A life circle planning rationality evaluation method and system based on people flow prediction

Abstract

The application discloses a life circle planning rationality evaluation method and system based on people flow prediction, and belongs to the field of city planning. The life circle planning rationality evaluation method based on people flow prediction comprises the following steps: collecting multi-source data, and constructing a community space-time database; constructing a community people flow trajectory prediction model based on the community space-time database, and predicting a community inflow trajectory vector set and an outflow trajectory vector set; simulating a community life circle based on the community inflow trajectory vector set and the outflow trajectory vector set; performing planning rationality evaluation on the simulated life circle; and outputting an evaluation result. The method uses OD flow prediction data to simulate a life circle, obtains community outflow and inflow trajectories, and further predicts a life circle service range, thereby breaking through the technical method that can only identify a community life circle based on present data, providing a simulation and rationality evaluation method for a planning community life circle after a community planning scheme is implemented, and improving evaluation accuracy.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of urban planning, and particularly relates to a life circle planning rationality evaluation method and system based on people flow prediction. BACKGROUND

[0002] Life circle planning and construction can improve the quality and satisfaction of residents' life, and therefore the life circle concept definition and service range demarcation problem has attracted widespread attention. However, the current life circle demarcation technology still adopts the method of taking life facilities as the center and directly demarcating according to the service radius, which has strong subjectivity and does not consider the real space-time activity of residents; the demarcated life circle deviates from the life circle generated by the actual activity of residents, which is not conducive to the implementation of community planning. Moreover, whether the community planning scheme is reasonable, such as whether the new land layout increases the commuting distance or causes uneven distribution of facilities, needs to be determined by the living range of residents after the scheme is implemented, however, the existing technology cannot predict and evaluate the situation after the implementation of the scheme relying on the existing facilities resources. SUMMARY

[0003] In view of the deficiencies of the prior art, the purpose of the present application is to provide a life circle planning rationality evaluation method and system based on people flow prediction, which relies on people flow data prediction technology to simulate and evaluate the rationality of the community life circle in the area after the implementation of the community planning scheme.

[0004] The purpose of the present application can be achieved by the following technical solutions:

[0005] A life circle planning rationality evaluation method based on people flow prediction, comprising the following steps:

[0006] Collecting multi-source data to construct a community space-time database;

[0007] Constructing a community people flow trajectory prediction model based on the community space-time database, and predicting a community inflow trajectory vector set and an outflow trajectory vector set;

[0008] Simulating a community life circle based on the community inflow trajectory vector set and the outflow trajectory vector set;

[0009] Evaluating the rationality of the simulated life circle;

[0010] Outputting the evaluation result.

[0011] Further, the step of constructing the community space-time database is:

[0012] S11, collecting geographic space-time data and community information data, and pre-processing the collected data to measure the basic attributes of the travel time and travel direction of the individual space-time trajectory data of the community regular users;

[0013] S12, spatially match the geospatial data and community information data with the community administrative boundary, extract the spatiotemporal flow data set and built environment data set of each community;

[0014] S13, data preprocessing is performed on the spatiotemporal flow data set and built environment data set of each community, and a community spatiotemporal flow database is constructed by summarizing each community as an object.

[0015] Further, in S11, the preprocessing of the collected data includes land unit division and identification of permanent users;

[0016] The land unit division is to take the community land function data as a basic unit boundary and number it.

[0017] The permanent user identification is that the preliminary screening of community permanent users according to the residence time means that the time difference between two adjacent and both located in the specified city positioning of the same user within a specified time interval is added to the cumulative time T of the corresponding user; whether the cumulative time T of the corresponding user is greater than or equal to the experience threshold value is judged, if T is greater than or equal to the experience threshold value, the corresponding user is a permanent resident, otherwise, the corresponding user is not a permanent resident.

[0018] Further, in S13, the data preprocessing includes land unit OD flow clustering and OD flow flow calculation.

[0019] The land unit OD flow clustering is to spatially summarize the individual spatiotemporal trajectory data of the land unit to generate the land unit OD flow.

[0020] The OD flow flow calculation process is to calculate the 24h trajectory flow of different land unit OD flows, remove the land unit OD flow whose 24h trajectory flow sum is less than 20, number the remaining land unit OD flows j, obtain the trajectory flow of the land unit OD flow numbered j in the time period t within 24 hours, construct a spatiotemporal complex clustering algorithm, cluster the individual spatiotemporal trajectory data in the land unit OD flow combined with the travel direction, and calculate the flow of each type of OD flow.

[0021] Further, the construction and prediction steps of the community flow trajectory prediction model include:

[0022] S21, according to the start and end point land properties, identify the community outflow trajectory and the community inflow trajectory; calculate the trajectory travel frequency, cluster the community inflow trajectory and the community outflow trajectory respectively, and use the word embedding model for vectorization processing to obtain the inflow trajectory vector training set and the outflow trajectory vector training set of each community;

[0023] S22, taking the community inflow trajectory vector training set, the outflow trajectory vector training set and the community land unit independent variable index as input variables, constructing a structural equation model, and extracting regression variable indexes between the variables;

[0024] S23, automatically generating the inflow trajectory vector set and the outflow trajectory vector set of each community in the target administrative area by the community trajectory prediction model constructed in S22.

[0025] Further, the regression variable indexes are solved by a gradient descent algorithm to obtain the goodness of fit, and if each item of the goodness of fit index of the structural equation model meets the fitting degree index adaptation standard, the final structural equation model is taken as the community people flow trajectory prediction model:

[0026] X1=Λ1x+δ1 (1)

[0027] X2=Λ2x+δ2 (2)

[0028] η=βx+Γx2+...+ζ (3)

[0029] Wherein, formula (1) and (2) are measurement models, formula (3) is a structural model, X1 is the outflow trajectory flow of the community, X2 is the inflow trajectory flow of the community, η is the matrix set of dependent variables X1 and X2; x is the independent variable; β is the coefficient matrix of the dependent variable; Γ is the coefficient matrix of the independent variable; ζ represents the residual error.

[0030] Further, the specific steps of simulating the community life circle include:

[0031] S31, marking the community inflow trajectory vector set in S24 as V1={V i}, marking the community outflow trajectory vector set as V2={V i}, filtering each prediction set according to the travel frequency and the distance interval d=(d min ,d max ), extracting the trajectory endpoints in the community outflow trajectory vector set to form a point set P1, and extracting the trajectory starting points in the community outflow trajectory vector set to form a point set P2;

[0032] S32, calculating the spatial range of the point sets P1 and P2 by using the standard deviation ellipse tool in the geographic information platform, and forming the simulated life circle {C, S} of each community.

[0033] Further, the calculation method of the standard deviation ellipse tool is:

[0034]

[0035]

[0036]

[0037]

[0038] Wherein, x and y are the coordinates of the point set P1, P2 respectively, {x, y} represents the average center of the point set, and n is the total number of points in each point set.

[0039] Further, the step of evaluating the simulated life circle is:

[0040] S41, evaluating the life circle planning rationality based on a compactness index, the compactness index being calculated as the ratio of the area of the community simulated life circle to the area of its minimum circumscribed circle;

[0041] S42, evaluating the life circle planning rationality based on a deviation index, the deviation index being calculated as the ratio of the difference between the distance of the minimum circumscribed circle center of the community simulated life circle and the community centroid and the average radius of the basic life circle;

[0042] The basic life circle refers to a walking range with the community centroid as the center and the 15min walking distance of the residents as the radius.

[0043] A life circle planning rationality evaluation system based on human flow prediction, comprising:

[0044] A data collection module: collecting multi-source data to construct a community space-time database;

[0045] A prediction model construction module: constructing a community human flow trajectory prediction model based on the community space-time database and predicting the community inflow trajectory vector set and outflow trajectory vector set;

[0046] A life circle simulation module: simulating the community life circle based on the community inflow trajectory vector set and outflow trajectory vector set;

[0047] An evaluation module: evaluating the planning rationality of the simulated life circle;

[0048] A result output module: outputting the evaluation result.

[0049] The beneficial effects of the present application are:

[0050] 1. The present application innovatively simulates the life circle by using OD flow prediction data, obtains the outflow and inflow point sets of the community, and further predicts the service range of the life circle, thereby breaking through the technical method of identifying the community life circle only based on the present situation data, providing a simulation and rationality evaluation method for the planning community life circle after the implementation of the community planning scheme, and significantly improving the evaluation accuracy.

[0051] 2、The application evaluates the planning rationality of the simulation life circle from the compactness index and the deviation index, effectively judges the influence of the land layout of the current community planning scheme on future human flow, and improves the design efficiency of designers. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0053] Figure 1 is a flow framework diagram of the method of the present application;

[0054] Figure 2 is a land unit division diagram in the present application;

[0055] Figure 3 is a variable relationship diagram of the structural equation model in the present application;

[0056] Figure 4 is a simulation schematic diagram of the life circle in the present application;

[0057] Figure 5 is a life circle rationality evaluation flow diagram in the present application;

[0058] Figure 6 is an evaluation result output schematic diagram in the present application. DETAILED DESCRIPTION

[0059] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0060] The technical solutions of the present application will be described in detail below with reference to a case of a certain place;

[0061] As shown in the figure, a life circle planning rationality evaluation method based on human flow prediction comprises the following steps: Figure 1

[0062] S1, collect multi-source data and construct a community space-time database;

[0063] Specifically, it comprises:

[0064] S11, collect geographic space-time data and community information data;

[0065] ​Specifically: obtaining geographic spatio-temporal data and community information data within the target city range from the regional data opening platform, the community information data including community administrative boundary vector data, industry POI point data, road network data, land use function data, building data, community population data, and family annual income data, and the geographic spatio-temporal data being community present population spatio-temporal distribution data represented by mobile location service data (LBS data). The mobile location service data is preprocessed, and the basic attributes of travel time and travel direction of individual spatio-temporal trajectory data of community regular users are measured.

[0066] The preprocessing of the mobile location service data includes: land unit division and identification of regular users; as shown in Figure 2 the land unit division is to take the community land function data as the basic unit boundary and number i (i = 1, 2,..., n); and the identification of regular users is to preliminarily screen community regular users according to the residence time, that is, to accumulate the time difference between two adjacent and both located in the specified city positioning of the same user within a set time interval into the cumulative time T of the corresponding user; to determine whether the cumulative time T of the corresponding user is greater than or equal to the empirical threshold value, if T is greater than or equal to the empirical threshold value, the corresponding user is a regular resident, otherwise, the corresponding user is not a regular resident.

[0067] In addition, the basic attributes of travel time and travel direction of individual spatio-temporal trajectory data are measured in the following way:

[0068] The travel time is to divide the individual spatio-temporal trajectory data into 24 categories according to the time of reaching the trajectory endpoint, and to identify the travel time T (T = 0, 1, 2,..., 23);

[0069] The travel direction refers to the individual spatio-temporal trajectory data whose starting land unit number is i x and the ending land unit is i y .

[0070] The community regular users are preliminarily screened according to the residence time, that is, to accumulate the time difference between two adjacent and both located in the specified city positioning of the same user within a set time interval into the cumulative time T of the corresponding user; to determine whether the cumulative time T of the corresponding user is greater than or equal to the empirical threshold value, if T is greater than or equal to the empirical threshold value, the corresponding user is a regular resident, otherwise, the corresponding user is not a regular resident.

[0071] The above data is converted into a skp format file and accurately positioned to a specific spatial location; and stored on a workstation configured with an Intel Xeon Processor E5-2620V4 processor, 512G SSD, and 128G dDDR4 memory, and specific operations are performed on the workstation.

[0072] S12, Extracting the spatiotemporal flow dataset and the built environment dataset: Spatially match the geographic spatiotemporal data, community information data and community administrative boundaries to extract the spatiotemporal flow dataset and the built environment dataset for each community.

[0073] Taking community land use units as the object, based on the current data, the relevant indicators of the built environment of each community are calculated: community economic development characteristics, community spatial form characteristics, community transportation accessibility characteristics, community business facilities characteristics, and community population composition characteristics;

[0074] The characteristics of community economic development are: annual household income, in units of 10,000 yuan / year;

[0075] Community spatial morphology features include plot ratio and building density. Plot ratio is the ratio of the total building area to the land area of ​​the community, and building density is the ratio of the sum of the base areas of the buildings to the community land area.

[0076] Community accessibility features include road network density and transportation facility density. Road network density is the ratio of the sum of the centerline lengths of all roads in the community to the community area. Transportation facility density is the density of public transportation facilities, namely the density of subway stations and bus stops within the community.

[0077] The characteristics of community business facilities include the density of various business resource points. The facilities are classified into 3 major categories and 13 subcategories (as shown in Table 1). The density of business resource points refers to the ratio of the number of various business resource points in each community to the community land area.

[0078] Table 1 Facility Classification Table

[0079]

[0080]

[0081] The characteristics of the community's population composition include the proportion of migrant population and the proportion of elderly population. The data were obtained from the local subdistrict office or government. The proportion of migrant population is the ratio of residents with household registration outside the community to the total population of the community, and the proportion of elderly population is the ratio of residents aged 60 and above to the total population of the community.

[0082] S13, Constructing a community spatiotemporal flow database: Preprocess the spatiotemporal flow datasets and built environment datasets of each community in S12, and summarize and construct a community spatiotemporal flow database for each community as an object;

[0083] The preprocessing steps include: land use unit OD flow clustering and OD flow calculation;

[0084] Land use unit OD flow clustering is: spatially summarizing individual spatiotemporal trajectory data by land use unit to generate land use unit OD flow;

[0085] The OD flow calculation process is: calculating the 24-hour trajectory flow of different land unit OD flows, removing the land unit OD flows with a 24-hour trajectory flow total less than 20, numbering the remaining land unit OD flows j, obtaining the trajectory flow Qjt of the land unit OD flow numbered j in the time period t within 24 hours; constructing a spatio-temporal complex clustering algorithm, clustering the individual spatio-temporal trajectory data in the land unit OD flow in combination with the travel direction, and calculating the OD flow flow (Qjt-type) of each type.

[0086] In combination with the clustering results of the individual spatio-temporal trajectory data and the OD flow flow, the community spatio-temporal flow database is respectively summarized and constructed for each community as an object; the community spatio-temporal flow database includes the OD flow flow (Qjt-type) of each direction in each time period of 24 hours between two land units, the urban built environment data corresponding to the start and end point land x, y.

[0087] S2, constructing a human flow trajectory prediction model based on the community spatio-temporal flow database in S1, and predicting the trajectory of each community in the target administrative area; specifically including:

[0088] S21, constructing an inflow and outflow trajectory vector training set based on the community spatio-temporal flow database in S1;

[0089] According to the properties of the start and end point land, the community outflow trajectory and the community inflow trajectory are identified; the trajectory travel frequency is calculated, the community inflow trajectory and the community outflow trajectory are respectively clustered, and the vectorization processing is performed by using a word embedding model to obtain the inflow trajectory vector training set and the outflow trajectory vector training set of each community;

[0090] The trajectory vector training set includes the travel direction, the travel distance, and the corresponding travel frequency information.

[0091] It is worth mentioning that the land unit with a trajectory residence time proportion at night (22:00 to 06:00) greater than 90% is a residential area, and the land unit with a trajectory residence time proportion during the day (09:00 to 18:00) greater than 90% is an employment area.

[0092] The travel frequency of the trajectory is: the number of travel segments after the travel trajectory is interrupted by the residence point within 24 hours;

[0093] In addition, the trajectory with a residential land as a starting point and other community employment lands as end points in each community spatio-temporal flow database is taken as a community outflow trajectory; the trajectory with other community residential lands as starting points and the community employment lands as end points is taken as a community inflow trajectory. With the community administrative boundary as an object, all the community outflow trajectories and the community inflow trajectories in the same community are respectively clustered by using OPTICS, and input into a word embedding model to obtain the inflow trajectory vector set and the outflow trajectory vector set of each community.

[0094] S22, constructing a community people flow trajectory prediction model according to the inflow trajectory vector training set and the outflow trajectory vector training set based on the community spatio-temporal flow database obtained in S21;

[0095] The community inflow trajectory vector training set, the outflow trajectory vector training set and the community land unit independent variable index are taken as input variables to construct a structural equation model, and a regression variable index table between variables is extracted; the regression variable index is extracted by using a gradient descent algorithm to solve the goodness of fit, and if each goodness of fit index of the structural equation model meets the goodness of fit index adaptation standard, the finally obtained structural equation model is taken as the community people flow trajectory prediction model:

[0096] The basic framework of the equation is as follows:

[0097] X1=Λ1x+δ1 (1)

[0098] X2=Λ2x+δ2 (2)

[0099] η=βx+Γx2+...+ζ (3)

[0100] Wherein, formula 1 and formula 2 are measurement models, formula 3 is a structural model, X1 is community outflow trajectory flow, X2 is community inflow trajectory flow, η is a matrix set of dependent variables X1 and X2; x is an independent variable; β is a coefficient matrix of dependent variables; Γ is a coefficient matrix of independent variables; ζ represents a residual, which is a part that cannot be explained within the model, and the model variable construction is as shown in Figure 3

[0101] Further, the independent variable index table is obtained by extracting the independent variable index of each community built environment data set; the specific content of the index table is as follows:

[0102] Table 2 Independent variable index table

[0103] Serial number Index Code 1 Annual household income e1 2 Floor area ratio e2 3 Building density e3 4 Road network density e4 5 Traffic facility density e5 6 Format facility density e6 7 Proportion of floating population e7 8 Proportion of elderly population e8

[0104] Further, if each goodness of fit index of the structural equation model meets the goodness of fit index adaptation standard, the finally obtained structural equation model is taken as the community people flow trajectory prediction model, wherein according to the goodness of fit index adaptation standard, the model needs to be modified, mainly by arranging the model modification value in descending order and establishing the correlation between variables in order to modify the model, or deleting insignificant variables and paths to modify the model.

[0105] S23, target administrative district community trajectory model prediction;

[0106] ​Input the target administrative district where the community planning scheme data is located, and the built environment data form of the target administrative district is the same as S1; automatically generate the inflow trajectory vector set and outflow trajectory vector set of each community in the target administrative district by the community trajectory prediction model constructed by S22.

[0107] S3, based on the inflow trajectory vector set and outflow trajectory vector set of each community in the target administrative district predicted by S23, simulate the community life circle; the specific steps include:

[0108] S31, screen the prediction point set;

[0109] Mark the community inflow trajectory vector set in S24 as V1={V i}, and mark the community outflow trajectory vector set as V2={V i}, screen each prediction set according to the travel frequency and distance interval d=(d min ,d max ), and extract the trajectory endpoints in the community outflow trajectory vector set to form the point set P1; extract the trajectory starting points in the community inflow trajectory vector set to form the point set P2.

[0110] S32, simulate the community life circle;

[0111] As shown in Figure 4 , calculate the spatial range of the point sets P1 and P2 in the geographic information platform using the standard deviation ellipse tool, mark the spatial range of P1 as the travel life circle of each community {C=C i}, mark the spatial range of P2 as S, which is the service life circle of each community {S=S i}, and form the simulated life circle of each community {C, S}.

[0112] The calculation method of the standard deviation ellipse tool is as follows:

[0113]

[0114]

[0115]

[0116]

[0117] Wherein, x and y are the coordinates of the point sets P1 and P2, respectively, and n is the total number of points in each point set.

[0118] S4, evaluate the planning rationality of the simulated life circle of S3;

[0119] As shown in Figure 5The specific evaluation steps are shown as follows:

[0120] S41, evaluating the planning rationality of the life circle based on the compact index;

[0121] The compact index P is calculated as the ratio of the area A of the community simulation life circle {C, S} to the area A0 of the minimum circumscribed circle thereof. If the travel life circle compact index P_c is greater than or equal to 0.8, the life circle is marked as a “compliant life circle”, otherwise, Ci is marked as a “to-be-modified life circle”. The service life circle compact index P_s is determined in the same way. If both {C, S} are determined to be compliant, S42 is entered, otherwise, S51 is entered.

[0122] For the simulation life circle {C, S} in the embodiment, P_c = 0.9 and P_s = 0.81, both of which are determined to be compliant.

[0123] S42, evaluating the planning rationality of the simulated life circle based on the deviation index;

[0124] The deviation index Q is used to reflect the position deviation of each life circle in the simulation life circle {C, S} relative to the basic life circle Gi. The basic life circle refers to a walking range with the community centroid as the center and a 15-minute walking distance as the radius. The extraction method is to use the computational geometry tool to obtain the community centroid, and use the routing algorithm tool to obtain the range domain of the 15-minute basic life circle in combination with the community road network data, which is marked as Gi.

[0125] The calculation method of the deviation index Q is the ratio of the difference between the distance from the minimum circumscribed circle center of the community simulation life circle {C, S} and the community centroid to the average radius of Gi. The travel life circle deviation index is denoted as Q_c, and the service life circle deviation index is denoted as Q_s. If Q_c < 0.2 and Q_s < 0.2, it is determined to be a “travel and service dual-rational life circle”. If Q_c ≥ 0.2 and Q_s < 0.2, it is determined to be a “service rational life circle”. If Q_c < 0.2 and Q_s ≥ 0.2, it is determined to be a “travel rational life circle”. If any of the above conditions is met, it can jump to S53. If Q_c ≥ 0.2 and Q_s ≥ 0.2, it is marked as a “to-be-modified life circle” and S51 is entered.

[0126] For the embodiment area {C, S}, Q_c = 18% and Q_s = 40%, which are determined to be a “service rational life circle”.

[0127] S5, feedback and output of the evaluation result; the specific steps include:

[0128] S51, feedback of the evaluation result,

[0129] The communities determined as "to-be-modified life circle" in S41 and S42 are exported to the geographic spatial information platform, and the adjusted community geographic information data is fed back to S24 by changing the type, layout and quantity of community facilities in the geographic spatial information platform; until the compact index is determined to be compliant and the deviation index evaluation result is "travel service double reasonable life circle", "service reasonable life circle" and "travel reasonable life circle", S52 is entered.

[0130] S52, life circle simulation and evaluation thematic map generation;

[0131] The life circle simulation results and life circle planning rationality evaluation results in S32 are visualized in the geographic information platform.

[0132] The visualization mode of the life circle simulation results is that the community simulation life circle {C, S} in S32 is superimposed on the target administrative district map respectively, and is exported as a life circle simulation thematic map; the visualization mode of the life circle planning rationality evaluation results is that the spatial distribution of "travel service double reasonable life circle", "service reasonable life circle" and "travel reasonable life circle" in S41 and S42 is displayed in grades, and the compact index and deviation type information are output as a life circle planning rationality evaluation chart.

[0133] S53, life circle simulation and evaluation result output.

[0134] The obtained life circle simulation and planning rationality evaluation results are connected with the database of each planning management department, and a community life circle simulation and planning rationality evaluation report is output by a Stratasys industrial 3D printer, which is used to guide the land optimization decision of community planning scheme; as shown in Figure 6 .

[0135] In the description of the specification, the description of the terms "one embodiment", "example", "specific example" and the like means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the described specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0136] The basic principles, main features and advantages of the present application are shown and described above. It should be understood by those skilled in the art that the present application is not limited by the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application.

Claims

1. A method for evaluating the rationality of a life circle plan based on human flow prediction, characterized in that, The method comprises the following steps: Collecting multi-source data and constructing a community space-time database; Based on the community space-time database, a community flow trajectory prediction model is constructed, and a community inflow trajectory vector set and a community outflow trajectory vector set are predicted; Based on the community inflow trajectory vector set and the community outflow trajectory vector set, a community life circle is simulated; The simulated life circle is evaluated for planning rationality; The evaluation result is outputted; The construction and prediction of the community flow trajectory prediction model comprises the following steps: S21, according to the starting and ending point land use property, the community outflow trajectory and the community inflow trajectory are identified; the trajectory travel frequency is calculated, the community inflow trajectory and the community outflow trajectory are clustered respectively, and a word embedding model is used for vectorization processing to obtain an inflow trajectory vector training set and an outflow trajectory vector training set of each community; S22, the community inflow trajectory vector training set, the community outflow trajectory vector training set and the community land unit independent variable index are taken as input variables to construct a structural equation model, and regression variable indexes between variables are extracted; S23, the community inflow trajectory vector set and the community outflow trajectory vector set in the target administrative area are automatically generated through the community trajectory prediction model constructed in S22; The specific steps of simulating the community life circle comprise: S31, mark the community inflow trajectory vector set in S23 as V1={V i}, mark the community outflow trajectory vector set as V2={V i}, filter each prediction set according to the travel frequency and the distance interval d=(d min ,d max ), and extract the terminal points of each trajectory in the community outflow trajectory vector set to form a point set P1; extract the starting points of each trajectory in the community outflow trajectory vector set to form a point set P2; S32, the spatial range of the point set P1 and P2 is calculated by using a standard deviation ellipse tool in a geographic information platform, and a simulated life circle {C, S} of each community is formed; The calculation method of the standard deviation ellipse tool is: where x and y are the coordinates of the point sets P1, P2, respectively, denotes the average center of the point sets, and n is the total number of points in each point set. The steps of evaluating the simulated life circle are: S41, the planning rationality of the life circle is evaluated based on a compactness index, and the calculation method of the compactness index is the ratio of the area of the community simulated life circle to the area of its minimum circumscribed circle; S42, the planning rationality of the life circle is evaluated based on a deviation index; the calculation method of the deviation index is the ratio of the difference between the distance of the minimum circumscribed circle center of the community simulated life circle and the community centroid to the average radius of the basic life circle; The basic life circle refers to a walking range with the community centroid as the center and the 15-minute walking distance of a resident as the radius. 2.The method of claim 1, wherein, The steps of constructing the community space-time database are: S11, collecting geographic space-time data and community information data, and preprocessing the collected data to measure the basic attributes of individual space-time trajectory data of community resident users, such as travel time and travel direction; S12, the geographic space-time data and the community information data are spatially matched with the community administrative boundary to extract a space-time flow data set and a built environment data set of each community; S13, the space-time flow data set and the built environment data set of each community are preprocessed to construct a community space-time flow database for each community. 3.The method of claim 2, wherein, In S11, the preprocessing of the collected data comprises land unit division and identification of resident users; The land unit division is to take the community land function data as the basic unit boundary and number. The identifying the permanent user comprises: preliminarily screening the community permanent user according to the time left, that is, adding the time difference between two adjacent positioning points of the same user in a set time interval and located in a specified city into the accumulated time T of the corresponding user; judging whether the accumulated time T of the corresponding user is greater than or equal to the experience threshold value, if the T is greater than or equal to the experience threshold value, the corresponding user is a permanent resident, otherwise, the corresponding user is not a permanent resident. 4.The method of claim 2, wherein, In S13, the data preprocessing comprises: land unit OD flow clustering and OD flow flow calculation; The land unit OD flow clustering is: spatially aggregating individual space-time trajectory data in land units to generate land unit OD flow; The OD flow flow calculation process is: calculating 24h trajectory flow of different land unit OD flows, removing land unit OD flows with 24h trajectory flow sum less than 20, numbering the remaining land unit OD flows j to obtain the trajectory flow of land unit OD flow numbered j in period t within 24 hours; constructing a space-time complex clustering algorithm, clustering individual space-time trajectory data in land unit OD flow combined with travel direction, and calculating OD flow flow of each class. 5.The method of claim 1, wherein, The regression variable index adopts a gradient descent algorithm to solve the goodness of fit, if each goodness of fit index of the structural equation model meets the fitting degree index adaptation standard, the finally obtained structural equation model is taken as the community flow trajectory prediction model: X1=Λ1x+δ1 (1) X2=Λ2x+δ2 (2) η=βx+Γx2+...+ζ (3) Wherein, formula (1) and (2) are measurement models, formula (3) is a structure model, X1 is community outflow trajectory flow, X2 is community inflow trajectory flow, η is a matrix set of dependent variables X1 and X2; x is an independent variable; β is a coefficient matrix of dependent variable; Γ is a coefficient matrix of independent variable; ζ represents residual error.

6. A life circle planning rationality evaluation system based on people flow prediction, which executes the evaluation method of claim 1, characterized by It comprises: A data acquisition module: acquires multi-source data and constructs a community space-time database; A prediction model construction module: constructs a community flow trajectory prediction model based on the community space-time database, and predicts the community inflow trajectory vector set and the outflow trajectory vector set; A life circle simulation module: simulates the community life circle based on the community inflow trajectory vector set and the outflow trajectory vector set; An evaluation module: evaluates the planning rationality of the simulated life circle; A result output module: outputs the evaluation result.

Images