., is obtained through API calls. The Python requests library is used to make API requests, and data is obtained and stored in JSON or CSV format. The extracted data is organized into a structured format (such as pandas DataFrame), and each parking lot node is marked for subsequent processing. The address of each parking lot is geocoded and converted into geographic coordinates (longitude and latitude). A geocoding service (such as Google Maps API, OpenStreetMap, etc.) is used. The geopy library in Python is used to obtain the longitude and latitude coordinates of the parking lot through address resolution. Errors or missing values ​​are handled to ensure that each parking lot has corresponding coordinates. The geographic location coordinates of all parking lots are extracted from the database to ensure that longitude and latitude information is included. A suitable spatial analysis method, such as K-means clustering, heat map analysis, or spatial statistical analysis, is selected to identify the spatial distribution characteristics of parking lots. A heat map of parking lots is generated using the seaborn or folium library to display the spatial distribution of parking lots in the city. Cluster analysis is performed on parking lots to identify high-density parking lots. For parking lot areas, use the KMeans clustering method in scikit-learn, and store the generated urban spatial distribution data (such as heat maps and clustering results) in the database for subsequent use. Select a suitable distribution fitting method, such as Gaussian distribution fitting, Poisson distribution fitting, or use graph theory methods to build a network. Perform discrete fitting on the spatial distribution data of parking lots. Analyze the distance and relative position between parking lots. Use the scipy.stats library in Python for distribution fitting to determine the distribution characteristics of parking lots. Convert the fitted data into a network model. Nodes represent parking lots, and edges represent the connection relationship between parking lots (such as distance or traffic flow). Use the NetworkX library to build a parking lot urban distribution network. Analyze the connectivity between nodes. Organize the geographic location coordinates of parking lots and their related attributes. Prepare data for three-dimensional visualization. Select a suitable three-dimensional visualization tool, such as matplotlib's 3D drawing function, Plotly, or Mayavi. Use the selected tool to draw a three-dimensional distribution network of parking lots. Nodes represent parking lots, and edges represent connection relationships. Set the size and color of nodes according to the capacity or other attributes of the parking lot to enhance the visualization effect. Save the generated three-dimensional distribution visualization results as image files or interactive web pages for easy display and sharing.

[0112] In this embodiment, the specific steps of step S15 are:

[0113] Calculate the vehicle entry and exit times of each parking lot in the city, and extract the vehicle entry and exit timestamps of each parking lot;

[0114] Calculate the available parking spaces in each parking lot based on the vehicle entry and exit timestamps to generate the number of available parking spaces in each parking lot;

[0115] Perform real-time status analysis of vacant parking spaces based on the number of vacant parking spaces in each parking lot, and generate real-time status data of vacant parking spaces;

[0116] Based on the real-time status data of vacant parking spaces, the city distribution network of parking lots is mapped in real time to obtain a vacant parking space status mapping distribution network;

[0117] Carry out topological correlation analysis between different parking lots in the urban distribution network of parking lots and extract the topological correlation features between parking lots;

[0118] Based on the topological association characteristics between parking lots, the three-dimensional distribution visualization of the vacant parking space status mapping distribution network is performed to construct a three-dimensional distribution visualization network of parking lots.

[0119] In this embodiment, the vehicle entry and exit records of each parking lot are extracted from the database, including vehicle ID, entry timestamp, exit timestamp and other information. The data is read and cleaned using Python's pandas library to ensure that the timestamp format is consistent. The entry and exit times of each parking lot are sorted to form a list containing entry and exit timestamps. The vehicle entry and exit timestamps of each parking lot are stored in a new data table for subsequent analysis. The entry and exit timestamps of each parking lot are sorted, and a timeline index is established. The number of available parking spaces in each parking lot is initialized (based on the parking lot capacity). The entry and exit timestamps are traversed to calculate the number of available parking spaces at each time point. The number of vehicles in the parking lot is used to deduce the number of vacant parking spaces. A counter is used to record the current number of vehicles: when a vehicle enters the parking lot, the number of vacant parking spaces is reduced, and when a vehicle leaves the parking lot, the number of vacant parking spaces is increased. The number of vacant parking spaces in each parking lot at different time points is stored in a data table, and the real-time status classification of the vacant parking spaces is determined, such as "full", "half full", and "empty". Based on the data on the number of vacant parking spaces, a threshold is set for status classification: for example, if the number of vacant parking spaces is 0, the status is "full"; if it is less than 50% of the total number of parking spaces, the status is "half full"; otherwise, it is "empty". The real-time status data of each parking lot is generated, and the data is recorded at each time point. The status of the parking lot and the number of available parking spaces are combined, and a parking lot status mapping scheme is designed to associate the real-time status with the parking lot city distribution network. The real-time status information of the parking lot is mapped to the parking lot city distribution network using Python's NetworkX library. The color or size of each node indicates its real-time status, such as "full" is represented by red, "half full" is represented by orange, and "empty" is represented by green. The network data of the status mapping is stored for subsequent visualization and analysis. The topological characteristics between parking lots, such as adjacency, connectivity, and flow, are determined. The topological structure of the parking lot city distribution network is analyzed using the NetworkX library, and the association characteristics between parking lots, such as degree centrality and betweenness centrality, are extracted. The connection status and relative importance between parking lots are identified. Appropriate three-dimensional visualization tools are selected, such as matplotlib's 3D function, Plotly, etc. are used for visualization. The extracted topological association features and real-time status data are used to draw a three-dimensional distribution network of the parking lot. In the three-dimensional visualization, the node represents the parking lot, the color and size of the node indicate the real-time status of the available parking spaces, and the connecting line indicates the association between parking lots. The generated three-dimensional distribution visualization results are saved as image files or interactive charts for easy display and sharing.

[0120] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0121] Step S21: Acquire urban road traffic flow data based on multi-source sensors;

[0122] Step S22: removing abnormal values ​​from the urban road traffic flow data to obtain abnormal filtered traffic flow data;

[0123] Step S23: Perform real-time vehicle dynamic movement analysis on the abnormal filtered traffic flow data to generate real-time dynamic trajectory of vehicles;

[0124] Step S24: performing regional trajectory distribution analysis on the real-time dynamic trajectory of the vehicle to obtain regional trajectory distribution characteristics of the vehicle;

[0125] Step S25: performing multi-time point distribution evolution on the regional trajectory distribution characteristics of the vehicle to generate regional multi-time point trajectory distribution data;

[0126] Step S26: Conduct in-depth distribution law mining on the regional multi-time point trajectory distribution data to construct a digital traffic trajectory portrait of the city.

[0127] In this embodiment, multi-source sensors are arranged on the main roads and intersections of the city to obtain traffic flow data. The sensors include cameras, geomagnetic sensors, radar sensors, etc. The traffic flow data from the sensors is obtained in real time through API or data interface, and the data is ensured to contain timestamp, vehicle count, speed, direction and other information. The acquired traffic flow data is stored in the database, and an appropriate data format (such as time series data) is used for subsequent analysis. A suitable outlier detection method is selected, such as Z-score, IQR (interquartile range) or machine learning method (such as isolation forest). The traffic flow data is analyzed using Python's pandas library, and the traffic statistical characteristics at each time point are calculated. The selected The outlier detection method selected is used to identify and remove abnormal traffic data, and the traffic flow data after abnormal filtering is stored in a new data table for subsequent analysis. The dynamic information of the vehicle is extracted from the traffic flow data after abnormal filtering, ensuring that it contains timestamps, vehicle IDs, location coordinates, etc. The dynamic movement of the vehicle is analyzed using a trajectory analysis algorithm (such as trajectory smoothing based on Kalman filtering), and the real-time dynamic trajectory of each vehicle is generated. The displacement and speed of each vehicle in different time periods are calculated, and the generated real-time dynamic trajectory data of the vehicle is stored in the database, ensuring that each trajectory contains vehicle ID, timestamp and location information. The city is divided into several areas according to its geographic information to ensure that the regional division is consistent with the traffic flow characteristics. According to the requirements of the analysis, the real-time dynamic trajectory data of vehicles is regionalized for analysis. The number of vehicle trajectories, average speed and other indicators in each area are counted. The pandas and geopandas libraries are used for regional trajectory statistics. The vehicle trajectory distribution characteristic data of each area is generated. The vehicle activities in each area are recorded. According to the analysis requirements, the time is divided into multiple time periods (such as peak hours, non-peak hours, weekdays and weekends, etc.). The vehicle trajectories in each area in different time windows are counted. The trend of the number of trajectories and the distribution characteristics are analyzed. The time series analysis method is used to analyze the traffic changes at different time points. The regional multi-time point trajectory distribution data is generated, and the vehicle activities in each area in different time windows are recorded. The regional The multi-time point trajectory distribution data of the domain is used to extract key features, such as vehicle flow, speed, and dwell time. Appropriate deep learning models (such as LSTM, GRU) or clustering analysis methods (such as K-means, DBSCAN) are selected to explore the deep laws of traffic trajectories. Deep learning frameworks such as TensorFlow or PyTorch are used to train the model to identify the potential patterns and laws of traffic flow. Through clustering analysis, the traffic flow characteristics of different regions or time periods are identified. Based on the extracted deep distribution laws, a digital urban traffic trajectory portrait is constructed to depict the spatial and temporal distribution characteristics of urban traffic flow. The generated traffic trajectory portrait is visualized to facilitate the display and analysis of the characteristics of urban traffic flow.

[0128] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include:

[0129] Step S31: Obtain historical city parking operation logs;

[0130] Step S32: mining the personalized parking behavior of car owners on historical urban parking operation logs to extract the personalized parking behavior characteristics of car owners;

[0131] Step S33: Performing a spatiotemporal dynamic parking demand driving analysis based on the personalized parking behavior characteristics of the car owner and the urban digital traffic trajectory portrait to generate spatiotemporal parking demand data within the city;

[0132] Step S34: performing parking demand trend evolution on the spatiotemporal parking demand data in the city to generate urban parking demand trend data;

[0133] Step S35: Perform dynamic parking demand forecasting for future time periods on the urban parking demand trend data, thereby obtaining dynamic parking demand forecasting data.

[0134] In this embodiment, the source of the historical parking operation log is determined, including data from the city parking management system, the smart parking cloud platform or the traffic management department. The parking operation log is extracted using SQL query or API call to ensure that the data contains information such as timestamp, parking lot ID, owner ID, parking duration, entry and exit time, etc. The extracted historical parking operation log is stored in a database, and an appropriate data format (such as CSV or SQL table) is used for subsequent analysis. The historical parking operation log is organized into a structured format to ensure that necessary fields are included for analysis. The parking data is analyzed using the pandas library to extract the personalized parking behavior characteristics of the owner, such as: parking frequency: the number of times each owner parks in a certain period of time, parking duration: the duration of each parking, parking preference: the parking lot or area that the owner prefers to use, cluster analysis (such as K-means or DBSCAN) is used to identify the parking behavior pattern of the owner in order to classify different types of owners, integrate personalized parking behavior characteristics with urban digital traffic trajectory portrait data to ensure the temporal and spatial consistency of the data, analyze the parking demand in different areas and time periods based on personalized parking behavior characteristics, and use time series analysis methods (such as ARIMA model) to capture Capture the changes in parking demand in different time periods, combine real-time traffic flow data, generate spatiotemporal parking demand data in the city, record the demand of each parking lot in different time periods, use tools such as pandas and numpy to perform statistical analysis on the spatiotemporal parking demand data in the city, calculate the historical average, maximum and minimum values ​​of the demand, select a suitable trend analysis model, such as moving average method, exponential smoothing method or linear regression analysis, apply the selected model to analyze the parking demand data, identify the evolution trend of demand, generate urban parking demand trend data, store the evolved parking demand trend data in the database for subsequent dynamic prediction, select a suitable dynamic demand prediction model, such as time series prediction model (ARIMA, SARIMA), machine learning model (such as random forest, XGBoost) or deep learning model (such as LSTM), prepare the data set required for prediction, including historical parking demand data, parking behavior characteristics and spatiotemporal demand data, use the training set to train the selected model, and use the test set to verify it, evaluate the performance of the model (such as MAE, RMSE, etc.), predict the parking demand in the future period based on the trained model, and generate dynamic parking demand prediction data.

[0135] In this embodiment, the specific steps of step S32 are:

[0136] Calculate the owner's historical parking time based on the historical city parking operation log to obtain the owner's historical parking time;

[0137] Identify the parking location distribution of car owners based on historical city parking operation logs and extract the parking location distribution data of car owners in the city;

[0138] Carry out parking location preference analysis on the parking location distribution data of car owners in the city to obtain the parking location preference characteristics of car owners;

[0139] The personalized parking behavior of car owners is mined based on their historical parking duration and parking location preference characteristics to extract the personalized parking behavior characteristics of car owners.

[0140] In this embodiment, relevant data including the vehicle owner ID, entry time and exit time are extracted from the historical city parking operation log, the entry time and exit time are converted into date and time format, and the parking time (in hours) is calculated to obtain the historical parking time of each vehicle owner, the parking location data in each parking record is extracted to ensure that the identification information of the parking lot or parking space is included, the parking data is grouped and counted, the usage frequency of each parking location is analyzed, the parking locations frequently used by the vehicle owner are identified, and the parking preference characteristics of the vehicle owner, such as the frequently used parking location, parking frequency and average parking time, are determined, and the parking location distribution data is used to identify the preferred parking location and parking behavior of each vehicle owner, the parking habits and patterns are analyzed, and the historical parking time and parking location preference characteristics of the vehicle owner are integrated to form a complete personalized parking behavior data set, and the parking behavior of the vehicle owner is classified by cluster analysis and other methods to identify different types of parking habits and behavior patterns.

[0141] In this embodiment, step S4 includes the following steps:

[0142] Step S41: extracting the user's current parking demand request data;

[0143] Step S42: Calculate the user's real-time location based on the user's current parking demand request data to obtain the user's real-time location;

[0144] Step S43: Calculate the user's estimated parking time based on the user's current parking demand request data to obtain the user's estimated parking demand timestamp;

[0145] Step S44: Calculate the optimal parking time for the user's real-time location and the user's estimated parking demand timestamp based on the dynamic parking demand prediction data, thereby obtaining the user's current optimal parking time;

[0146] Step S45: Based on the user's current optimal parking time, intelligent parking demand recommendation is performed on the three-dimensional distribution visualization network of the parking lot, so as to obtain intelligent recommended parking lot location data.

[0147] In this embodiment, a good data interface is established with the front-end application (such as a mobile application or a web page) of the user's parking demand request. When the user initiates a parking demand request, the request data is extracted, including information such as the user ID, request time, and requested parking type (such as temporary parking or long-term parking). The user's real-time location data is obtained using GPS or other positioning services to ensure the accuracy of the data. The user's longitude and latitude coordinates are obtained by calling a positioning API (such as Google Maps API or OpenStreetMap), and the positioning error is processed to ensure the accuracy of the user's location. The user's estimated parking time (such as the parking time entered by the user or the default time) is obtained through the user's parking demand request data. The user's estimated parking time is added to the current time to calculate the user's estimated parking demand timestamp. For example, if the current time is T and the estimated parking time is H, the estimated parking timestamp is T+H. Dynamic parking demand prediction data is extracted from the database to ensure that the data contains future The parking demand forecast information within the time period is analyzed based on the user's real-time location and expected parking timestamp combined with the dynamic parking demand forecast data. An algorithm (such as linear programming or heuristic algorithm) is implemented to calculate the user's current optimal parking time. Factors considered include: real-time demand: which areas have the lowest parking demand during the expected parking time period, distance: the distance from the user's current location to each parking lot, parking space availability: the predicted parking space availability of the parking lot, ensure that the three-dimensional distribution visualization network of the parking lot has been constructed, including the parking lot location, real-time status and available parking space data, select a suitable recommendation algorithm (such as content-based recommendation, collaborative filtering or hybrid recommendation) to generate intelligent recommendations, based on the user's optimal parking time, real-time location and parking lot distribution information, apply the recommendation algorithm to generate a list of recommended parking lots, calculate the priority of each recommended parking lot, consider factors including distance, available parking spaces, user's historical parking preferences, etc., display the recommended parking lot location data to the user in a visual form, and mark the recommended parking lots on the map.

[0148] In this embodiment, the specific steps of step S5 are:

[0149] Step S51: performing a real-time road traffic status assessment on the intelligently recommended parking lot location data based on the city digital traffic trajectory portrait to obtain parking lot location traffic status assessment data;

[0150] Step S52: performing congestion risk analysis on the parking lot location traffic status assessment data to generate road traffic congestion risk data;

[0151] Step S53: Performing dynamic parking path planning based on the road traffic congestion risk data, thereby generating a dynamic parking path;

[0152] Step S54: making dynamic parking scheduling decisions for the dynamic parking path and constructing a dynamic parking scheduling strategy.

[0153] In this embodiment, the urban digital traffic trajectory portrait data is extracted from the database to ensure that the data contains real-time traffic flow, road conditions, speed and other information. The intelligent recommended parking lot location data is extracted to ensure the location information and related attributes of each parking lot. The location of the recommended parking lot is evaluated in real time using a traffic state evaluation model (such as a traffic flow model or a prediction model based on machine learning). The traffic state indicators around the parking lot are calculated, including: real-time traffic flow. Average vehicle speed. The impact of traffic events (such as accidents, construction, etc.). The evaluated parking lot location traffic state data is stored in the database, including the traffic state indicators of each parking lot. Determine the indicators for congestion risk analysis, such as the ratio of traffic flow to road capacity, the average speed reduction, etc. Based on the parking lot location traffic state evaluation data, the congestion risk value around each parking lot is calculated. Use statistical analysis methods (such as Z-score or IQR) to identify potential congestion risk areas. Generate road traffic congestion risk data and identify the congestion risk level (such as low, medium, and high) around each parking lot. Select an appropriate path planning algorithm (such as Dijkstra algorithm, A* algorithm or graph-based path planning method) to calculate the dynamic parking path from the user's current location to the recommended parking lot. Integrate real-time road traffic status data and congestion risk data to provide the latest road condition information for path planning. Calculate the optimal dynamic parking path based on the user's real-time location, the recommended parking lot location and road traffic status. Consider factors such as the shortest path, the lowest congestion risk and real-time traffic flow. Store the generated dynamic parking path in the database and ensure that it is accessible at any time. Determine the goals of dynamic parking scheduling, such as reducing the owner's search time and optimizing the parking space utilization rate of the parking lot. Based on the dynamic parking path and real-time traffic status data, design a scheduling strategy to determine the parking order and path of the vehicle. Use optimization algorithms (such as linear programming, genetic algorithms, etc.) to optimize parking scheduling decisions. Based on the scheduling strategy, generate a real-time parking scheduling plan and match it with the user's parking needs. Store the constructed dynamic parking scheduling strategy in the database for real-time scheduling and subsequent analysis.

[0154] In this embodiment, the specific steps of step S6 are:

[0155] Step S61: Performing real-time intelligent parking scheduling based on a dynamic parking scheduling strategy and monitoring real-time parking scheduling data;

[0156] Step S62: performing multi-time point parking scheduling simulation on the real-time parking scheduling data, and extracting parking scheduling simulation data at multiple time points;

[0157] Step S63: performing a city parking demand peak prediction on the parking scheduling simulation data at multiple time points to generate parking demand peak time point prediction data;

[0158] Step S64: performing coordinated dispatch of parking demand in adjacent suburbs on the parking demand peak time point prediction data to obtain a coordinated dispatch strategy for suburban parking;

[0159] Step S65: Perform deep parking migration optimization on the suburban parking collaborative scheduling strategy and the dynamic parking scheduling strategy to build an intelligent parking collaborative scheduling model.

[0160] In this embodiment, according to the dynamic parking scheduling strategy, the parking space allocation and vehicle guidance of the parking lot are adjusted in real time, the scheduling system is used to receive the user's parking request, and the decision is made based on the real-time data (such as parking space availability, traffic conditions, etc.), a real-time monitoring system is implemented to track the parking space usage and vehicle entry and exit records of the parking lot, sensors and surveillance cameras are used to collect data to ensure real-time update of the parking status, and the real-time parking scheduling data (such as scheduling decisions, parking space usage, etc.) are stored in the database for subsequent analysis and optimization, a multi-time point parking scheduling simulation model is designed, the simulation time range (such as peak period, non-peak period) and scheduling strategy are determined, necessary information is extracted from the real-time parking scheduling data and historical parking data, and the data set required for the simulation is constructed, and the parking scheduling simulation is run using a simulation tool (such as AnyLogic, MATLAB, or a custom Python simulation script) to generate parking scheduling data at multiple time points, analyze the scheduling effects at different time points, record information such as vehicle flow and parking space changes, analyze the parking scheduling simulation data at multiple time points, extract key features of parking demand (such as demand, utilization rate, etc.), and select a suitable time series prediction model (such as ARIMA , seasonal decomposition model or machine learning model) to predict the peak of parking demand, use the selected prediction model to analyze the parking demand, generate parking demand peak time point prediction data, identify the time period of peak demand, output the prediction results, investigate the parking demand in the adjacent suburbs, collect the capacity, utilization rate and user demand characteristics of suburban parking lots, design a collaborative scheduling strategy for parking demand in the adjacent suburbs, consider factors such as the geographical location of the parking lot, available parking spaces and real-time demand, implement a collaborative scheduling strategy for suburban parking lots based on the prediction of the peak time point of parking demand, optimize the use of parking resources in the adjacent areas, use scheduling algorithms (such as genetic algorithms, simulated annealing, etc.) to optimize the resource allocation between parking lots, design a deep parking migration optimization model, including elements of dynamic scheduling strategy and suburban collaborative scheduling strategy, determine the optimization goals, such as reducing the search time of car owners, improving the utilization rate of parking spaces, etc., integrate the data of dynamic parking scheduling strategy and suburban parking collaborative scheduling strategy, prepare for deep optimization analysis, use optimization algorithms (such as linear programming, nonlinear programming or machine learning methods) to conduct deep analysis on parking migration, generate the optimal parking scheduling plan, simulate the scheduling effects under different scenarios, and evaluate the performance of the model.

[0161] In this embodiment, a big data analysis system for smart parking is provided, which is used to execute the big data analysis method for smart parking as described above, including:

[0162] The three-dimensional distribution visualization module is used to obtain all parking lot information in the city based on the parking management platform and mark parking lot nodes; based on all parking lot information in the city, the three-dimensional distribution visualization of parking lot nodes is visualized to build a three-dimensional distribution visualization network of parking lots;

[0163] Traffic trajectory module, used to obtain urban road traffic flow data based on multi-source sensors; conduct real-time vehicle dynamic movement analysis on urban road traffic flow data, and perform multi-time point distribution evolution to build a digital urban traffic trajectory portrait;

[0164] The parking demand prediction module is used to obtain historical city parking operation logs; perform parking demand trend evolution on historical city parking operation logs, and then perform dynamic parking demand prediction for future periods, thereby obtaining dynamic parking demand prediction data;

[0165] The parking demand recommendation module is used to calculate the optimal parking time and intelligently recommend parking demands for the three-dimensional distribution visualization network of parking lots based on dynamic parking demand prediction data, thereby obtaining intelligently recommended parking lot location data;

[0166] The parking scheduling decision module is used to plan the dynamic parking path for the intelligent recommended parking lot location data based on the city's digital traffic trajectory portrait, and then make dynamic parking scheduling decisions to build a dynamic parking scheduling strategy;

[0167] The collaborative scheduling module is used to simulate multi-time point parking scheduling based on dynamic parking scheduling strategies, perform in-depth parking migration optimization, and build an intelligent parking collaborative scheduling model.

[0168] The present invention uses three-dimensional distribution visualization to intuitively present the location and distribution of all parking lots in the city, helping users to better understand the distribution of parking resources. Building a three-dimensional distribution visualization network of parking lots helps city planners and users to quickly find surrounding parking lots, improve parking retrieval efficiency, and optimize parking experience. Real-time vehicle dynamic movement analysis and multi-time point distribution evolution can help urban traffic managers better grasp urban road traffic conditions and adjust traffic management strategies in a timely manner. Building a digital urban traffic trajectory portrait helps to deeply understand the changing trend of urban traffic flow and provide data support for traffic planning and optimization. Dynamic parking demand prediction data helps the parking management platform to effectively plan parking resources, allocate parking resources in advance, and improve parking efficiency. Through the analysis of historical urban parking operation logs and the evolution of parking demand trends, the system can more accurately predict parking demand in future time periods and provide smarter parking services. , the optimal parking time calculation and intelligent parking demand recommendation can help drivers find the best parking lot and parking time, reduce the time and cost of finding parking spaces, and provide intelligent recommended parking lot location data to optimize parking lot resource utilization, alleviate parking congestion, and improve urban parking efficiency. Dynamic parking path planning and parking scheduling decisions based on urban digital traffic trajectory portraits can optimize urban parking processes, improve parking efficiency, and reduce traffic congestion. The construction of dynamic parking scheduling strategies dynamically adjusts parking resource allocation according to real-time traffic conditions and parking needs, providing more flexible and efficient parking services. Multi-time point parking scheduling simulation and deep parking migration optimization improve the overall utilization of parking resources, optimize collaborative scheduling between parking lots, and reduce urban parking pressure. The construction of an intelligent parking collaborative scheduling model helps to achieve the optimal configuration of parking resources, provide smarter and more efficient parking services, and improve the level of urban parking management.

[0169] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0170] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A big data analysis method for intelligent parking, characterized in that: The following steps are involved: Step S1: Based on the parking management platform, all parking lot information in the city is obtained and parking lot nodes are marked; based on the information of all parking lots in the city, the parking lot nodes are visualized in three dimensions to construct a three-dimensional distribution visualization network of parking lots; Step S2: Acquire urban road traffic flow data based on multi-source sensors; perform real-time vehicle dynamic movement analysis on urban road traffic flow data, and perform multi-time point distribution evolution to construct a digital urban traffic trajectory portrait; Step S3: Obtain historical city parking operation logs; perform parking demand trend evolution on the historical city parking operation logs, and then perform dynamic parking demand forecasting for future periods, thereby obtaining dynamic parking demand forecasting data; Step S4: Calculate the optimal parking time and intelligently recommend parking requirements for the three-dimensional distribution visualization network of the parking lot based on the dynamic parking demand prediction data, thereby obtaining intelligently recommended parking lot location data; Step S5: Perform dynamic parking path planning on the location data of the intelligently recommended parking lot based on the city's digital traffic trajectory portrait, and then make dynamic parking scheduling decisions to build a dynamic parking scheduling strategy; Step S6: Based on the dynamic parking scheduling strategy, multi-time point parking scheduling simulation is performed, and deep parking migration optimization is performed to build an intelligent parking collaborative scheduling model; the specific steps of step S6 are: Step S61: Performing real-time intelligent parking scheduling based on a dynamic parking scheduling strategy and monitoring real-time parking scheduling data; Step S62: performing multi-time point parking scheduling simulation on the real-time parking scheduling data, and extracting parking scheduling simulation data at multiple time points; Step S63: performing a city parking demand peak prediction on the parking scheduling simulation data at multiple time points to generate parking demand peak time point prediction data; Step S64: performing coordinated dispatch of parking demand in adjacent suburbs on the parking demand peak time point prediction data to obtain a coordinated dispatch strategy for suburban parking; Step S65: Perform deep parking migration optimization on the suburban parking collaborative scheduling strategy and the dynamic parking scheduling strategy to build an intelligent parking collaborative scheduling model.

2. The big data analysis method for intelligent parking according to claim 1 is characterized in that: The specific steps of step S1 are: Step S11: Obtain all parking lot information in the city based on the parking management platform and mark the parking lot nodes; Step S12: performing geospatial positioning calculation on parking lot nodes to extract the geographical location coordinates of each parking lot; Step S13: Performing urban spatial distribution analysis on the geographical location coordinates of each parking lot to generate urban spatial distribution data of parking lot nodes; Step S14: performing node distribution discrete fitting on the parking lot node urban spatial distribution data to construct a parking lot urban distribution network; Step S15: Based on the information of all parking lots in the city, the three-dimensional distribution visualization of the parking lot city distribution network is performed to construct a three-dimensional parking lot distribution visualization network.

3. The big data analysis method for intelligent parking according to claim 2 is characterized in that: The specific steps of step S15 are: Calculate the vehicle entry and exit times of each parking lot in the city, and extract the vehicle entry and exit timestamps of each parking lot; Calculate the available parking spaces in each parking lot based on the vehicle entry and exit timestamps to generate the number of available parking spaces in each parking lot; Perform real-time status analysis of vacant parking spaces based on the number of vacant parking spaces in each parking lot, and generate real-time status data of vacant parking spaces; Based on the real-time status data of vacant parking spaces, the city distribution network of parking lots is mapped in real time to obtain a vacant parking space status mapping distribution network; Carry out topological correlation analysis between different parking lots in the urban distribution network of parking lots and extract the topological correlation features between parking lots; Based on the topological association characteristics between parking lots, the three-dimensional distribution visualization of the vacant parking space status mapping distribution network is performed to construct a three-dimensional distribution visualization network of parking lots.

4. The big data analysis method for intelligent parking according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: Acquire urban road traffic flow data based on multi-source sensors; Step S22: removing abnormal values ​​from the urban road traffic flow data to obtain abnormal filtered traffic flow data; Step S23: Perform real-time vehicle dynamic movement analysis on the abnormal filtered traffic flow data to generate real-time dynamic trajectory of vehicles; Step S24: performing regional trajectory distribution analysis on the real-time dynamic trajectory of the vehicle to obtain regional trajectory distribution characteristics of the vehicle; Step S25: performing multi-time point distribution evolution on the regional trajectory distribution characteristics of the vehicle to generate regional multi-time point trajectory distribution data; Step S26: Conduct in-depth distribution law mining on the regional multi-time point trajectory distribution data to construct a digital traffic trajectory portrait of the city.

5. The big data analysis method for intelligent parking according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: Obtain historical city parking operation logs; Step S32: mining the personalized parking behavior of car owners on historical urban parking operation logs to extract the personalized parking behavior characteristics of car owners; Step S33: Performing a spatiotemporal dynamic parking demand driving analysis based on the personalized parking behavior characteristics of the car owner and the urban digital traffic trajectory portrait to generate spatiotemporal parking demand data within the city; Step S34: performing parking demand trend evolution on the spatiotemporal parking demand data in the city to generate urban parking demand trend data; Step S35: Perform dynamic parking demand forecasting for future time periods on the urban parking demand trend data, thereby obtaining dynamic parking demand forecasting data.

6. The big data analysis method for intelligent parking according to claim 5 is characterized in that: The specific steps of step S32 are: Calculate the owner's historical parking time based on the historical city parking operation log to obtain the owner's historical parking time; Identify the parking location distribution of car owners based on historical city parking operation logs and extract the parking location distribution data of car owners in the city; Carry out parking location preference analysis on the parking location distribution data of car owners in the city to obtain the parking location preference characteristics of car owners; The personalized parking behavior of car owners is mined based on their historical parking duration and parking location preference characteristics to extract the personalized parking behavior characteristics of car owners.

7. The big data analysis method for intelligent parking according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: extracting the user's current parking demand request data; Step S42: Calculate the user's real-time location based on the user's current parking demand request data to obtain the user's real-time location; Step S43: Calculate the user's estimated parking time based on the user's current parking demand request data to obtain the user's estimated parking demand timestamp; Step S44: Calculate the optimal parking time for the user's real-time location and the user's estimated parking demand timestamp based on the dynamic parking demand prediction data, thereby obtaining the user's current optimal parking time; Step S45: Based on the user's current optimal parking time, intelligent parking demand recommendation is performed on the three-dimensional distribution visualization network of the parking lot, so as to obtain intelligent recommended parking lot location data.

8. The big data analysis method for intelligent parking according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: performing a real-time road traffic status assessment on the intelligently recommended parking lot location data based on the city digital traffic trajectory portrait to obtain parking lot location traffic status assessment data; Step S52: performing congestion risk analysis on the parking lot location traffic status assessment data to generate road traffic congestion risk data; Step S53: Performing dynamic parking path planning based on the road traffic congestion risk data, thereby generating a dynamic parking path; Step S54: making dynamic parking scheduling decisions for the dynamic parking path and constructing a dynamic parking scheduling strategy.

9. A big data analysis system for intelligent parking, characterized in that: The big data analysis method for executing the smart parking according to claim 1 comprises: The three-dimensional distribution visualization module is used to obtain all parking lot information in the city based on the parking management platform and mark parking lot nodes; based on all parking lot information in the city, the three-dimensional distribution visualization of parking lot nodes is visualized to build a three-dimensional distribution visualization network of parking lots; Traffic trajectory module, used to obtain urban road traffic flow data based on multi-source sensors; conduct real-time vehicle dynamic movement analysis on urban road traffic flow data, and perform multi-time point distribution evolution to build a digital urban traffic trajectory portrait; The parking demand prediction module is used to obtain historical city parking operation logs; perform parking demand trend evolution on historical city parking operation logs, and then perform dynamic parking demand prediction for future periods, thereby obtaining dynamic parking demand prediction data; The parking demand recommendation module is used to calculate the optimal parking time and intelligently recommend parking demands for the three-dimensional distribution visualization network of parking lots based on dynamic parking demand prediction data, thereby obtaining intelligently recommended parking lot location data; The parking scheduling decision module is used to plan the dynamic parking path for the intelligent recommended parking lot location data based on the city's digital traffic trajectory portrait, and then make dynamic parking scheduling decisions to build a dynamic parking scheduling strategy; The collaborative scheduling module is used to simulate multi-time point parking scheduling based on dynamic parking scheduling strategies, perform in-depth parking migration optimization, and build an intelligent parking collaborative scheduling model.

Images