CN109740292A - A kind of urban population spatial distribution evaluation method and device based on multiple agent - Google Patents
A kind of urban population spatial distribution evaluation method and device based on multiple agent Download PDFInfo
- Publication number
- CN109740292A CN109740292A CN201910095266.1A CN201910095266A CN109740292A CN 109740292 A CN109740292 A CN 109740292A CN 201910095266 A CN201910095266 A CN 201910095266A CN 109740292 A CN109740292 A CN 109740292A
- Authority
- CN
- China
- Prior art keywords
- building
- multiple agent
- attraction
- synthesis
- agent
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to population spatial distribution fields, disclose a kind of urban population spatial distribution evaluation method and device based on multiple agent.Method includes multiple building periphery public service facility environmental factors in the survey region for the influence city multiple agent population spatial distribution for obtaining input;Establish city multiple agent population spatial distribution appraising model;The synthesis attraction of owned building in multiple agent search range is calculated and compared according to model and multiple agent is currently located the synthesis attraction of building, and judge whether the multiple agent is moved to any building in search range, when reaching preset model running termination condition, end of run, export operation result, otherwise, model is returned to continue to calculate;Export result.The present invention establishes urban population spatial distribution appraising model according to the building periphery public service facility environmental factor of selection of influencing to live, can obtain and live in the higher population spatial distribution result of actual conditions degree of conformity.
Description
Technical field
The present invention relates to population spatial distribution field, discloses a kind of urban population spatial distribution based on multiple agent and estimate
Calculate method and device.
Background technique
City is gathering for population, capital and technology, and as the settlement of the mankind, the development and population distribution in city have weight
It contacts.Population spatial distribution refers to regular hour point, the distribution situation of population geographically, be Population process spatially
The form of expression and population geoegraphy research key problem.The planning construction in population spatial distribution and city, economic development
There is inseparable relationship, and solves natural resources shortage and the unbalanced key of regional development.
Being currently being widely used population distribution data is usually to pass through generaI investigation, sampling statistics etc. using administrative division as unit
Mode summarizes the typical population distributed data of acquisition step by step, exist in practical applications it is following three aspect deficiency: first, the time
Resolution ratio is low, and the national census every 10 years primary, and the data update cycle is long, it is difficult to accurate to disclose population change situation;Second,
Spatial resolution is low, be in administration cell using administrative area as the demographic data that unit obtains it is equally distributed, people cannot be embodied
The spatial distribution characteristic of mouth data;Third is unfavorable for multisource data fusion and general space analysis, using administrative area as the system of unit
It counts and there are problems that Space mismatching with physical geographical unit, limit the multidisciplinary friendship based on consensus data
Mutually.
With being constantly progressive for science and technology, especially Earth Information Science is advanced by leaps and bounds in recent years, " from bottom to top "
The modeling approach of multiple agent model start in the research applied to urban population spatial distribution, by between multiple agent,
Interaction between multiple agent and environment studies population space distribution rule from microcosmic angle, solves tradition research side
The problems such as model existing for " from top to bottom " meterological model that method uses is complicated, parameter is more and calculation amount is larger, has more
Independence, intelligence and mobility are more suitable for the research of complex geographical question.
Using multi-agent modeling technology as geographical simulation system (the Geographical Simulation of representative
System new solution) is provided for problems.Multi-agent system is widely used in geography simulation in recent years
In research, the space structure variation power mechanism especially in geography is studied, and is made some progress.In urban geography
In, this method influences the various power mechanisms of urban development, the developing direction of simcity and the population in city to simulate
Spatial distribution and land use/expansion.The concept of intelligent body (Agent) is can be traced earliest to Carl Hewitt's in 1977
" Viewing Control Structures as Patterns of Passing Message " text.In this text, Carl
Hewitt defines the object with self-compatibility, interactivity and concurrent processing mechanism, referred to as " Actor ", which has envelope
The internal storage state closed, and message transmission and feedback can be carried out with other homogeneous objects.The famous computer section of MIT in 1986
One of scholar and the founder of artificial intelligence subject M.Minsky propose the general of intelligent body in " Society of mind "
It reads, and attempts for the concept of social cooperative behaviors to be introduced into computer system.2001, concept of the M.Minsky to Agent
Further instruction is made, he thinks that intelligent body (Agent) is some individuals with special technical ability.The Agent of Britain is theoretical
Expert Wooldridge and Jennings is an autonomous program in proposition Agent in 2009, it can be based on it to environment
Understand, the decision behavior to draw oneself up of having the ability, reaches one or more targets to pursue.Multi-agent system is by a fixed number
The intelligent body of amount shows orderly cooperative motion and behavior by cooperating with each other and self-organizing in collective's level.In system
Each intelligent body has relatively simple function and limited information collection, processing, ability to communicate, however passes through part individual
Between information transmitting and reciprocation after, whole system often shown in group's level efficient cooperative cooperating ability and
High-grade intelligent is horizontal, to realize various arduous, complicated, required precision the is high tasks that single intelligent body cannot be completed.It
It is the fusion of Complex Adaptive Systems, artificial life and distributed artificial intelligence technology, has become a kind of carry out complication system
The way of thinking and tool of analysis and simulation.
But multiple agent model in the prior art consider to influence resident live selection factor when consider is room
The attribute factor of room building itself, and building construction periphery public service facility environmental factor is not considered,
And with the fast development of urban economy, the inhabitation selection of resident is actually more related to the public clothes on building construction periphery
It is engaged in facility situation, therefore the urban population spatial distribution that obtains of prior art research and the actual conditions in resident living meet
Degree is lower, and realistic meaning is not strong.
Summary of the invention
(1) goal of the invention
To overcome above-mentioned at least one defect existing in the prior art, the present invention provides a kind of city based on multiple agent
The evaluation method and device of city's population distribution, the factor that this method is considered when estimating urban population spatial distribution includes resident's
It lives and selects the public service facility factor on the building construction periphery being more related to, with the practical feelings in resident living now
Condition matching degree is higher, and will affect the live building construction attribute factor of selection of resident and be also included in and consider in range, more
Add and tally with the actual situation, there is very strong realistic meaning.
(2) technical solution
As the first aspect of the present invention, the invention discloses a kind of Urban Population Distribution estimation side based on multiple agent
Method, comprising:
It is public to obtain multiple building peripheries in the survey region of the influence city multiple agent population spatial distribution of input
Service facility environmental factor;
The search range of the multiple agent is determined according to multiple income attributes of the multiple agent, and according to described more
Influence and multiple agent of a building periphery public service facility environmental factor to the multiple agent population spatial distribution move
Dynamic rule, establishes city multiple agent population spatial distribution appraising model;
According to owned building in model calculating and more the multiple income attribute multiple agent search range
Comprehensive attraction and the multiple agent are currently located the synthesis attraction of building, and according to the multiple agent movement rule
Judge whether the multiple agent is moved to any building within the scope of described search, when reaching preset model running
When termination condition, end of run exports operation result, otherwise, returns to the model and continues to calculate the multiple agent search model
The synthesis attraction and the multiple agent of enclosing interior owned building are currently located the synthesis attraction of building;
Export the population spatial distribution result after the model running.
In a kind of possible embodiment, multiple income attributes according to the multiple agent determine described more
It further include population, architectural environment and the income of residents data obtained in the survey region before the search range of intelligent body, and
The quantity of income attribute multiple agents multiple in the survey region and spatial distribution are initialized to obtain initial value.
In a kind of possible embodiment, the multiple building periphery public service facility environmental factor is to described more
The influence of intelligent body population spatial distribution passes through to the building and the building periphery public service facility environmental factor
Including transport advantages degree, medical facilities environment, at least one of any in culture and education environment and amusement and recreation environmental factor
Different brackets carry out assignment after calculate and normalized obtain.
In a kind of possible embodiment, own in the calculating the multiple income attribute multiple agent search range
The synthesis attraction of building and the multiple agent are currently located the synthesis attraction of building, comprising:
The building periphery public service facility environment attraction and build that synthesis attraction based on the building includes
The population attraction in object is built, the multiple agent is established and is currently located the comprehensive attraction that building generates the multiple agent
The calculation formula of power;
The multiple agent is currently located the calculation formula for the synthesis attraction that building generates the multiple agent are as follows:
Pi=WiQi, the multiple agent for being A for attribute, Qi=1, the multiple agent for being B for attribute,The multiple agent for being C for attribute,
Wherein, PiThe synthesis attraction that building i generates the multiple agent, W are currently located for multiple agentiFor the periphery building i
The attraction that public service facility generates multiple agent, QiDemographic situation is currently located in building i to institute for multiple agent
The attraction of multiple agent generation is stated, P is the synthesis attraction of owned building in the multiple agent search range, DiTo build
Build object current census's number, EiFor the maximum population that building can accommodate, EiCalculation formula are as follows:S is research area's per-capita housing, XnTo influence the factor of selection of living, xnFor the factor pair
The weight answered, n be influence live selection factor quantity.
In a kind of possible embodiment, the multiple agent movement rule includes being determined using extended pattern mole neighborhood
The search range of the multiple agent of the multiple income attribute, and according to owned building in the multiple agent search range
The size for the synthesis attraction that comprehensive attraction and the multiple agent are currently located building is to judge the multiple agent
No movement.
In a kind of possible embodiment, the model further includes according to influence city multiple agent population spatial distribution
Survey region in the model established of building attribute factor.
As a second aspect of the invention, the invention discloses a kind of, and the urban population spatial distribution based on multiple agent is estimated
Calculate device, comprising:
Acquiring unit multiple is built in the survey region of the influence city multiple agent population spatial distribution for obtaining input
Build object periphery public service facility environmental factor;
Model foundation unit determines the search of the multiple agent for multiple income attributes according to the multiple agent
Range, and according to the multiple building periphery public service facility environmental factor to the multiple agent population spatial distribution
Influence and multiple agent movement rule, establish city multiple agent population spatial distribution appraising model;
Model running unit, for according to model calculating and more the multiple income attribute multiple agent search model
The synthesis attraction and the multiple agent of enclosing interior owned building are currently located the synthesis attraction of building, and according to described
Multiple agent movement rule judges whether the multiple agent is moved to any building within the scope of described search, pre- when reaching
When the model running termination condition first set, end of run exports operation result, otherwise, returns to the model and continues to calculate institute
The synthesis attraction and the multiple agent of stating owned building in multiple agent search range are currently located the synthesis of building
Attraction;
As a result output unit, for exporting the model running after population spatial distribution result.
In a kind of possible embodiment, the model foundation unit includes search subelement, for according to described more
Multiple income attributes of intelligent body determine the search range of the multiple agent;It further include model initialization subelement, for obtaining
Take population, architectural environment and the income of residents data in the survey region, and by income attributes multiple in the survey region
The quantity and spatial distribution of multiple agent are initialized to obtain initial value.
In a kind of possible embodiment, the model foundation unit includes:
First computation subunit, for calculating building and building periphery public service described in simultaneously normalized
In transport advantages degree, medical facilities environment, culture and education environment and the amusement and recreation environmental factor that facilities environment factor includes
At least one of any different grades of assignment, to obtain the multiple building periphery public service facility environmental factor to institute
State the influence of multiple agent population spatial distribution.
In a kind of possible embodiment, how intelligent the multiple income attribute of calculating in the model running unit is
The synthesis attraction of owned building and the multiple agent are currently located the synthesis attraction of building in body search range, packet
It includes:
Second computation subunit, the building periphery public service for including for the synthesis attraction based on the building
Population attraction in facilities environment attraction and building establishes the multiple agent and is currently located building to more intelligence
The calculation formula for the synthesis attraction that energy body generates;
The multiple agent is currently located the calculation formula for the synthesis attraction that building generates the multiple agent are as follows:
Pi=WiQi, the multiple agent for being A for attribute, Qi=1, the multiple agent for being B for attribute,The multiple agent for being C for attribute,
Wherein, PiThe synthesis attraction that building i generates the multiple agent, W are currently located for multiple agentiFor the periphery building i
The attraction that public service facility generates multiple agent, QiDemographic situation is currently located in building i to institute for multiple agent
The attraction of multiple agent generation is stated, P is the synthesis attraction of owned building in the multiple agent search range, DiTo build
Build object current census's number, EiFor the maximum population that building can accommodate, EiCalculation formula are as follows:S is research area's per-capita housing, XnTo influence the factor of selection of living, xnFor the factor pair
The weight answered, n be influence live selection factor quantity.
(3) beneficial effect
A kind of Urban Population Distribution evaluation method and device based on multiple agent provided by the invention has following beneficial
Effect:
Multiple buildings in the survey region that the present invention passes through the influence city multiple agent population spatial distribution for obtaining input
Object periphery public service facility environmental factor determines searching for the multiple agent according to multiple income attributes of the multiple agent
Rope range, and according to the multiple building periphery public service facility environmental factor to the multiple agent population spatial distribution
Influence and multiple agent movement rule, establish city multiple agent population spatial distribution appraising model, estimation urban population
The factor considered when spatial distribution includes that the public service on the building construction periphery that the inhabitation selection of resident is more related to is set
Factor is applied, it is higher with the actual conditions matching degree in resident living now, and will affect the live house of selection of resident and build
It builds object attribute factor and is also included in and consider in range, improve the urban population spatial distribution precision that appraising model obtains, more accord with
Actual conditions are closed, there is very strong realistic meaning.
Detailed description of the invention
It is exemplary below with reference to the embodiment of attached drawing description, it is intended to for the explanation and illustration present invention, and cannot manage
Solution is the limitation to protection scope of the present invention.
Fig. 1 is a kind of Urban Population Distribution evaluation method based on multiple agent of first embodiment provided by the invention
Flow chart;
It is that the survey region obtains that Fig. 2, which is first embodiment provided by the invention with Yongding Road, Haidian District, Beijing City street,
Population spatial distribution result schematic diagram.
Fig. 3 is the preferable implementation flow chart of first embodiment provided by the invention.
Fig. 4 is that a kind of Urban Population Distribution based on multiple agent of second embodiment provided by the invention estimates device
Structural schematic diagram.
Specific embodiment
To keep the purposes, technical schemes and advantages of the invention implemented clearer, with reference to the accompanying drawings and examples to this
Invention is further described in more detail.
It should be understood that described embodiments are some of the embodiments of the present invention, instead of all the embodiments,
In the case where not conflicting, the features in the embodiments and the embodiments of the present application be can be combined with each other.Based on the reality in the present invention
Example is applied, every other embodiment obtained by those of ordinary skill in the art without making creative efforts all belongs to
In the scope of protection of the invention.
Herein, " first ", " second ", attribute " A ", " B " " C " etc. are only used for mutual differentiation, rather than indicate them
Significance level and sequence etc..
The division of unit, subelement herein is only a kind of division of logic function, can be had in actual implementation
Other division modes, such as multiple units and/or subelement can be combined or are integrated in another system.As separation unit
Part illustrate may be physically separated, be also possible to indiscrete.Therefore it can select according to actual needs wherein
Some or all of unit realize the scheme of embodiment.
A kind of Urban Population Distribution estimation side based on multiple agent provided below with reference to Fig. 1 the present invention is described in detail
Method: the present embodiment is mainly used in the estimation of urban population spatial distribution.
As shown, the Urban Population Distribution evaluation method provided in this embodiment based on multiple agent, comprising:
In step s101, multiple in the survey region of the influence city multiple agent population spatial distribution of input build is obtained
Build object periphery public service facility environmental factor;The multiple building periphery public service facility environmental factor includes that traffic is excellent
Gesture degree, medical facilities environment, culture and education environment and amusement and recreation environmental factor;The multiple agent can be handed over environment
Mutually, multiple agent receives the perception in environment by sensor and inputs, and then output action acts on environment, the spy of multiple agent
Property depend on the specific application purpose of its model and establish the intention of model;The inhabitation selection including resident is considered more to be related to
Building construction periphery public service facility factor so that the population spatial distribution result obtained in resident living now
Actual conditions matching degree it is higher, have very strong realistic meaning;
In step s 102, the search model of the multiple agent is determined according to multiple income attributes of the multiple agent
It encloses, and according to the multiple building periphery public service facility environmental factor to the shadow of the multiple agent population spatial distribution
Loud and multiple agent movement rule, establishes city multiple agent population spatial distribution appraising model;Population is as the residence in model
For building in people Agent and survey region there are many-to-one relationship, each building can accommodate multiple intelligent bodies;It is described
Multiple income attributes of multiple agent include low income, middle income and booming income;
In step s 103, according in model calculating and more the multiple income attribute multiple agent search range
The synthesis attraction of owned building and the multiple agent are currently located the synthesis attraction of building, and according to more intelligence
Energy body movement rule judges whether the multiple agent is moved to any building within the scope of described search, sets in advance when reaching
When fixed model running termination condition, end of run exports operation result, otherwise, return the model continue to calculate it is described more
The synthesis attraction of owned building and the multiple agent are currently located the comprehensive of building and attract in intelligent body search range
Power;The movement rule of the multiple agent is based on bounded rationality selection mechanism, and bounded rationality refers in view of at activist's information
Reason ability is limited, because will receive the constraint of emotion, theory, experience and processing capacity, the reason of activist's decision
And " best " is not always pursued according to Rational principle, but pursue " satisfaction " established psychologically, i.e., where intelligent body
The size of the attraction of the attraction and other buildings of building, when in the presence of the attraction for being currently located building greater than intelligent body
When force value, intelligent body random movement is greater than to one to be currently located on the building of building attraction;If it does not exist, then it does not send out
Raw movement, but a random factor can be generated at this time, make intelligent body there is certain probability random movement to search to its attribute is corresponding
On any building within the scope of rope, model terminates to run when can preset intelligent body moving ratio less than 5%.
In step S104, the population spatial distribution result after the model running is exported.
In a kind of possible embodiment, multiple income attributes according to the multiple agent determine described more
It further include population, architectural environment and the income of residents data obtained in the survey region before the search range of intelligent body, and
The quantity of income attribute multiple agents multiple in the survey region and spatial distribution are initialized to obtain initial value;Just
The quantity of intelligent body is set as the size of population in the survey region after beginningization, original demographic's quantity root in each building
The average population quantity on the floor area of building of region and the product of each floor area of building are set according to the study;In the model
The quantitative proportion of the income attribute of multiple agent and each income attribute is according to the income of residents in the yearbook data of the survey region
Data determine;That is, being determined in the multiple agent of the model according to the income of residents data, income attribute belongs to low income
Intelligent body, income attribute belongs to the intelligent body of middle income and income attribute belongs to the respective quantity ratio of intelligent body of booming income
Example;By in the survey region population and building quantity initialization, the two data are exactly set as default value, i.e., initially
Value, prepares for model running.
In a kind of possible embodiment, the multiple building periphery public service facility environmental factor is to described more
The influence of intelligent body population spatial distribution passes through to the building and the building periphery public service facility environmental factor
Including transport advantages degree, medical facilities environment, at least one of any in culture and education environment and amusement and recreation environmental factor
Different brackets carry out assignment after calculate and normalized obtain;The present embodiment is chosen Yongding Road, Haidian District, Beijing City street and is made
For survey region, building construction newness degree has typicalness, and includes in geographical national conditions monitoring building construction data
Multilayer, short and single family house building, and the public service facility on street inside and periphery is more sound;Acquired data packet
It includes: geographical national conditions monitoring building construction data, Haidian District, Beijing City yearbook income of residents data, survey region in 2017 in 2017
Remote sensing image datas in 2017 and Beijing POI point vector data;The transport advantages degree can be according to building construction
It is no subway station 1000m, bus stop 300m coverage area in delimited, if the building construction simultaneously in subway station
With bus station in coverage area, then dominance is set as 1;If only being set in coverage area in subway station
Dominance is 0.7;If only setting dominance as 0.5 in coverage area in bus stop;Neither in subway station nor
In bus station in coverage area, then dominance is set as 0.3;The medical facilities environment, culture and education environment and stop
Not busy entertainment environment can be according to difference of hospital, school and the park in POI point data layer at a distance from building construction
Grade carries out assignment, and the different brackets medical facilities environment includes health station, large hospital, different grades of culture and education ring
Border includes kindergarten, primary school, junior middle school, senior middle school, and different grades of amusement and recreation environment includes park plaza, shopping mall market;
In addition, geographical national conditions monitor the not effective residential building of building construction collected, so need to be monitored using geographical national conditions
Cities and towns comprehensive function unit figure layer and night lights remotely-sensed data in data extract industrial and mining enterprises therein, Unit courtyard and
The building construction figure spot etc. that long-term lamplight brightness is 0, to obtain effective residential building data.
In a kind of possible embodiment, own in the calculating the multiple income attribute multiple agent search range
The synthesis attraction of building and the multiple agent are currently located the synthesis attraction of building, comprising:
The building periphery public service facility environment attraction and build that synthesis attraction based on the building includes
The population attraction in object is built, the multiple agent is established and is currently located the comprehensive attraction that building generates the multiple agent
The calculation formula of power;
The multiple agent is currently located the calculation formula for the synthesis attraction that building generates the multiple agent are as follows:
Pi=WiQi, the multiple agent for being A for attribute, Qi=1, the multiple agent for being B for attribute,The multiple agent for being C for attribute,
Wherein, PiThe synthesis attraction that building i generates the multiple agent, W are currently located for multiple agentiFor the periphery building i
The attraction that public service facility generates multiple agent, QiDemographic situation is currently located in building i to institute for multiple agent
The attraction of multiple agent generation is stated, P is the synthesis attraction of owned building in the multiple agent search range, DiTo build
Build object current census's number, EiFor the maximum population that building can accommodate, EiCalculation formula are as follows:S is research area's per-capita housing, XnTo influence the factor of selection of living, xnFor the factor pair
The weight answered, n be influence live selection factor quantity.
In a kind of possible embodiment, the multiple agent movement rule includes being determined using extended pattern mole neighborhood
The search range of the multiple agent of the multiple income attribute, and according to owned building in the multiple agent search range
The size for the synthesis attraction that comprehensive attraction and the multiple agent are currently located building is to judge the multiple agent
No movement;In standard CA model, mole neighborhood is made of 8 grey cellulars adjacent around the cellular of center, if normalized molar is adjacent
The extension radius in domain is 1, and extended pattern mole neighborhood is then started with center cellular, is extended the extension that for example radius is 2 outward
There can be 5 × 5 structure in type mole neighborhood, that is, there are 25 " cellular ", the search model of the present embodiment middle and low income intelligent body
It encloses and can be set to one 3 × 3 region of search, the region of search that the search range of middle income intelligent body is one 5 × 5, height is received
Enter the region of search that the search range of intelligent body is one 7 × 7, the intelligent body of difference income attribute perceives in its search range
The attraction of periphery residential building is to determine whether be moved on other buildings.
In a kind of possible embodiment, the model further includes according to influence city multiple agent population spatial distribution
Survey region in the model established of building attribute factor;The building attribute factor includes building type, builds
Object newness degree is built, about building construction type, according to the 5th class building construction area in geographical ten major class of national conditions database
Ground outdoor scene photo and acquisition instruction in data, building type is classified, and according to the superiority and inferiority assignment of grade, is thus obtained
Building type;It will affect the live building construction attribute factor of selection of resident to be also included in and consider in range, improve estimation
The urban population spatial distribution precision that model obtains, is more in line with actual conditions.
Reference table 1, table 1 are the preferable building construction type declaration tables of first embodiment provided by the invention.
- 1 building construction type declaration table of table
Note: above-mentioned field can be not limited in specific implementation.
About building construction newness degree, according to on-site inspection, survey region building construction is according to " house is damaged etc.
Grade evaluation criteria " it is broadly divided into intact room, substantially intact room and a small number of general damage rooms, different grades of building construction is in height
Resolution remote sense image shows different colors, therefore the corresponding number of plies of binding region building construction, regional architecture
Feature and investigation, and residential building is divided by remote sensing image using supervised classification method: intact room, substantially intact and general
House is damaged, and assignment 1,0.75 and 0.5 respectively;It is public using very poor standardization in order to eliminate the dimension impact of each impact factor
Each impact factor is normalized in formula, so that the influence value of each impact factor is distributed between [0,1], than less
With impact factor to the importance of population spatial distribution.For positive impact factor, i.e., more bigger the more excellent type factor, standardization are public
Formula are as follows:
For negative sense impact factor, i.e., smaller more excellent type, such as medical facilities environment, culture and education environment, amusement and recreation ring
Border etc. standardizes formula are as follows:
X in formulamaxRepresent the maximum value of certain a kind of impact factor, XminRepresent the minimum value of such impact factor, XiIt represents
The value of such impact factor of current building, X represent the standardized value of such impact factor.
The evaluation of the effect on environment factor, then since the difference of income leads to that different knots can be generated between intelligent body individual
Fruit.Intelligent body is due to itself economic or actual demand difference, to show the preference different to impact factor.Different Effects
The weighing factor of factor pair difference income intelligent body can be determined using ranking method.
In a kind of possible embodiment, the population spatial distribution result is provided to responsible city or city office
The management system of the department of the work such as traffic, security, municipal administration is shown in portion region with carrying out mapable, and the management system can
To carry out the integrated dispatch of traffic, security, municipal administration etc. resource, or progress and people based on the population spatial distribution result
The relevant urban planning of mouth spatial distribution;Fig. 2 is first embodiment provided by the invention with Yongding Road, Haidian District, Beijing City street
The population spatial distribution result schematic diagram obtained for the survey region.
Below with reference to the preferable implementation flow chart for the first embodiment that Fig. 3 the present invention is described in detail provides, comprising:
S11 obtains multiple building periphery public service facility environmental factors of input;
S12 establishes city multiple agent population spatial distribution appraising model;
The synthesis of owned building in S13, moving model, calculating and more multiple income attribute multiple agent search ranges
Attraction P and multiple agent are currently located the synthesis attraction P of buildingi, judge whether there is and be greater than comprehensive attraction PiIt is comprehensive
Attraction P is closed, is to execute S14, otherwise executes S15;
S14, multiple agent move;
S15, multiple agent do not move;
S16, mobile multiple agent quantity accounts for the ratio of sum and judges whether to reach pre- after computation model operation is primary
First setting value is to execute S17, otherwise executes S12;
S17, model running terminate, and export result.
A kind of urban population spatial distribution estimation based on multiple agent provided below with reference to Fig. 4 the present invention is described in detail
Device second embodiment: the present embodiment is mainly used in the estimation of urban population spatial distribution.
As shown, the urban population spatial distribution provided in this embodiment based on multiple agent estimates device, comprising:
Acquiring unit 201 is more in the survey region for influencing city multiple agent population spatial distribution inputted for obtaining
A building periphery public service facility environmental factor;The multiple building periphery public service facility environmental factor includes handing over
Logical dominance, medical facilities environment, culture and education environment and amusement and recreation environmental factor;The multiple agent can with environment into
Row interaction, multiple agent receive the perception in environment by sensor and input, and then output action acts on environment, multiple agent
Characteristic depend on the specific application purpose of its model and establish the intention of model;
Model foundation unit 202, for determining the multiple agent according to multiple income attributes of the multiple agent
Search range, and according to the multiple building periphery public service facility environmental factor to multiple agent population space point
The influence of cloth and multiple agent movement rule establish city multiple agent population spatial distribution appraising model;Population is as model
In resident Agent and survey region in building there are many-to-one relationship, each building can accommodate multiple intelligence
Body;Multiple income attributes of the multiple agent include low income, middle income and booming income;Consider the inhabitation selection including resident
The public service facility factor on the building construction periphery being more related to so that obtain population spatial distribution result with now
Actual conditions matching degree in resident living is higher, has very strong realistic meaning;
Model running unit 203, for according to the model calculate and more the multiple income attribute multiple agent search
The synthesis attraction of owned building and the multiple agent are currently located the synthesis attraction of building within the scope of rope, and according to
The multiple agent movement rule judges whether the multiple agent is moved to any building within the scope of described search, when reaching
When to preset model running termination condition, otherwise end of run, output operation result returns to the model and continues to count
It calculates the synthesis attraction of owned building and the multiple agent in the multiple agent search range and is currently located building
Comprehensive attraction;The movement rule of the multiple agent is based on bounded rationality selection mechanism, and bounded rationality refers in view of activity
Person's information processing capability is limited, because will receive the constraint of emotion, theory, experience and processing capacity, activist determines
The reason of plan, simultaneously not always pursues " best " according to Rational principle, but pursues " satisfaction " established psychologically, that is, compares intelligence
The size of the attraction of the attraction and other buildings of building where energy body, is currently located building greater than intelligent body when existing
When the attraction force value of object, intelligent body random movement is greater than to one to be currently located on the building of building attraction;If not depositing
It is not moving then, but a random factor can be generated at this time, intelligent body is made to have certain probability random movement to its attribute
On any building in corresponding search range, model terminates to transport when can preset intelligent body moving ratio less than 5%
Row.
As a result output unit 204, for exporting the model running after population spatial distribution result.
In a kind of possible embodiment, the model foundation unit includes search subelement, for according to described more
Multiple income attributes of intelligent body determine the search range of the multiple agent;It further include model initialization subelement, for obtaining
Take population, architectural environment and the income of residents data in the survey region, and by income attributes multiple in the survey region
The quantity and spatial distribution of multiple agent are initialized to obtain initial value;The quantity of intelligent body is set as described after initialization
The size of population in survey region, original demographic's quantity in each building is according to being averaged on the survey region floor area of building
The size of population and the product of each floor area of building are set;The income attribute of multiple agent and each income belong in the model
Property quantitative proportion according in the yearbook data of the survey region income of residents data determine;By the people in the survey region
The two data are exactly set as default value, i.e. initial value, prepared for model running by mouth and the initialization of building quantity.
In a kind of possible embodiment, the model foundation unit includes:
First computation subunit, for calculating building and building periphery public service described in simultaneously normalized
In transport advantages degree, medical facilities environment, culture and education environment and the amusement and recreation environmental factor that facilities environment factor includes
At least one of any different grades of assignment, to obtain the multiple building periphery public service facility environmental factor to institute
State the influence of multiple agent population spatial distribution;The present embodiment chooses Yongding Road, Haidian District, Beijing City street as survey region,
Its building construction newness degree has typicalness, and include multilayer in geographical national conditions monitoring building construction data, it is short and
Single family house building, and the public service facility on street inside and periphery is more sound;Acquired data include: ground in 2017
It manages national conditions and monitors building construction data, Haidian District, Beijing City yearbook income of residents data in 2017, survey region remote sensing in 2017
Image data and Beijing POI point vector data;Whether the transport advantages degree can be according to building construction in subway station
1000m, bus stop 300m coverage area in delimited, if the building construction is simultaneously in subway station and bus station
In coverage area, then dominance is set as 1;If only in subway station in coverage area, set dominance as
0.7;If only setting dominance as 0.5 in coverage area in bus stop;Neither in subway station nor in bus station
In coverage area, then set dominance as 0.3;The medical facilities environment, culture and education environment and amusement and recreation ring
Border can be carried out according to different brackets of hospital, school and the park in POI point data layer at a distance from building construction
Assignment, the different brackets medical facilities environment include health station, large hospital, and different grades of culture and education environment includes children
Youngster garden, primary school, junior middle school, senior middle school, different grades of amusement and recreation environment include park plaza, shopping mall market;In addition, geographical
National conditions monitor the not effective residential building of building construction collected, so the city in geographical national conditions monitoring data need to be utilized
Town comprehensive function unit figure layer and night lights remotely-sensed data extract industrial and mining enterprises, Unit courtyard and long-term light therein
The building construction figure spot etc. that brightness is 0, to obtain effective residential building data.
In a kind of possible embodiment, how intelligent the multiple income attribute of calculating in the model running unit is
The synthesis attraction of owned building and the multiple agent are currently located the synthesis attraction of building in body search range, packet
It includes:
Second computation subunit, the building periphery public service for including for the synthesis attraction based on the building
Population attraction in facilities environment attraction and building establishes the multiple agent and is currently located building to more intelligence
The calculation formula for the synthesis attraction that energy body generates;
The multiple agent is currently located the calculation formula for the synthesis attraction that building generates the multiple agent are as follows:
Pi=WiQi, the multiple agent for being A for attribute, Qi=1, the multiple agent for being B for attribute,The multiple agent for being C for attribute,
Wherein, PiThe synthesis attraction that building i generates the multiple agent, W are currently located for multiple agentiFor the periphery building i
The attraction that public service facility generates multiple agent, QiDemographic situation is currently located in building i to institute for multiple agent
The attraction of multiple agent generation is stated, P is the synthesis attraction of owned building in the multiple agent search range, DiTo build
Build object current census's number, EiFor the maximum population that building can accommodate, EiCalculation formula are as follows:S is research area's per-capita housing, XnTo influence the factor of selection of living, xnFor the factor pair
The weight answered, n be influence live selection factor quantity.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by those familiar with the art, all answers
It is included within the scope of the present invention.Therefore, protection scope of the present invention should be with the scope of protection of the claims
It is quasi-.
Claims (10)
1. a kind of urban population spatial distribution evaluation method based on multiple agent characterized by comprising
Obtain multiple building peripheries public service in the survey region of the influence city multiple agent population spatial distribution of input
Facilities environment factor;
The search range of the multiple agent is determined according to multiple income attributes of the multiple agent, and is built according to the multiple
Build the mobile rule of influence and multiple agent of the object periphery public service facility environmental factor to the multiple agent population spatial distribution
Then, city multiple agent population spatial distribution appraising model is established;
According to the synthesis of owned building in model calculating and more the multiple income attribute multiple agent search range
Attraction and the multiple agent are currently located the synthesis attraction of building, and are judged according to the multiple agent movement rule
Whether the multiple agent is moved to any building within the scope of described search, terminates when reaching preset model running
When condition, otherwise end of run, output operation result returns to the model and continues to calculate in the multiple agent search range
The synthesis attraction of owned building and the multiple agent are currently located the synthesis attraction of building;
Export the population spatial distribution result after the model running.
2. the method according to claim 1, wherein multiple income attributes according to the multiple agent
It further include population, architectural environment and the resident obtained in the survey region before the search range for determining the multiple agent
Income data, and the quantity of income attribute multiple agents multiple in the survey region and spatial distribution are initialized to obtain
Obtain initial value.
3. method according to claim 1 or 2, which is characterized in that the multiple building periphery public service facility ring
Influence of the border factor to the multiple agent population spatial distribution passes through to the building and the public clothes in building periphery
In transport advantages degree, medical facilities environment, culture and education environment and the amusement and recreation environmental factor that business facilities environment factor includes
At least one of any different brackets carry out assignment after calculate and normalized obtains.
4. method according to claim 1 or 2, which is characterized in that described to calculate the multiple income attribute multiple agent
The synthesis attraction of owned building and the multiple agent are currently located the synthesis attraction of building in search range, packet
It includes:
The building periphery public service facility environment attraction and building that synthesis attraction based on the building includes
Interior population attraction establishes the multiple agent and is currently located synthesis attraction that building generates the multiple agent
Calculation formula;
The multiple agent is currently located the calculation formula for the synthesis attraction that building generates the multiple agent are as follows: Pi=
WiQi, the multiple agent for being A for attribute, Qi=1, the multiple agent for being B for attribute,
The multiple agent for being C for attribute, Wherein, PiIt is current for multiple agent
The synthesis attraction that place building i generates the multiple agent, WiIt is building i periphery public service facility to mostly intelligent
The attraction that body generates, QiThe attraction that demographic situation generates the multiple agent in building i is currently located for multiple agent
Power, P are the synthesis attraction of owned building in the multiple agent search range, DiFor building current census's number, EiTo build
Build the maximum population that object can accommodate, EiCalculation formula are as follows:S is that research area lives face per capita
Product, XnTo influence the factor of selection of living, xnFor the corresponding weight of the factor, n be influence to live selection factor quantity.
5. according to the method described in claim 4, it is characterized in that, the multiple agent movement rule includes being rubbed using extended pattern
Your neighborhood determines the search range of the multiple agent of the multiple income attribute, and according to institute in the multiple agent search range
There is the synthesis attraction of building and the size of synthesis attraction that the multiple agent is currently located building is described to judge
Whether multiple agent moves.
6. according to the method for any of claims 1 or 2, which is characterized in that the model further includes more according to city is influenced
The model that building attribute factor in the survey region of intelligent body population spatial distribution is established.
7. a kind of urban population spatial distribution based on multiple agent estimates device, which is characterized in that including
Acquiring unit, for obtaining multiple buildings in the survey region for influencing city multiple agent population spatial distribution inputted
Periphery public service facility environmental factor;
Model foundation unit determines the search model of the multiple agent for multiple income attributes according to the multiple agent
It encloses, and according to the multiple building periphery public service facility environmental factor to the shadow of the multiple agent population spatial distribution
Loud and multiple agent movement rule, establishes city multiple agent population spatial distribution appraising model;
Model running unit, for calculating all in the multiple income attribute multiple agent search range build according to the model
The synthesis attraction and the multiple agent of building object are currently located the synthesis attraction of building, and are moved according to the multiple agent
Dynamic rule judges whether the multiple agent is moved to any building within the scope of described search, when reaching preset mould
Type run abort condition when, end of run, output operation result otherwise return to the model and continue to calculate the multiple agent
The synthesis attraction of owned building and the multiple agent are currently located the synthesis attraction of building in search range;
As a result output unit, for exporting the model running after population spatial distribution result.
8. device according to claim 7, which is characterized in that the model foundation unit includes search subelement, is used for
The search range of the multiple agent is determined according to multiple income attributes of the multiple agent;It further include that model initialization is single
Member, for obtaining the population in the survey region, architectural environment and income of residents data, and will be multiple in the survey region
The quantity and spatial distribution for taking in attribute multiple agent are initialized to obtain initial value.
9. device according to claim 7 or 8, which is characterized in that the model foundation unit includes:
First computation subunit, for calculating building and building periphery public service facility described in simultaneously normalized
It is any in transport advantages degree, medical facilities environment, culture and education environment and the amusement and recreation environmental factor that environmental factor includes
The different grades of assignment of at least one, to obtain the multiple building periphery public service facility environmental factor to described more
The influence of intelligent body population spatial distribution.
10. device according to claim 9, which is characterized in that the multiple receipts of calculating in the model running unit
Enter the synthesis attraction of owned building and the multiple agent in attribute multiple agent search range and is currently located building
Comprehensive attraction, comprising:
Second computing unit, the building periphery public service facility ring for including for the synthesis attraction based on the building
Population attraction in border attraction and building, establishes the multiple agent and is currently located building and produce to the multiple agent
The calculation formula of raw synthesis attraction;
The multiple agent is currently located the calculation formula for the synthesis attraction that building generates the multiple agent are as follows: Pi=
WiQi, the multiple agent for being A for attribute, Qi=1, the multiple agent for being B for attribute,It is right
In attribute be C multiple agent, Wherein, PiFor the current institute of multiple agent
In the synthesis attraction that building i generates the multiple agent, WiIt is the periphery building i public service facility to multiple agent
The attraction of generation, QiIt is currently located the attraction that demographic situation generates the multiple agent in building i for multiple agent,
P is the synthesis attraction of owned building in the multiple agent search range, DiFor building current census's number, EiFor building
The maximum population that object can accommodate, EiCalculation formula are as follows:S is that research area lives face per capita
Product, XnTo influence the factor of selection of living, xnFor the corresponding weight of the factor, n be influence to live selection factor quantity.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910095266.1A CN109740292A (en) | 2019-01-30 | 2019-01-30 | A kind of urban population spatial distribution evaluation method and device based on multiple agent |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910095266.1A CN109740292A (en) | 2019-01-30 | 2019-01-30 | A kind of urban population spatial distribution evaluation method and device based on multiple agent |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109740292A true CN109740292A (en) | 2019-05-10 |
Family
ID=66366950
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910095266.1A Pending CN109740292A (en) | 2019-01-30 | 2019-01-30 | A kind of urban population spatial distribution evaluation method and device based on multiple agent |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109740292A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110309953A (en) * | 2019-05-28 | 2019-10-08 | 特斯联(北京)科技有限公司 | Using the city safety monitoring layout system and method for object mobility forecast of distribution |
CN111986215A (en) * | 2020-08-31 | 2020-11-24 | 四川省地震局减灾救助研究所 | Population calculation method and system based on high-precision building data |
CN117272787A (en) * | 2023-09-04 | 2023-12-22 | 中山大学 | Residence land growth simulation method and system based on multiple intelligent agents |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102129623A (en) * | 2011-03-15 | 2011-07-20 | 宝鸡四维衡器有限公司 | Method and system for monitoring and managing municipal refuse transport vehicle |
CN102750411A (en) * | 2012-06-19 | 2012-10-24 | 中国地质大学(武汉) | Urban dynamic micro-simulation method based on multi-agent discrete choice model |
US20140065115A1 (en) * | 2011-01-28 | 2014-03-06 | Probiotical S.P.A. | Effervescent composition in solid form for use in vaginal applications for the treatment of vaginal infections |
CN104765924A (en) * | 2015-04-13 | 2015-07-08 | 清华大学 | Urban traffic emergency evacuation simulation system and control method |
CN104933286A (en) * | 2015-03-13 | 2015-09-23 | 华南理工大学 | City spatial quality evaluation method based on big data |
CN106455058A (en) * | 2016-12-02 | 2017-02-22 | 中国联合网络通信集团有限公司 | Method and device for determining population distribution condition |
CN107516168A (en) * | 2017-08-28 | 2017-12-26 | 中国测绘科学研究院 | A kind of Synthetic Assessment of Eco-environment Quality method |
CN107801203A (en) * | 2017-11-13 | 2018-03-13 | 毛国强 | The evaluation method and its system of the density of population and mobility based on multi-data fusion |
-
2019
- 2019-01-30 CN CN201910095266.1A patent/CN109740292A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140065115A1 (en) * | 2011-01-28 | 2014-03-06 | Probiotical S.P.A. | Effervescent composition in solid form for use in vaginal applications for the treatment of vaginal infections |
CN102129623A (en) * | 2011-03-15 | 2011-07-20 | 宝鸡四维衡器有限公司 | Method and system for monitoring and managing municipal refuse transport vehicle |
CN102750411A (en) * | 2012-06-19 | 2012-10-24 | 中国地质大学(武汉) | Urban dynamic micro-simulation method based on multi-agent discrete choice model |
CN104933286A (en) * | 2015-03-13 | 2015-09-23 | 华南理工大学 | City spatial quality evaluation method based on big data |
CN104765924A (en) * | 2015-04-13 | 2015-07-08 | 清华大学 | Urban traffic emergency evacuation simulation system and control method |
CN106455058A (en) * | 2016-12-02 | 2017-02-22 | 中国联合网络通信集团有限公司 | Method and device for determining population distribution condition |
CN107516168A (en) * | 2017-08-28 | 2017-12-26 | 中国测绘科学研究院 | A kind of Synthetic Assessment of Eco-environment Quality method |
CN107801203A (en) * | 2017-11-13 | 2018-03-13 | 毛国强 | The evaluation method and its system of the density of population and mobility based on multi-data fusion |
Non-Patent Citations (2)
Title |
---|
LAWRENCE A. BROWN ET AL.: "The Intra-Urban Migration Process: A Perspective", 《GEOGRAFISKA ANNALER. SERIES B, HUMAN GEOGRAPHY》 * |
卓莉等: "基于多智能体模型与建筑物信息的高空间分辨率人口分布模拟", 《地理研究》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110309953A (en) * | 2019-05-28 | 2019-10-08 | 特斯联(北京)科技有限公司 | Using the city safety monitoring layout system and method for object mobility forecast of distribution |
CN111986215A (en) * | 2020-08-31 | 2020-11-24 | 四川省地震局减灾救助研究所 | Population calculation method and system based on high-precision building data |
CN117272787A (en) * | 2023-09-04 | 2023-12-22 | 中山大学 | Residence land growth simulation method and system based on multiple intelligent agents |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhou et al. | Understanding crowd behaviors in a social event by passive wifi sensing and data mining | |
Agnew | Space and place | |
CN104867357B (en) | Multiple no-manned plane scheduling and mission planning method towards Emergency Response to Earthquake | |
CN109740292A (en) | A kind of urban population spatial distribution evaluation method and device based on multiple agent | |
CN109583698A (en) | A kind of Urban Streets spatial shape comprehensive estimation method based on fuzzy mathematics theory | |
He et al. | What is the developmental level of outlying expansion patches? A study of 275 Chinese cities using geographical big data | |
Zhou et al. | Digital manufacturing and urban conservation based on the Internet of Things and 5 G technology in the context of economic growth | |
CN112949997A (en) | System and method for community portrayal in urban planning design | |
CN109214065A (en) | The crowd evacuation emulation method and system of Q table are shared based on multi-Agent | |
Xiao et al. | Does the culture service supply of green spaces match the demand of residents in a new district? A perspective from China. | |
CN111784830B (en) | Rule-based three-dimensional geographic information model space analysis method and system | |
Zhou et al. | Village-town system in suburban areas based on cellphone signaling mining and network hierarchy structure analysis | |
Mansoor | Conceptualizing and realizing a smart city model for Bangladesh | |
Dong et al. | 5G virtual reality in the design and dissemination of contemporary urban image system under the background of big data | |
Abdullahi et al. | Urban compactness assessment | |
Chen | Strategy of artificial intelligence assisted health tourism in the perspective of global region based on Markov chain model | |
Zhou et al. | Population mobility network and spatial accessibility based on peer-to-peer interactive energy management | |
Shao | Research on architectural planning and landscape design of smart city based on computational intelligence | |
Xie et al. | A modeling framework for individual-based urban mobility based on data fusion | |
CN117079124B (en) | Urban and rural landscape image quantification and promotion method based on community differentiation | |
Leonidis et al. | Interaction in Smart Cities | |
Gan et al. | Infrastructure and Industry Economy | |
ZHANG | Research on Vitality and Sustainable Development of Urban Villages from the Urban Perspective | |
Pang et al. | Community environmental detection and community outdoor space health promotion based on cloud resource scheduling mechanism | |
Zhang et al. | Smart Urban and Rural Spatial Planning Methods Based on the Era of Big Data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190510 |
|
RJ01 | Rejection of invention patent application after publication |