A method for full population synthesis and travel element construction based on multi-data fusion

Abstract

The present invention provides a method for constructing a full population synthesis and travel elements based on multi-data fusion, using population census data, economic census data, resident travel survey data and Internet location data as data sources, using an iterative proportional update model IPU to generate full household information and population information, and updating population travel element attributes according to the statistical results of multi-source data. The present invention comprehensively utilizes multiple data sources to obtain population characteristics of different dimensions and levels, and improves the accuracy and credibility of population generation results; and uses an iterative update algorithm IPU for population synthesis, which can iteratively update sample data according to actual conditions, so that the population synthesis results are closer to the actual situation, with high calculation efficiency, and can better adapt to changes in different population characteristics.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic planning and urban planning, and relates to a method for synthesizing a full population and constructing travel elements based on multi-data fusion. Background Art

[0002] In the field of urban planning and traffic management, the overall size of the population and travel determine the demand and operation status of the traffic system and are the basis for traffic demand analysis. The size of the population and travel factors can be used as the basic input of the traditional four-stage model in the traffic model, as well as the necessary data conditions for the travel chain model and the activity-based model and hybrid model.

[0003] The population synthesis model is a basic method for obtaining the attributes of families and individuals. The population synthesis model integrates different data sources to obtain detailed limited population sample data and full feature data, and fits the limited sample data under the full feature constraints to generate full population data, so that the characteristics of the fitted full population data are consistent with the overall characteristics.

[0004] However, the results of the population synthesis model lack the main attributes of travel factors such as work-residence relationship, family relationship, living status, school location, school distance, work-residence, etc., which in turn affects the judgment of the starting and ending points of steady-state travel activities.

[0005] The current massive amount of traffic operation monitoring data and Internet location data provide a new data source for population synthesis and travel element construction, provide a new opportunity for more comprehensive urban observation, and provide a new perspective for the reliability test of the results of population synthesis and travel element construction, making it possible to obtain the job-residence OD relationship based on traffic zoning.

[0006] For the population synthesis model, this paper proposes a method for full population synthesis and travel element construction based on multi-data fusion. Multi-data fusion technology is used to integrate multiple data sources such as population census data, economic census data, resident travel survey data, and Internet location data, and an iterative proportional update algorithm is used to generate a population set with similar characteristics in a data-driven manner. At the same time, through long-term spatiotemporal location data, school district division and other data resources, individual travel element attributes are mined from the group, and the travel element attributes of the population synthesis results are updated. The use of real overall data and large-sample personal travel information for population synthesis provides the possibility of causal inference for the destination selection of steady-state travel activities. Summary of the invention

[0007] The purpose of the present invention is to overcome the problems in the prior art and provide a method for synthesizing the total population and constructing travel elements based on multi-data fusion, using population census data, economic census data, resident travel survey data and Internet location data as data sources, adopting an iterative proportional update model to generate total family information and population information, and updating the population travel element attributes according to the statistical results of multi-source data. Specifically, the following steps are included: (1) Establishing a basic information database of urban families and population, storing and obtaining population census data, economic census data, resident travel survey data, Internet location data, traffic areas, traffic centers, traffic areas, administrative areas and corresponding spatial scopes. (2) According to the original data, obtaining detailed sample data of individuals and families in different spatial scopes, statistical data, the proportion of non-working population attending university, the OD matrix of work and residence, the distribution matrix of school and traffic areas of different residential communities of different school groups, the average school distance and the probability list of living on campus at different school distances. (3) Using an iterative proportional update model, using family information and population information sample data to fit and generate total family information and population information. (4) Updating the family relationship travel elements in the total personal information according to the age and gender of family members. (5) Update the travel elements of the living status of the entire personal information according to the individual's age, gender, and occupation. (6) Construct the total sample library of the random sampling model without replacement based on the statistical results of the job-residence OD matrix. Use the random sampling model without replacement to obtain and update the transportation community of the place of employment for the working population in the entire personal information. (7) Update the attributes of the transportation community where the school population is located, the distance to school, and the travel element attributes of the information on the residence and school according to the probability distribution matrix of the residential transportation community, the distance distribution matrix, and the probability list of living on campus at different school distances combined with individual attributes. This method provides causal inference for the destination selection of steady-state travel activities of personnel.

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

[0009] A method for synthesizing a full population and constructing travel elements based on multi-data fusion, the method comprising the following steps:

[0010] S1. Establish a basic information database of urban households and population and obtain original data; the original data include: population census data, economic census data, resident travel survey data and Internet location data within the research scope, and original data on the spatial scope and corresponding relationship of transportation areas, transportation centers, transportation communities, and administrative areas;

[0011] S2. Obtain statistical data based on the original data; store the original data and statistical data in the family population basic information database; the data statistics steps are as follows:

[0012] S21. Obtain individual and household statistics in different areas of transportation communities, transportation centers, transportation regions, and administrative districts based on population census data and Internet location data;

[0013] S22. Obtain detailed household information and personal information sample data based on residents’ travel survey data;

[0014] S23. Obtain a list of the proportion of non-working population aged 18 to 30 years old attending university based on census data, INSH;

[0015] S24, determining the job-residence OD matrix A based on the traffic community according to the economic census data and the Internet location data;

[0016] S25. Obtain the distribution probability matrix P of residential transportation communities and school transportation communities of different primary schools, junior high schools, high schools, and universities based on the residents' travel survey data. type , average travel distance matrix D type , the probability distribution dictionary of students living on campus at different school distances LSH type ;

[0017] S3. Expand the sample data; obtain input sample data and constraint data, use the iterative proportional update model to fit the population with the sample data, generate full family information and population information, and store them in the urban family and population basic information database;

[0018] The specific steps are: expand the sample data of family information and personal information to generate full family information and personal information data; obtain family information in the urban population basic information database, use the personal information sample data as the algorithm input sample data, use the population and family statistical data as the constraint data, use the iterative proportional update model to fit the population with the sample data, generate full family information and population information, and store them in the urban family and population basic information database;

[0019] S4. Update the family relationship travel element in the full amount of personal information in the urban family and population basic information database, obtain the full amount of personal information generated in step S3 from the urban family and population basic information database, group it by family, and judge the relationship between husband and wife, intergenerational relationship, and parent-child relationship between family members based on the age of the family members and initialize it;

[0020] S5. Update the travel elements of the full amount of personal information in the urban household and population basic information database, and judge the individual's living status based on the individual's age, gender, and occupation in the full amount of personal information table;

[0021] S6. Update the employment location and transportation community travel elements of the employed population in the full personal information of the urban household and population basic information database. According to the employment-residence OD matrix A in the urban household and population basic information database, build a total sample library of the random sampling model without replacement; add the employment transportation community field to the full personal information table, iteratively update the full personal information, and use the random sampling model without replacement to assign and update the employment location and transportation community of the employed population in the full personal information; the specific method is as follows:

[0022] S61. According to the job-residence OD matrix A, the number of people living in traffic area m and working in traffic area n is a mn Construct a list of job distribution corresponding to the job-residence-traffic community mn :

[0023] t mn ={n,n,n,…,n},T mn Contains a mn n

[0024] S62, merge all corresponding job distribution lists of residents in traffic community m into T m :

[0025] T m ={1,1,1…1,2,2,…,2,…n,n,n,…,n}

[0026] S63. Filter out all employed people, traverse the employed people, obtain the numbers of the residential transportation communities of the employed people, and obtain the corresponding job distribution list T m , for T m Perform sampling without replacement, obtain the employment location transportation district number, and update the employment transportation district field;

[0027] S7. Update the travel elements of the school-going population's full personal information in the urban household and population basic information database, and add the school-going transportation community field, school-going distance field, and school-living situation field to the full personal information table; obtain all the school-going population in the full personal information table, iterate the school-going population, and update the school-going transportation community field, school-going distance field, and school-living situation field for the school-going population.

[0028] Preferably, in step S21, the individual and family statistical data include the total family size, family size with different number of people, total population size, population size with different ages, population size with different genders, population size with different occupations, and population size data with different number of vehicles owned within different traffic areas, traffic centers, traffic communities, and administrative areas;

[0029] In step S22, the family information includes: family number, major traffic area, central traffic area, small traffic area, administrative area, number of family members, number of vehicles owned, and number of employed people in the family;

[0030] Personal information includes: ID number, residential transportation area, residential transportation area, residential transportation area, residential administrative area, family number, member number, age, gender, occupation, income;

[0031] In step S23:

[0032] INSH=[insh 18 ,insh 19 ,insh 20 ,…,insh age ,…,insh 30 ]

[0033] insh 18 is the proportion of college-going population among the non-employed population aged 18, insh age is the proportion of the non-employed population aged age who attend college;

[0034] In step S24, the job-residence OD matrix A is as follows: mn is the number of people living in transportation area m and employed in transportation area n;

[0035]

[0036] In step S25, the traffic zone distribution probability matrix P type , average travel distance matrix D type , the probability distribution dictionary of students living on campus at different school distances LSH type As follows: type is the type of school-going population;

[0037]

[0038] p type,m,n is the probability that a student of school type type who lives in traffic community m attends school in traffic community n;

[0039]

[0040] d type,m,n The minimum travel distance for a student of type type living in transportation zone m and studying in transportation zone n;

[0041] LSH type =[lsh type,0-2 ,lsh type,2-4 ,lsh type,4-6 ,…,lshtype,18-20 ,lsh type,＞20 ]

[0042] lsh type is the probability value of living on campus for different types of students and different distances to school; lsh type,0-2 is the probability value of students with school type type and school distance less than or equal to 2 kilometers living on campus; lsh type,＞20 is the probability value of living on campus for students whose school type is type and whose school distance is greater than 20 kilometers.

[0043] Preferably, the specific method of grouping the personal information data by household and iteratively updating it in step S4 is as follows:

[0044] S41. Add and update the generation code of the individual in the personal information, sort the family members by age from oldest to youngest, and build the intergenerational relationship between family members based on the age difference. The minimum age difference between generations is 20 years old, the oldest generation code is 1, and the generation number increases in descending order of age;

[0045] S42. Add a spouse ID field to the personal information and update it. Construct a husband-wife relationship between family members based on the family member level and family member gender. If the family intergenerational level has only one level or the family intergenerational level is greater than one level and the individual intergenerational level is not the last level, family members at the same level are sorted by age. If two family members with the same generational code and adjacent ages have different genders and the male is older than 22 years old and the female is older than 20 years old, then the husband-wife relationship between the two family members is constructed.

[0046] S43. Add a child ID list field in the personal information and update it. Construct the relationship between parents and children among family members based on the adjacent generational relationship of the family members. If the individual's family generation code is i and there are family members with a family generation code of i+1, then add the IDs of all family members with a generation code of i+1 to the child ID list in the personal information.

[0047] Preferably, in step S5, the life status travel elements include infant, toddler, primary school, middle school, university, working, retired, and unemployed; a life status field is added to the full personal information table, and the full personal information is iteratively updated. The specific method is as follows:

[0048] S51. If the occupation information in the personal information is not unemployed, the life status field is updated to work, and the update is completed. Otherwise, proceed to the next step.

[0049] S52. Determine the living status of personal information based on age and gender, and determine the information as follows:

[0050] A. Those under 3 years old are considered infants;

[0051] B. 3 to 6 years old living in toddler status, living in kindergarten;

[0052] C. Between the ages of 6 and 12, the living conditions are in primary school;

[0053] D. Between the ages of 12 and 18, the living status is in middle school;

[0054] E. From the city family and population basic information database, obtain the university enrollment ratio list INSH of different ages of non-working population from 18 to 30 years old, and obtain the university enrollment ratio insh according to age age , the unemployment rate is 1-insh age , a probability sampling method was used to obtain the status of non-working population as college students or unemployed, and to update their living status;

[0055] F. The living status of personal information is determined to be retired based on gender and statutory retirement age. If a man is over 60, his living status is updated to retired; if a woman is over 55, her living status is determined to be retired.

[0056] Preferably, the specific steps of step S7 are as follows:

[0057] S71) According to the individual school type type, obtain the residential transportation community and school transportation community distribution probability matrix P from the urban family and population basic information database type , average travel distance matrix D type And the probability distribution dictionary LSH of students living on campus at different school distances type ;

[0058] S72) According to the school population living in the traffic area m, obtain the corresponding school traffic area probability distribution list P type,m =[p type,m,0 ,p type,m,1 ,…,p type,m,n ], obtain the traffic zone number of the school through the probability sampling model, and update the traffic zone number of the school;

[0059] S73) According to the living in traffic area m and the school in traffic area n, from D type Get the travel distance d from school type,m,n , closer school distance field;

[0060] S74) According to d type,m,n And the probability distribution dictionary LSH of students living on campus at different school distances type , get the probability value of students living on campus lsh type , the probability of not living on campus is 1-lsh type, a probability sampling method is used to obtain individual schooling status and update the school-enrolled population’s boarding information.

[0061] The present invention provides a method for synthesizing the entire population and constructing travel elements based on multi-data fusion, which uses an iterative proportional update model to generate the entire household information and population information, and updates the population travel element attributes according to the multivariate data statistical results. Compared with the prior art, the present invention has the following advantages:

[0062] (1) Multivariate data fusion for population synthesis and travel element construction can integrate information from different data sources to obtain richer and more comprehensive data. It can also comprehensively utilize multiple data sources to obtain population characteristics at different dimensions and levels, thereby improving the accuracy and credibility of population generation results.

[0063] (2) The iterative update algorithm is used for population synthesis, which can iteratively update the sample data according to the actual situation, making the population synthesis results closer to the actual situation, with high calculation efficiency and better adaptability to changes in different population characteristics.

[0064] (3) Using the spatial location data and correlation relationships of transportation areas, transportation centers, transportation areas, and administrative areas as constraints for population synthesis and travel element construction can ensure the spatial consistency of data and the spatial accuracy of synthetic population data.

[0065] (4) Using data resources such as long-term location data and school district divisions, and adopting probability sampling methods, we can explore individual travel factors from the group, save survey costs and time, ensure that the sample represents the population, reduce selection bias in the travel factor construction process, and make the results more universal and reliable, which better supports the research and analysis of population travel factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. The features and advantages of the present invention will be more clearly understood by referring to the drawings. The drawings are schematic and should not be understood as limiting the present invention in any way. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0067] Figure 1 It is a method for synthesizing the entire population and constructing travel elements based on multi-data fusion. DETAILED DESCRIPTION

[0068] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.

[0069] The steps of the method for synthesizing the entire population and constructing travel elements based on multi-data fusion provided in this embodiment are as follows: Figure 1 As shown, the specific steps are as follows:

[0070] Step 1: Establish a basic information database of urban households and population, obtain population census data, economic census data, resident travel survey data and Internet location data within the research scope, and original data on the spatial scope and corresponding relationships of major transportation areas, central transportation areas, small transportation areas, and administrative areas.

[0071] Step 2: Obtain statistical data based on the original data. Store the original data and statistical data in the family population basic information database. The data statistics steps are as follows:

[0072] (1) Obtain individual and household statistics in different areas of transportation communities, transportation centers, transportation regions, and administrative districts based on census data and Internet location data.

[0073] Individual and family statistics include total family size, family size of different number of people, total population size, population size of different ages, population size of different genders, population size of different occupations, and population size data of different number of vehicles in different transportation areas, transportation centers, transportation communities, and administrative areas. Some sample data are shown in Tables 1 to 4.

[0074] Table 1 Sample statistics of family members

[0075]

[0076]

[0077] Table 2 Sample household vehicle statistics

[0078] Number of households Car-free A car Two cars Three cars 459757 282290 164595 12413 459

[0079] Table 3 Sample of personal age statistics

[0080]

[0081] Table 4 Sample statistics of personal occupation and gender

[0082] Total number of people Unemployed Office Serve retail hotel factory educate male female 1329220 522383 238827 129093 114570 28239 270290 25818 681294 647926

[0083] (2) Obtain detailed household information and personal information sample data based on residents’ travel survey data.

[0084] The family information includes: family number, transportation area, transportation center, transportation community, administrative area, family size, number of vehicles, and number of employed people in the family. The sample data of family information is shown in Table 5.

[0085] Table 5 Family information sample data example

[0086]

[0087]

[0088] Personal information includes: ID number, residential transportation area, residential transportation area, residential transportation area, residential administrative area, family number, member number, age, gender, occupation, income. The sample data of personal information is shown in Table 6.

[0089] Table 6 Sample table of personal information sample data

[0090]

[0091] (3) Based on the census data, we obtain the list of university enrollment rates for non-working population aged 18 to 30 years old, INSH. We count the non-working population and university enrollment population aged 18 to 30 years old, and calculate the university enrollment rates for different ages.

[0092] INSH=[insh 18 ,insh 19 ,insh 20 ,…,insh age ,…,insh 30 ]

[0093] insh 18 is the proportion of college-going population among the non-employed population aged 18, insh age is the proportion of the non-employed population aged age who attend college.

[0094] (4) According to the economic census data and Internet location data, the job-residence OD matrix A based on the traffic community is calculated, a mn is the number of people living in transportation area m and employed in transportation area n.

[0095]

[0096] (5) Based on the resident travel survey data, the distribution probability matrix P of residential transportation areas and school transportation areas for different groups of students in different primary schools, junior high schools, high schools, and universities is obtained. type , average travel distance matrix D type, the probability distribution dictionary of students living on campus at different school distances LSH type . type is the type of school-going population.

[0097]

[0098] p type,m,n is the probability that a person with school type type and living in transportation zone m goes to school in transportation zone n.

[0099]

[0100] d type,m,n The minimum travel distance for a student of school type type who lives in transportation zone m and studies in transportation zone n.

[0101] LSH type =[lsh type,0-2 ,lsh type,2-4 ,lsh type,4-6 ,…,lsh type,18-20 ,lsh type,＞20 ]

[0102] lsh type is the probability value of living on campus for different types of students and different distances to school; lsh type,0-2 is the probability value of students with school type type and school distance less than or equal to 2 kilometers living on campus; lsh type,＞20 is the probability value of living on campus for students whose school type is type and whose school distance is greater than 20 kilometers.

[0103] Step 3: Expand the sample data of family information and personal information to generate full family information and personal information data. Obtain family information from the basic information database of the urban population. The sample data of personal information is used as the sample data for algorithm input. Population and family statistics include the total size of families, the size of families with different numbers of people, the total population size, the size of populations of different ages, the size of populations of different genders, the size of populations of different occupations, and the size of populations with different numbers of vehicles as constraint data. The iterative proportional updating model is used to fit the population with the sample data to generate full family information and population information, which are stored in the basic information database of urban families and population.

[0104] Table 7 shows some samples of the synthesized complete population household information data.

[0105] Table 7 Sample data table of some family information of population synthesis

[0106] Family ID Traffic area Traffic Central District Traffic District Number of people Number of vehicles Family employment population 359741 10 25 451 2 1 0 359742 10 25 451 2 0 0 359743 10 25 451 3 1 2 359744 10 25 451 5 2 2 359745 10 25 451 3 1 2 359746 10 25 451 2 0 1 359747 10 25 451 1 0 0 359748 10 25 451 4 2 2 359749 10 25 451 3 1 1 359750 10 25 451 5 3 3 359751 10 25 451 1 0 1 359752 10 25 451 2 1 1 359753 10 25 451 3 1 2 359754 10 25 451 4 1 2 359755 10 25 451 2 0 2 359756 10 25 451 1 0 0

[0107] An example of the synthesized full amount of personal information data is shown in Table 8.

[0108] Table 8 Sample data table of some personal information of population synthesis

[0109]

[0110] Step 4: Update the family relationship travel elements in the full amount of personal information in the urban family and population basic information database. Obtain the full amount of personal information generated in step 2 from the urban family and population basic information database, group it by family, and use the age of family members as the basis to determine the relationship between husband and wife, intergenerational relationship, and parent-child relationship between family members and initialize it. Group the personal information data by family and iteratively update it; the specific operations are as follows:

[0111] (1) Add and update the intergenerational code of individuals in personal information, sort family members by age from oldest to youngest, and construct the intergenerational relationship between family members based on the age difference. The minimum age difference between generations is 20 years old, and the generation code of the oldest is 1. The generation code increases in descending order of age.

[0112] (2) Add a spouse ID field to the personal information and update it. Construct the husband-wife relationship between family members based on the family member level and gender. If the family intergenerational level has only one level or the family intergenerational level is greater than one level and the individual intergenerational level is not the last level, family members at the same level are sorted by age. If two family members with the same generational code and adjacent ages are of different genders and the male is older than 22 years old and the female is older than 20 years old, then the husband-wife relationship between the two family members is constructed.

[0113] (3) Add a child ID list field in the personal information and update it. Construct the relationship between the parents and children of the family members based on the adjacent generational relationship of the family members. If the individual's family generation code is i and there are family members with a family generation code of i+1, then add the IDs of all family members with a generation code of i+1 to the child ID list in the personal information.

[0114] Step 5: Update the life status travel elements of all personal information in the urban household and population basic information database, and judge the individual life status based on the individual age, gender, and occupation in the full personal information table. The life status travel elements include infant, toddler, elementary school, middle school, university, work, retirement, and unemployed. Add a life status field to the full personal information table and iteratively update the full personal information. The specific method is as follows:

[0115] (1) If the occupation information in the personal information is not unemployed, update the life status field to working and the update is complete. Otherwise, proceed to step (2)

[0116] (2) The living status of personal information is determined based on age and gender, and the determination information is as follows:

[0117] A. Those under 3 years old are considered infants;

[0118] B. 3 to 6 years old living in toddler status, living in kindergarten;

[0119] C. Between the ages of 6 and 12, the living conditions are in primary school;

[0120] D. Between the ages of 12 and 18, the living status is in middle school;

[0121] E. From the city family and population basic information database, obtain the university enrollment ratio list INSH of different ages of non-working population from 18 to 30 years old, and obtain the university enrollment ratio insh according to age age , the unemployment rate is 1-insh age , a probability sampling method is used to obtain the status of non-working population attending university or unemployed, and to update their living status.

[0122] F. The living status of personal information is determined to be retired based on gender and statutory retirement age. If a man is over 60, his living status is updated to retired; if a woman is over 55, her living status is determined to be retired.

[0123] Step 6: Update the employment location and transportation community travel elements of the employed population in the full personal information of the urban household and population basic information database, and build a total sample library of the random sampling model without replacement based on the employment-residence OD matrix A in the urban household and population basic information database. Add the employment transportation community field to the full personal information table, iteratively update the full personal information, and use the random sampling model without replacement to assign and update the employment location and transportation community of the employed population in the full personal information. The specific method is as follows:

[0124] (1) According to the job-residence OD matrix A, the number of people living in transportation community m and working in transportation community n is a mn Construct a list of job distribution corresponding to the job-residence-traffic community mn :

[0125] t mn ={n,n,n,…,n},T mn Contains a mn n

[0126] (2) Combine all corresponding job distribution lists of residents living in traffic community m into T m :

[0127] T m ={1,1,1…1,2,2,…,2,…n,n,n,…,n}

[0128] (3) Filter out all employed people, traverse the employed population, obtain the number of the residential transportation community where the employed population lives, and obtain the corresponding job distribution list T m , for T m Perform sampling without replacement, obtain the transportation district number of the employment location, and update the employment transportation district field.

[0129] Step 7: Update the travel elements of the schooling area and transportation community of the full personal information of the urban household and population basic information database, and add the schooling transportation community field, schooling distance field, and school accommodation field to the full personal information table. Get all the schooling population in the full personal information table, iterate the schooling population, and update the schooling transportation community field, schooling distance field, and school accommodation field of the schooling population. The specific method is as follows:

[0130] (1) According to the individual school type type, obtain the residential transportation community and school transportation community distribution probability matrix P from the urban family and population basic information database type , average travel distance matrix D type And the probability distribution dictionary LSH of students living on campus at different school distances type .

[0131] (2) According to the school population living in the traffic area m, obtain the corresponding school traffic area probability distribution list P type,m =

[0132] [p type,m,0 ,p type,m,1 ,…,p type,m,n ], obtain the traffic zone number of the school through the probability sampling model, and update the traffic zone number of the school.

[0133] (3) Based on the number of people living in traffic area m and studying in traffic area n, type Get the travel distance d from school type,m,n , closer to the school distance field.

[0134] (4) According to d type,m,n And the probability distribution dictionary LSH of students living on campus at different school distances type , get the probability value of students living on campus lsh type , the probability of not living on campus is 1-lsh type , a probability sampling method is used to obtain individual schooling status and update the school-enrolled population’s boarding information.

[0135] After completing the update of all personal information travel elements through steps 4 to 5, some sample data in the full personal information table is shown in Table 9, and the empty table indicates that the data is empty.

[0136] Table 9 Some sample data in the full personal information table

[0137]

[0138] The above description is only a preferred embodiment of the present invention. For example, there may be multiple combinations of different configurations and characteristic parameter settings. The embodiments of the present invention only show example parameters and are not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for synthesizing the entire population and constructing travel elements based on multi-data fusion, characterized in that: The method comprises the following steps: S1. Establish a basic information database of urban households and population and obtain original data; the original data include: population census data, economic census data, resident travel survey data and Internet location data within the research scope, and original data on the spatial scope and corresponding relationship of transportation areas, transportation centers, transportation communities, and administrative areas; S2. Obtain statistical data based on the original data; store the original data and statistical data in the family population basic information database; the data statistics steps are as follows: S21. Obtain individual and household statistics in different areas of transportation communities, transportation centers, transportation regions, and administrative districts based on population census data and Internet location data; S22. Obtain detailed household information and personal information sample data based on residents’ travel survey data; S23. Obtain a list of the proportion of non-working population aged 18 to 30 years old attending university based on census data, INSH; S24, determining the job-residence OD matrix A based on the traffic community according to the economic census data and the Internet location data; S25. Obtain the distribution probability matrix P of residential transportation communities and school transportation communities of different primary schools, junior high schools, high schools, and universities based on the residents' travel survey data. type , average travel distance matrix D type , the probability distribution dictionary of students living on campus at different school distances LSH tvpe ; S3. Expand the sample data; obtain input sample data and constraint data, use the iterative proportional update model to fit the population with the sample data, generate full family information and population information, and store them in the urban family and population basic information database; The specific steps are: expand the sample data of family information and personal information to generate full family information and personal information data; obtain family information in the urban population basic information database, use the personal information sample data as the algorithm input sample data, use the population and family statistical data as the constraint data, use the iterative proportional update model to fit the population with the sample data, generate full family information and population information, and store them in the urban family and population basic information database; S4. Update the family relationship travel element in the full amount of personal information in the urban family and population basic information database, obtain the full amount of personal information generated in step S3 from the urban family and population basic information database, group it by family, and judge the relationship between husband and wife, intergenerational relationship, and parent-child relationship between family members based on the age of the family members and initialize it; S5. Update the travel elements of the full amount of personal information in the urban household and population basic information database, and judge the individual's living status based on the individual's age, gender, and occupation in the full amount of personal information table; S6. Update the employment location and transportation community travel elements of the employed population in the full personal information of the urban household and population basic information database. According to the employment-residence OD matrix A in the urban household and population basic information database, build a total sample library of the random sampling model without replacement; add the employment transportation community field to the full personal information table, iteratively update the full personal information, and use the random sampling model without replacement to assign and update the employment location and transportation community of the employed population in the full personal information; the specific method is as follows: S61. According to the job-residence 0D matrix A, the number of people living in traffic community m and working in traffic community n is a mn Construct a list of job distribution corresponding to the job-residence-traffic community mn : t mn ={n,n,n,…,n},T mn Contains a mn n S62, merge all corresponding job distribution lists of residents in traffic community m into T m : T m ={1,1,1…1,2,2,…,2,…n,n,n,…,n} S63. Filter out all employed people, traverse the employed people, obtain the numbers of the residential transportation communities of the employed people, and obtain the corresponding job distribution list T m , for T m Perform sampling without replacement, obtain the employment location transportation district number, and update the employment transportation district field; S7. Update the travel elements of the school-going population's full personal information in the urban household and population basic information database, and add the school-going transportation community field, school-going distance field, and school-living situation field to the full personal information table; obtain all the school-going population in the full personal information table, iterate the school-going population, and update the school-going transportation community field, school-going distance field, and school-living situation field for the school-going population.

2. According to the method of full population synthesis and travel element construction based on multi-data fusion according to claim 1, it is characterized in that: In step S21, the individual and family statistical data include the total family size, family size with different number of people, total population size, population size with different ages, population size with different genders, population size with different occupations, and population size data with different number of vehicles owned within different traffic areas, traffic centers, traffic communities, and administrative areas; In step S22, the family information includes: family number, major traffic area, central traffic area, small traffic area, administrative area, number of family members, number of vehicles owned, and number of employed people in the family; Personal information includes: ID number, residential transportation area, residential transportation area, residential transportation area, residential administrative area, family number, member number, age, gender, occupation, income; In step S23: INSH=[insh 18 ,insh 19 ,insh 20 ,...,insh age ,...,insh 30 , insh 18 is the proportion of college-going people among the non-employed population aged 18, insh age is the proportion of the non-employed population aged age who attend college; In step S24, the job-residence OD matrix A is as follows: mn is the number of people living in transportation area m and employed in transportation area n; In step S25, the traffic zone distribution probability matrix P type , average travel distance matrix D type , the probability distribution dictionary of students living on campus at different school distances LSH type As follows: type is the type of school-going population; p type，m，n is the probability that a student of type type who lives in transportation community m attends transportation community n; d type，m，n The minimum travel distance for a student of type type living in transportation zone m and studying in transportation zone n; LSH type =[lsh type，0-2 ,lsh type，2-4 ,lsh type，4-6 ,...,lsh type，18-20 ,lsh type，>20 ] lsh type is the probability value of living on campus for different types of students and different distances to school; lsh type，0-2 is the probability value of students with school type type and school distance less than or equal to 2 kilometers living on campus; lsh type，＞20 is the probability value of living on campus for students whose school type is type and whose school distance is greater than 20 kilometers.

3. The method for synthesizing the total population and constructing travel elements based on multi-data fusion according to claim 1 is characterized in that: The specific method of grouping personal information data by household and iteratively updating in step S4 is as follows: S41. Add and update the generation code of the individual in the personal information, sort the family members by age from oldest to youngest, and build the intergenerational relationship between family members based on the age difference. The minimum age difference between generations is 20 years old, the oldest generation code is 1, and the generation number increases in descending order of age; S42. Add a spouse ID field to the personal information and update it. Construct a husband-wife relationship between family members based on the family member level and family member gender. If the family intergenerational level has only one level or the family intergenerational level is greater than one level and the individual intergenerational level is not the last level, family members at the same level are sorted by age. If two family members with the same generational code and adjacent ages have different genders and the male is older than 22 years old and the female is older than 20 years old, then the husband-wife relationship between the two family members is constructed. S43. Add a child ID list field in the personal information and update it. Construct the relationship between parents and children among family members based on the adjacent generational relationship of the family members. If the individual's family generation code is i and there are family members with a family generation code of i+1, then add the IDs of all family members with a generation code of i+1 to the child ID list in the personal information.

4. The method for synthesizing the total population and constructing travel elements based on multi-data fusion according to claim 1 is characterized in that: In step S5, the life status travel elements include infant, toddler, primary school, middle school, university, working, retired, and unemployed; a life status field is added to the full personal information table, and the full personal information is iteratively updated. The specific method is as follows: S51. If the occupation information in the personal information is not unemployed, the life status field is updated to work, and the update is completed. Otherwise, proceed to the next step. S52. Determine the living status of the personal information based on age and gender, and determine the information as follows: A. Those under 3 years old are considered infants; B. 3 to 6 years old living in toddler status, living in kindergarten; C. Between the ages of 6 and 12, the living conditions are in primary school; D. Between the ages of 12 and 18, the living status is in middle school; E. From the city family and population basic information database, obtain the university enrollment ratio list INSH of different ages of non-working population from 18 to 30 years old, and obtain the university enrollment ratio insh according to age age , the unemployment rate is 1-insh age , a probability sampling method was used to obtain the status of non-working population as college students or unemployed, and to update their living status; F. The living status of personal information is determined to be retired based on gender and statutory retirement age. If a man is over 60, his living status is updated to retired; if a woman is over 55, her living status is determined to be retired.

5. The method for synthesizing the total population and constructing travel elements based on multi-data fusion according to claim 1 is characterized in that: The specific steps of step S7 are as follows: S71) According to the individual school type type, obtain the residential transportation community and school transportation community distribution probability matrix P from the urban family and population basic information database type , average travel distance matrix D type And the probability distribution dictionary LSH of students living on campus at different school distances type ; S72) According to the school population living in the traffic area m, obtain the corresponding school traffic area probability distribution list P type，m =[p type，m，0 , p type，m，1 , ..., p type，m，n ], obtain the traffic zone number of the school through the probability sampling model, and update the traffic zone number of the school; S73) According to the living in traffic area m and the school in traffic area n, from D type Get the travel distance d from school type，m，n , closer school distance field; S74) According to d type，m，n And the probability distribution dictionary LSH of students living on campus at different school distances type , get the probability value of students living on campus lsh type , the probability of not living on campus is 1-lsh type , a probability sampling method is used to obtain individual schooling status and update the school-enrolled population’s boarding information.

Images