CN116756695B - Urban function collaborative optimization method integrating geographic features and flow features - Google Patents
Urban function collaborative optimization method integrating geographic features and flow features Download PDFInfo
- Publication number
- CN116756695B CN116756695B CN202310776586.XA CN202310776586A CN116756695B CN 116756695 B CN116756695 B CN 116756695B CN 202310776586 A CN202310776586 A CN 202310776586A CN 116756695 B CN116756695 B CN 116756695B
- Authority
- CN
- China
- Prior art keywords
- space
- features
- function
- flow
- city
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000005457 optimization Methods 0.000 title claims abstract description 52
- 230000006870 function Effects 0.000 claims abstract description 174
- 238000009826 distribution Methods 0.000 claims abstract description 11
- 230000002776 aggregation Effects 0.000 claims description 14
- 238000004220 aggregation Methods 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 11
- 230000003993 interaction Effects 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000005311 autocorrelation function Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 5
- 230000000737 periodic effect Effects 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 230000004931 aggregating effect Effects 0.000 claims description 2
- 230000000694 effects Effects 0.000 description 7
- 238000012549 training Methods 0.000 description 7
- 238000012360 testing method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000008447 perception Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- -1 biology Substances 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000007261 regionalization Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000002195 synergetic effect Effects 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
- G06F18/256—Fusion techniques of classification results, e.g. of results related to same input data of results relating to different input data, e.g. multimodal recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/29—Graphical models, e.g. Bayesian networks
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a city function collaborative optimization method integrating geographic features and flow features, which comprises the following steps: obtaining the geospatial features and the stream space features of the target city, and constructing a directed weighted graph according to the geospatial features and the stream space features; inputting the directional weighted graph into a pre-trained graph representation learning model to obtain city group characteristic information; inputting city group characteristic information into a pre-trained function identification model, and performing function classification on each geographic unit of a target city to obtain function type information of each geographic unit; and solving a preset objective function according to the information of each function type to obtain the optimal configuration information of the function distribution pattern of the target city. According to the method, the function type information is obtained through the geographic space features and the stream space features, and then the function type information is subjected to collaborative optimization, so that the urban function high-dynamic features can be rapidly extracted, and the urban function is adapted to actual use.
Description
Technical Field
The invention relates to the technical field of urban function area division, in particular to a urban function collaborative optimization method integrating geographic features and flow features.
Background
Global urbanization and urban regionalization have become prominent features and spatial aspects of economic development in the world today. With the advancement of the urban process, urban economy in the form of oversized urban areas in many countries and regions worldwide has generated tremendous scale benefits. The ultra-large urban area promotes the integration of regional economic and social traffic, and simultaneously, the social and economic elements such as capital, information, technology, labor force and the like overflow the traditional space-time limit, so that the problems of resource mismatch, competition homogeneity and difficult coupling are generated.
Traditional urban functions refer to the types of uses that can be identified based on the intensity of usage of people's common activities within a single city. However, urban functional research based on urban internal structuring and organizing survey data fails to consider the environmental background of urban production and living demands, and fails to provide powerful support for urban area cooperation. On the other hand, in the context of rapid urbanization, the functional utilization types of urban units are changing drastically, and the highly dynamic nature of urban functions is ignored for a long time. The city function layout and the actual use of the crowd have certain difference, and the problem of the uncomfortableness of the city function and the actual use is generated. The traditional urban function identification based on statistical investigation or remote sensing images is difficult to rapidly reveal urban function high dynamic characteristics, and the urban function adaptation problem is difficult to solve.
Accordingly, the prior art has drawbacks and needs to be improved and developed.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a city function collaborative optimization method integrating geographic features and flow features aiming at the defects in the prior art, and the method aims to solve the problems that the city functions are not adapted to actual use due to the fact that the high-dynamic characteristics of the city functions are difficult to extract rapidly in the region dividing method in the prior art.
The technical scheme adopted for solving the technical problems is as follows:
A city function collaborative optimization method integrating geographic features and flow features comprises the following steps:
obtaining the geospatial features and the flow space features of a target city, and constructing a directed weighted graph according to the geospatial features and the flow space features;
inputting the directional weighted graph into a pre-trained graph representation learning model to obtain city group characteristic information;
Inputting the city group characteristic information into a pre-trained function identification model, and performing function classification on each geographic unit of the target city to obtain function type information of each geographic unit;
And solving a preset objective function according to each function type information to obtain the optimal configuration information of the function distribution pattern of the target city.
Optionally, the step of obtaining the geospatial feature includes:
Acquiring landscape element information of a target city, and extracting physical characteristics from the landscape element information;
acquiring social and economic index information in a geographic unit in a target city, and extracting social features from the social and economic index information;
And fusing the physical characteristics and the social characteristics to obtain the geospatial characteristics.
Optionally, the step of acquiring the stream space feature includes:
acquiring multi-source flow data information of a target city, and extracting a space time sequence stay track from the multi-source flow data information;
Obtaining the flow distance and the flow density of the space-time flow according to the space time sequence stay track, and clustering the space-time flow to obtain a space aggregation mode of the space-time flow on the flow space, wherein the space aggregation mode comprises the following steps: a gather mode, a cluster mode, an equal length mode, and a propagation mode;
Acquiring the residence start time and the residence end time of the space-time flow in the space time sequence residence track, and calculating the residence position of the space-time flow based on a preset time sequence autocorrelation function;
clustering the space-time stream according to the stay position to obtain a space-time stream mode, wherein the space-time stream mode comprises the following steps: periodic mode, companion mode, and frequent mode;
and obtaining flow space characteristics according to the space time sequence stay track, the space aggregation mode and the space-time mode.
Optionally, constructing a directed weighted graph according to the geospatial feature and the stream space feature includes:
obtaining a node set according to geographic units connected with all space-time flows in the target city and the geographic space characteristics;
Obtaining an edge set according to the flow space characteristics of all the space-time flows in the target city;
calculating unit correlation between each node in the node set and each edge in the edge set, wherein the unit correlation comprises: a cell adjacency relationship, a cell distance relationship, and a flow feature interaction relationship between cells;
And constructing and obtaining a directed weighted graph according to the unit correlation.
Optionally, the graph represents that the learning model employs a graph roll-up neural network framework as the encoder;
Inputting the directional weighted graph into a pre-trained graph representation learning model to obtain city group characteristic information, wherein the method comprises the following steps:
inputting the directional weighted graph into a pre-trained graph representation learning model, and mapping each node of the directional weighted graph into a preset vector space to obtain characteristic information of each node;
Aggregating the characteristic information of the neighbor nodes in the directed weighted graph to a central node, and updating the central node;
after all nodes in the directed weighted graph are updated, graph representation codes of each node are obtained;
and obtaining a reconstruction adjacent matrix according to the graph representation codes of the nodes, and obtaining city group characteristic information according to the reconstruction adjacent matrix.
Optionally, the function type information includes: residential, industrial, commercial, educational, entertainment, and hybrid functions.
Optionally, the preset objective function includes a function population coverage maximization objective, a function reachability maximization objective, and a function type entropy value maximization objective.
The invention also provides a city function collaborative optimization device integrating geographic features and flow features, which comprises:
the construction module is used for acquiring the geospatial features and the stream space features of the target city and constructing a directed weighted graph according to the geospatial features and the stream space features;
the first input module is used for inputting the directional weighted graph into a pre-trained graph representation learning model to obtain city group characteristic information;
the second input module is used for inputting the city group characteristic information into a pre-trained function identification model, and performing function classification on each geographic unit of the target city to obtain function type information of each geographic unit;
and the solving module is used for solving a preset objective function according to each function type information to obtain the optimal configuration information of the function distribution pattern of the target city.
The invention also provides a terminal, comprising: the city function collaborative optimization method comprises the steps of a memory, a processor and a city function collaborative optimization program which is stored in the memory and can run on the processor and is used for fusing geographic features and flow features, wherein the city function collaborative optimization program fusing the geographic features and the flow features is executed by the processor.
The present invention also provides a computer readable storage medium storing a computer program executable for implementing the steps of the urban function collaborative optimization method of merging geographic and flow features as described above.
The invention provides a city function collaborative optimization method integrating geographic features and flow features, which comprises the following steps: obtaining the geospatial features and the flow space features of a target city, and constructing a directed weighted graph according to the geospatial features and the flow space features; inputting the directional weighted graph into a pre-trained graph representation learning model to obtain city group characteristic information; inputting the city group characteristic information into a pre-trained function identification model, and performing function classification on each geographic unit of the target city to obtain function type information of each geographic unit; and solving a preset objective function according to each function type information to obtain the optimal configuration information of the function distribution pattern of the target city. According to the invention, a directional weighted graph is constructed through the geospatial features and the flow space features, and then the graph representation learning model and the function recognition model are sequentially utilized to obtain the function type information, so that the function type information is subjected to collaborative optimization, and the urban function high-dynamic features can be rapidly extracted, so that the urban function is adapted to actual use.
Drawings
FIG. 1 is a flow chart of a preferred embodiment of the urban function collaborative optimization method of the present invention that merges geographic and flow features.
FIG. 2 is a functional block diagram of a preferred embodiment of the urban function co-optimization device of the present invention incorporating geographic features and flow features.
Fig. 3 is a functional block diagram of a preferred embodiment of the terminal of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more clear and clear, the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The invention utilizes the multi-source city perception big data to construct intelligent extraction and unified expression of city geographic-flow double space feature elements; the geographic space features and the stream space features are fused, a city feature multi-scale fusion representation learning model based on a graph embedded frame is developed, and finally, a city group function collaborative optimization method integrating a graph neural network and a geographic-stream double view angle is provided. The method breaks through a series of problems that the traditional method cannot rapidly reveal the high dynamic characteristics of the urban functions, is difficult to effectively evaluate the adaptation among the functions in the city, realizes the dynamic updating of the city according to the evaluation result and the like, and can realize the collaborative optimization of the urban functions under the support of big data and artificial intelligence.
Referring to fig. 1, fig. 1 is a flowchart of a city function collaborative optimization method that merges geographic and flow features according to the present invention. As shown in fig. 1, the city function collaborative optimization method integrating geographic features and flow features according to the embodiment of the invention includes the following steps:
And step S100, obtaining the geospatial features and the stream space features of the target city, and constructing a directed weighted graph according to the geospatial features and the stream space features.
The invention provides a brand-new urban function collaborative optimization method considering urban function characteristics and actual use demands of people, so as to realize optimization and management of urban function layout based on data driving. The geospatial feature W is a spatial feature affecting the functional structure of the city, and the stream spatial feature Y is a spatial feature affecting the functional structure of the city group.
In one embodiment, the step of obtaining the geospatial feature comprises: acquiring landscape element information of a target city, and extracting physical characteristics from the landscape element information; acquiring social and economic index information in a geographic unit in a target city, and extracting social features from the social and economic index information; and fusing the physical characteristics and the social characteristics to obtain the geospatial characteristics.
Specifically, the physical feature set W 1＝{w1,w2,…,w6 is extracted from the landscape elements of the target city Q, which are individual components constituting the landscape, including topography, climate, water, biology, soil and social culture factors, specifically including: building (w 1), road network (w 2), public facilities (w 3), water (w 4), green land (w 5), wherein w i of physical characteristics comprises the following attribute densitiesArea/>Quantity/>Length/>Height/>Shape/>Etc.
The method for extracting the geospatial social feature W 2＝{w7,w8,…,w10 from the socioeconomic index in the geographic unit specifically comprises the following steps: legal units (w 7), population density (w 8), house price index (w 9), POI type (w 10).
Wherein, the landscape elements and the physical characteristics can be added or changed according to the actual city distinction.
The invention fuses the physical feature W 1 and the social feature W 2 to construct the geospatial feature W= { W 1,W2 }. And further obtain spatial features affecting the urban functional structure.
In one embodiment, the step of obtaining the stream space feature comprises: acquiring multi-source flow data information of a target city, and extracting a space time sequence stay track from the multi-source flow data information; obtaining the flow distance and the flow density of the space-time flow according to the space time sequence stay track, and clustering the space-time flow to obtain a space aggregation mode of the space-time flow on the flow space, wherein the space aggregation mode comprises the following steps: a gather mode, a cluster mode, an equal length mode, and a propagation mode; acquiring the residence start time and the residence end time of the space-time flow in the space time sequence residence track, and calculating the residence position of the space-time flow based on a preset time sequence autocorrelation function; clustering the space-time stream according to the stay position to obtain a space-time stream mode, wherein the space-time stream mode comprises the following steps: periodic mode, companion mode, and frequent mode; and obtaining flow space characteristics according to the space time sequence stay track, the space aggregation mode and the space-time mode.
Specifically, the time-series dwell trajectory Y 1 is extracted from multi-source stream data including a people stream data source, a cell phone signaling data source, a geographic stream data source, and the like. Specifically, each dwell position (y location) in the time-series dwell trajectory is defined by spatial coordinates (longitudeDimension/>) Residence time/>End of residence time/>Etc.
Based on the space-time stream clustering method and the space-time interaction network construction method, the clustering of the streams is realized by redefining indexes such as the distance (d ij) and the density (lambda) of the streams, and then the aggregation mode Y 2 on the stream space is identified. The method comprises the following specific steps: the method comprises the steps of clustering the streams by using a density clustering method for traditional point set data, such as DBSCAN, OPTICS and the like, by taking the distance d ij of the streams and the density lambda of the streams as the characteristics of space-time streams, and finally identifying different modes aggregated on the stream space, wherein the modes comprise: a gather mode (y 1), a cluster mode (y 2), an equal length mode (y 3), and a propagation mode (y 4). The calculation formula of the flow distance d ij is as follows:
Wherein alpha is a preset first weight parameter, beta is a preset second weight parameter, O point is an outflow point, and D point is a point of a stream mesh; the alpha and beta are introduced to distinguish the effect of the distance between the O-point and D-point convection.
If the density of the flow in a certain area A is lambda, the calculation formula is as follows:
Where d i,1 represents the first-order proximity of the space-time stream F i, i.e., the distance between F i and the nearest stream thereto, and n is the number of space-time streams within region A.
The spatial features of the spatial streams are used to identify the spatial stream pattern Y 3. The method comprises the following specific steps: the residence position y location of the stream, residence start timeResidence stop time/>Regarding the three-dimensional characteristics of the space-time flow, constructing a time sequence-based autocorrelation function, wherein the formula of the autocorrelation function is as follows:
Where κ is the flow residence position and y location is the preset time residence.
Using kappa as a time sequence stay track characteristic, and using a clustering method such as DBSCAN, OPTICS and the like to identify different modes of the space-time stream, wherein the method comprises the following steps: periodic pattern (y 5), companion pattern (y 6), frequent pattern (y 7).
Finally, from the space-time sequence stay track Y 1, a space aggregation mode Y 2 and a space-time flow mode Y 3 are identified, and the specific relation is as follows: y 2∈Y1;Y3∈Y1;Y2＝{y1,y2,y3,y4};Y3 = { Y5, Y6, Y7}.
In one embodiment, constructing a directed weighted graph from the geospatial feature and the flow spatial feature includes: obtaining a node set according to geographic units connected with all space-time flows in the target city and the geographic space characteristics; obtaining an edge set according to the flow space characteristics of all the space-time flows in the target city; calculating unit correlation between each node in the node set and each edge in the edge set, wherein the unit correlation comprises: a cell adjacency relationship, a cell distance relationship, and a flow feature interaction relationship between cells; and constructing and obtaining a directed weighted graph according to the unit correlation.
Specifically, unified expression T of urban mass multi-element space-time geographic-stream double-space data can be realized based on the geographic space characteristics and the stream space characteristics. Firstly, constructing group activity space-time networks with different time-space scales, abstracting the group activity space-time networks into a directed weighted graph G, wherein geographic units connected with all space-time flows and geographic space features thereof are abstracted into a node set V, each grid is defined as a single node V epsilon V, and the feature x v is a multidimensional geographic space feature vector of the geographic unit plot; the whole space-time stream and the stream space characteristics thereof are abstracted into an edge set U, wherein each edge is defined as a single edge U epsilon U, the unit correlation of the edge characteristics of the connection node v and U is defined as x uv, and finally, the unified expression of the urban group multi-element space-time geographic-stream double-space data is realized.
The unit correlation is measured in three ways: cell adjacency, cell distance, flow feature interaction between cells.
The formula for the cell adjacency is as follows:
Wherein the method comprises the steps of Is the weight of node v i to node v j.
The formula of the cell distance relationship is as follows:
where d i-j is the spatial distance from node v i to node v j.
The formula of the flow characteristic interaction relationship between the units is as follows:
in addition, different modes of the stream space may be added or changed according to actual data.
As shown in fig. 1, the city function collaborative optimization method integrating the geographic feature and the flow feature further includes:
Step 200, inputting the directional weighted graph into a pre-trained graph representation learning model to obtain city group characteristic information.
In one embodiment, the graph representation learning model employs a graph roll-up neural network framework as the encoder. The step S200 specifically includes:
Step S210, inputting the directional weighted graph into a pre-trained graph representation learning model, and mapping each node of the directional weighted graph into a preset vector space to obtain characteristic information of each node;
step S220, feature information of neighbor nodes in the directed weighted graph is aggregated to a center node, and the center node is updated;
Step S230, after updating all nodes in the directed weighted graph is completed, graph representation codes of each node are obtained;
and step 240, obtaining a reconstructed adjacent matrix according to the graph representation codes of all the nodes, and obtaining city group characteristic information according to the reconstructed adjacent matrix.
Specifically, when the training chart represents the learning model, the processing of step S100 is performed on the sample city, and a directional weighted chart for the training chart represents the learning model is obtained. In particular, the goal of the geographic cell feature representation learning is to construct a mapMapping each node v of the graph into a dense low-dimensional vector space/>While preserving topology and node information of the network. Multidimensional feature representation learning using a fully unsupervised graph representation learning model employs a graph convolutional neural network (GCN) framework as an encoder.
After the directed weighted graph input diagram represents the learning model, the features of the neighbor nodes are aggregated to the central node, and then the central node is updated. The update node formula is as follows:
Where M l+1 denotes message passing, U l+1 denotes a status update operation, ne [ v ] denotes a set of its neighbor nodes, l denotes the first layer of graph convolution, Representing features of node v of layer I,/>Representing the representation features of the layer i node u, x uv represents the cell correlation.
During training, a proper loss function and an optimizer are required to be selected, through multi-layer stacking, loss value change in the training process is recorded, the training is stopped until the layer number reaches the expected depth, the graph representation code of each node is obtained, and the formula of the reconstructed adjacency matrix is as follows:
wherein Z represents the final new node matrix output representation after all nodes are updated, Representing the adjacency matrix after reconstruction, σ represents the Sigmoid function, and Z T represents the matrix transpose.
The formula of the loss function is: where a represents the adjacency matrix before reconstruction.
As shown in fig. 1, the city function collaborative optimization method integrating the geographic feature and the flow feature further includes:
And step S300, inputting the city group characteristic information into a pre-trained function recognition model, and performing function classification on each geographic unit of the target city to obtain function type information of each geographic unit.
In one embodiment, the function type information includes: residential, industrial, commercial, educational, entertainment, and hybrid functions. The method classifies the regional characteristics in the urban mass characteristic information, performs statistical analysis on each class, and realizes real-time monitoring of the urban mass.
Specifically, the node state is converted into a tag type by a local output function. The output node class labels P are set to 6 classes, namely, a residential function (P1), an industrial function (P2), a business function (P3), an educational function (P4), an entertainment function (P5), and a hybrid function (P6), respectively. All nodes T v are divided into training set D train, validation set D verify, and test set D test. By verifying the accuracy of classification of the node function labels P on the verification set D verify, the effect of the model is tested on the test set D test after training is completed. The local output function is o v＝g(hv,xv), where hv represents the encoding state, xv represents the node characteristics, and o v represents the class label. The category label includes: residential, industrial, commercial, educational, entertainment, and hybrid functions. That is, the city group feature information is input into the function identification model, each node feature in the city group feature information is extracted, the coding state of each node feature is identified, and then the function type information corresponding to each node feature is obtained, that is, the function type information of each region is obtained. In addition, the invention also adopts a graph space-time network frame (GSTN) in the function recognition model to establish the mapping relation between nodes and edge characteristics in different space-time scale active networks, and establishes a final function recognition model which can adapt to space-time dynamic changes by optimizing and correcting parameters of the function recognition model so as to realize the space-time dynamic monitoring of urban functions.
The expected depth of the functional identification model can be changed and adjusted according to actual data so as to achieve the optimal effect of feature classification.
As shown in fig. 1, the city function collaborative optimization method integrating the geographic feature and the flow feature further includes:
And step S400, solving a preset objective function according to each function type information to obtain the optimal configuration information of the function distribution pattern of the target city.
The invention designs the urban group function cooperative optimization strategy based on the urban group composite system function cooperative degree index difference distribution, and realizes the function optimization of the real-time urban group.
In one embodiment, the preset objective functions include a functional crowd coverage maximization objective, a functional reachability maximization objective, and a functional type entropy value maximization objective.
Specifically, the elements refer to abstract representations of geographic spatial features and stream spatial features in numerical values among regions and within regions, and different geographic region elements represent differences in features among regions. The synergy refers to efficient interconnection and optimal configuration of elements among areas and in the areas, the degree of synergy F i is an index for describing the synergistic effect of certain elements, and when the degree of synergy is larger, the better the synergy among the elements of the areas is. The formula is as follows: f i＝α*Xi; wherein X i is a numerical index of the element, the value ranges of the elements with different functions in the same area are different, and alpha is a weight value of the element.
The optimization strategy consists of three objective constituents: (1) functional crowd coverage is maximized. Defining the coverage of the functional crowd of the geographic unit i as s i, and the coverage of all units asWhere M is the number of geographic units and x i is the number of geographic units that indicate whether the geographic units cover the population. (2) functional reachability is maximized. Define the reachability of geographic element i as A i (L, t), the sum of reachability as(3) Function type entropy maximization,/>N is the number of function types. The formula of the objective function is as follows:
Wherein, α 1 is a preset first target weight, α 2 is a preset second target weight, and α 3 is a preset third target weight.
And counting the degree of synergy F of the plots P of the different geographic units, and considering that the plots P i need to be optimized when F is lower than a degree of synergy threshold F i. Modeling the optimization problem as a P-media problem, fusing the development hybrid heuristic algorithms such as self-adaptive large-scale neighborhood search, tabu search, genetic optimization and the like, rapidly solving the model, and optimally configuring the urban group function distribution pattern in a space-time dynamic environment. The synergy threshold is calculated by combining actual data through a probability model, so that an accurate threshold result is obtained.
The invention provides a city function collaborative optimization method integrating a graph neural network and a geographic-flow double view angle, which utilizes multi-source city perception big data, integrates geographic space characteristics and flow space characteristics to evaluate the high dynamic attribute of the current city function and the actual use function of crowd, and finally establishes a city function optimization scheme for a city. The space-time dynamic change of the urban function is considered, and the urban function collaborative optimization method integrating geographic-flow double views is provided based on a graph neural network framework. The invention achieves the following effects:
Firstly, the invention embeds and combines the multisource city perception data and the multi-level fusion map, realizes the characteristic expression of the complex city group system which can not be solved by the traditional method, and further provides a foundation for constructing the collaborative governance and optimization of the city group area.
Secondly, the invention combines the traditional application requirement with the novel city group function view angle, overcomes the defect that the traditional method cannot meet the requirement of city function real-time monitoring and dynamic planning, and provides professional insight for government regulation and optimization of resource allocation at regional scale.
Thirdly, innovation is realized on the city optimization formula, a regional coordination formula is defined, and the scalability of the inter-regional coordination degree is realized; and a land function optimization function is defined, so that the guidance of city function optimization is ensured.
Further, as shown in fig. 2, the present invention further provides a city function collaborative optimization device based on the above-mentioned city function collaborative optimization method integrating geographic features and flow features, which includes:
A construction module 100, configured to acquire a geospatial feature and a flow space feature of a target city, and construct a directed weighted graph according to the geospatial feature and the flow space feature;
The first input module 200 is configured to input the directional weighted graph into a pre-trained graph representation learning model to obtain city group feature information;
The second input module 300 is configured to input the city group feature information into a pre-trained function recognition model, and perform function classification on each region of the target city to obtain function type information of each region;
and the solving module 400 is configured to solve a preset objective function according to the function type information of each region, so as to obtain the optimal configuration information of the function distribution pattern of the target city.
Further, as shown in fig. 3, the present invention further provides a terminal based on the above-mentioned city function collaborative optimization method integrating geographic features and flow features, which includes: the system comprises a memory 20, a processor 10 and a city function co-optimization program 30 which is stored in the memory 20 and can run on the processor 10 and is used for fusing geographic features and flow features, wherein the city function co-optimization program 30 fusing geographic features and flow features realizes the steps of the city function co-optimization method fusing geographic features and flow features.
The present invention also provides a computer readable storage medium storing a computer program executable for implementing the steps of the urban function collaborative optimization method of merging geographic and flow features as described above.
In summary, the urban function collaborative optimization method integrating geographic features and flow features provided by the invention comprises the following steps: obtaining the geospatial features and the flow space features of a target city, and constructing a directed weighted graph according to the geospatial features and the flow space features; inputting the directional weighted graph into a pre-trained graph representation learning model to obtain city group characteristic information; inputting the city group characteristic information into a pre-trained function identification model, and performing function classification on each geographic unit of the target city to obtain function type information of each geographic unit; and solving a preset objective function according to each function type information to obtain the optimal configuration information of the function distribution pattern of the target city. According to the invention, a directional weighted graph is constructed through the geospatial features and the flow space features, and then the graph representation learning model and the function recognition model are sequentially utilized to obtain the function type information, so that the function type information is subjected to collaborative optimization, and the urban function high-dynamic features can be rapidly extracted, so that the urban function is adapted to actual use.
It is to be understood that the invention is not limited in its application to the examples described above, but is capable of modification and variation in light of the above teachings by those skilled in the art, and that all such modifications and variations are intended to be included within the scope of the appended claims.
Claims (9)
1. A city function collaborative optimization method integrating geographic features and flow features is characterized by comprising the following steps:
obtaining the geospatial features and the flow space features of a target city, and constructing a directed weighted graph according to the geospatial features and the flow space features;
inputting the directional weighted graph into a pre-trained graph representation learning model to obtain city group characteristic information;
Inputting the city group characteristic information into a pre-trained function identification model, and performing function classification on each geographic unit of the target city to obtain function type information of each geographic unit;
solving a preset objective function according to each function type information to obtain optimal configuration information of the function distribution pattern of the target city;
The step of obtaining the stream space feature includes:
acquiring multi-source flow data information of a target city, and extracting a space time sequence stay track from the multi-source flow data information;
Obtaining the flow distance and the flow density of the space-time flow according to the space time sequence stay track, and clustering the space-time flow to obtain a space aggregation mode of the space-time flow on the flow space, wherein the space aggregation mode comprises the following steps: a gather mode, a cluster mode, an equal length mode, and a propagation mode;
Acquiring the residence start time and the residence end time of the space-time flow in the space time sequence residence track, and calculating the residence position of the space-time flow based on a preset time sequence autocorrelation function;
clustering the space-time stream according to the stay position to obtain a space-time stream mode, wherein the space-time stream mode comprises the following steps: periodic mode, companion mode, and frequent mode;
and obtaining flow space characteristics according to the space time sequence stay track, the space aggregation mode and the space-time mode.
2. The method for collaborative optimization of urban functionality that merges geographic and flow features according to claim 1, wherein the step of obtaining geospatial features comprises:
Acquiring landscape element information of a target city, and extracting physical characteristics from the landscape element information;
acquiring social and economic index information in a geographic unit in a target city, and extracting social features from the social and economic index information;
And fusing the physical characteristics and the social characteristics to obtain the geospatial characteristics.
3. The method of collaborative optimization of urban functionality that merges geographic and flow features according to claim 1, wherein constructing a directed weighted graph from the geospatial features and the flow spatial features comprises:
obtaining a node set according to geographic units connected with all space-time flows in the target city and the geographic space characteristics;
Obtaining an edge set according to the flow space characteristics of all the space-time flows in the target city;
calculating unit correlation between each node in the node set and each edge in the edge set, wherein the unit correlation comprises: a cell adjacency relationship, a cell distance relationship, and a flow feature interaction relationship between cells;
Constructing and obtaining a directed weighted graph according to the unit correlation;
The formula of the flow characteristic interaction relationship between the units is as follows:
wherein, For the weight of node v i to node v j, d i-j is the spatial distance of node v i to node v j.
4. The urban function collaborative optimization method fusing geographic and flow features according to claim 1, wherein the graph representation learning model employs a graph roll-up neural network framework as an encoder;
Inputting the directional weighted graph into a pre-trained graph representation learning model to obtain city group characteristic information, wherein the method comprises the following steps:
inputting the directional weighted graph into a pre-trained graph representation learning model, and mapping each node of the directional weighted graph into a preset vector space to obtain characteristic information of each node;
Aggregating the characteristic information of the neighbor nodes in the directed weighted graph to a central node, and updating the central node;
after all nodes in the directed weighted graph are updated, graph representation codes of each node are obtained;
and obtaining a reconstruction adjacent matrix according to the graph representation codes of the nodes, and obtaining city group characteristic information according to the reconstruction adjacent matrix.
5. The urban function collaborative optimization method fusing geographical and flow features according to claim 1, wherein the function type information comprises: residential, industrial, commercial, educational, entertainment, and hybrid functions.
6. The urban function collaborative optimization method fusing geographic and flow features according to claim 1, wherein the preset objective functions include a function crowd coverage maximization objective, a function reachability maximization objective, and a function type entropy maximization objective.
7. A city function collaborative optimization device integrating geographic features and flow features, comprising:
the construction module is used for acquiring the geospatial features and the stream space features of the target city and constructing a directed weighted graph according to the geospatial features and the stream space features;
the first input module is used for inputting the directional weighted graph into a pre-trained graph representation learning model to obtain city group characteristic information;
the second input module is used for inputting the city group characteristic information into a pre-trained function identification model, and performing function classification on each geographic unit of the target city to obtain function type information of each geographic unit;
The solving module is used for solving a preset objective function according to each function type information to obtain the optimal configuration information of the function distribution pattern of the target city;
The step of obtaining the stream space feature includes:
acquiring multi-source flow data information of a target city, and extracting a space time sequence stay track from the multi-source flow data information;
Obtaining the flow distance and the flow density of the space-time flow according to the space time sequence stay track, and clustering the space-time flow to obtain a space aggregation mode of the space-time flow on the flow space, wherein the space aggregation mode comprises the following steps: a gather mode, a cluster mode, an equal length mode, and a propagation mode;
Acquiring the residence start time and the residence end time of the space-time flow in the space time sequence residence track, and calculating the residence position of the space-time flow based on a preset time sequence autocorrelation function;
clustering the space-time stream according to the stay position to obtain a space-time stream mode, wherein the space-time stream mode comprises the following steps: periodic mode, companion mode, and frequent mode;
and obtaining flow space characteristics according to the space time sequence stay track, the space aggregation mode and the space-time mode.
8. A terminal, comprising: the method comprises the steps of a memory, a processor and a city function collaborative optimization program which is stored in the memory and can run on the processor and is used for fusing geographic features and flow features, wherein the city function collaborative optimization program fusing geographic features and flow features is executed by the processor and is used for realizing the city function collaborative optimization method fusing geographic features and flow features according to any one of claims 1-6.
9. A computer readable storage medium, characterized in that it stores a computer program executable for implementing the steps of the urban function collaborative optimization method of merging geographical and flow features according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310776586.XA CN116756695B (en) | 2023-06-27 | 2023-06-27 | Urban function collaborative optimization method integrating geographic features and flow features |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310776586.XA CN116756695B (en) | 2023-06-27 | 2023-06-27 | Urban function collaborative optimization method integrating geographic features and flow features |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116756695A CN116756695A (en) | 2023-09-15 |
| CN116756695B true CN116756695B (en) | 2024-05-31 |
Family
ID=87949480
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310776586.XA Active CN116756695B (en) | 2023-06-27 | 2023-06-27 | Urban function collaborative optimization method integrating geographic features and flow features |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116756695B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116954156B (en) * | 2023-09-19 | 2023-12-15 | 中科航迈数控软件(深圳)有限公司 | Numerical control processing process route planning method, device, equipment and medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104992260A (en) * | 2015-05-22 | 2015-10-21 | 东南大学 | Urban land function suitability planning method based on multi-dimensional guiding |
| CN109189917A (en) * | 2018-06-27 | 2019-01-11 | 华南师范大学 | A kind of city function limited region dividing method and system merging landscape and social characteristic |
| CN110377807A (en) * | 2019-06-03 | 2019-10-25 | 华南农业大学 | Group of cities functional cohesion and Analysis of Spatial Distribution Pattern method and system |
| CN112559909A (en) * | 2020-12-18 | 2021-03-26 | 浙江工业大学 | Business area discovery method based on GCN embedded spatial clustering model |
| CN113077370A (en) * | 2021-03-31 | 2021-07-06 | 北京嘀嘀无限科技发展有限公司 | City function attribute identification method, equipment, medium and product of geographic tile |
| CN113158084A (en) * | 2021-03-25 | 2021-07-23 | 马琦伟 | Method and device for processing movement track data, computer equipment and storage medium |
| CN113806419A (en) * | 2021-08-26 | 2021-12-17 | 西北大学 | Urban area function identification model and method based on space-time big data |
| CN114626512A (en) * | 2022-05-17 | 2022-06-14 | 南京信息工程大学 | High-temperature disaster forecasting method based on directed graph neural network |
| CN115545758A (en) * | 2022-09-26 | 2022-12-30 | 苏州大学 | Method and system for self-adaptive incremental site selection of urban service facilities |
| WO2023050955A1 (en) * | 2022-06-01 | 2023-04-06 | 南京大学 | Urban functional zone identification method based on function mixing degree and ensemble learning |
| CN116227791A (en) * | 2023-05-08 | 2023-06-06 | 湖南工商大学 | Visual analysis method for exploring dynamic division of urban functional areas based on semantic fusion model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113705959B (en) * | 2021-05-11 | 2023-08-15 | 北京邮电大学 | Network resource allocation method and electronic equipment |
-
2023
- 2023-06-27 CN CN202310776586.XA patent/CN116756695B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104992260A (en) * | 2015-05-22 | 2015-10-21 | 东南大学 | Urban land function suitability planning method based on multi-dimensional guiding |
| CN109189917A (en) * | 2018-06-27 | 2019-01-11 | 华南师范大学 | A kind of city function limited region dividing method and system merging landscape and social characteristic |
| CN110377807A (en) * | 2019-06-03 | 2019-10-25 | 华南农业大学 | Group of cities functional cohesion and Analysis of Spatial Distribution Pattern method and system |
| CN112559909A (en) * | 2020-12-18 | 2021-03-26 | 浙江工业大学 | Business area discovery method based on GCN embedded spatial clustering model |
| CN113158084A (en) * | 2021-03-25 | 2021-07-23 | 马琦伟 | Method and device for processing movement track data, computer equipment and storage medium |
| CN113077370A (en) * | 2021-03-31 | 2021-07-06 | 北京嘀嘀无限科技发展有限公司 | City function attribute identification method, equipment, medium and product of geographic tile |
| CN113806419A (en) * | 2021-08-26 | 2021-12-17 | 西北大学 | Urban area function identification model and method based on space-time big data |
| CN114626512A (en) * | 2022-05-17 | 2022-06-14 | 南京信息工程大学 | High-temperature disaster forecasting method based on directed graph neural network |
| WO2023050955A1 (en) * | 2022-06-01 | 2023-04-06 | 南京大学 | Urban functional zone identification method based on function mixing degree and ensemble learning |
| CN115545758A (en) * | 2022-09-26 | 2022-12-30 | 苏州大学 | Method and system for self-adaptive incremental site selection of urban service facilities |
| CN116227791A (en) * | 2023-05-08 | 2023-06-06 | 湖南工商大学 | Visual analysis method for exploring dynamic division of urban functional areas based on semantic fusion model |
Non-Patent Citations (5)
| Title |
|---|
| Urban function classification at road segment level using taxi trajectory data: A graph convolutional neural network approach;Sheng Hu等;《Computers, Environment and Urban Systems 87 (2021) 101619》;20210126;1-14 * |
| 图神经网络在复杂图挖掘上的研究进展;刘杰等;软件学报;20220222;第33卷(第10期);3582-3618 * |
| 地理流的空间模式：概念与分类;裴韬等;地球信息科学;20200131;第22卷(第1期);30-40 * |
| 大数据驱动的个体出行模式与城市空间结构交互研究;曹劲舟;中国博士学位论文全文数据库 工程科技Ⅱ辑;20200615;1-96 * |
| 语义轨迹建模与挖掘研究进展;赵斌等;地球信息科学;20200430;第22卷(第4期);842-856 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116756695A (en) | 2023-09-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109508360B (en) | Geographical multivariate stream data space-time autocorrelation analysis method based on cellular automaton | |
| CN106780089B (en) | Permanent basic farmland planning method based on neural network cellular automaton model | |
| Zhou et al. | Urban expansion simulation and development-oriented zoning of rapidly urbanising areas: A case study of Hangzhou | |
| CN112163367B (en) | Firefly algorithm and cellular automaton fused city expansion simulation prediction method | |
| CN116756695B (en) | Urban function collaborative optimization method integrating geographic features and flow features | |
| Fragkias et al. | Modeling urban growth in data-sparse environments: a new approach | |
| CN102646164A (en) | Land use change modeling method and system implemented in combination with spatial filtering | |
| CN113297174B (en) | Land utilization change simulation method based on deep learning | |
| CN115545758A (en) | Method and system for self-adaptive incremental site selection of urban service facilities | |
| CN116756828A (en) | Urban space planning method, medium and system | |
| CN112949997A (en) | System and method for community portrayal in urban planning design | |
| CN116017407A (en) | Method for reliably identifying resident trip mode driven by mobile phone signaling data | |
| Dadashpour Moghaddam et al. | A GIS-based assessment of urban tourism potential with a branding approach utilizing hybrid modeling | |
| CN113159371B (en) | Unknown target feature modeling and demand prediction method based on cross-modal data fusion | |
| CN113240219A (en) | Land utilization simulation and prediction method | |
| CN110264010B (en) | Novel rural power saturation load prediction method | |
| CN116167254A (en) | Multidimensional city simulation deduction method and system based on city big data | |
| CN116258612A (en) | Urban land utilization distribution evaluation method based on knowledge embedding model | |
| CN112818923A (en) | Urban mass living space construction time identification method | |
| CN110909765B (en) | Pedestrian behavior pattern classification method for big track data | |
| Provenzano et al. | Clustering of tourism patterns with self-organizing maps: The case of sicily | |
| Gao et al. | Integrating multi-source geographic big data to delineate urban growth boundary: A case study of Changsha | |
| CN117150698B (en) | Digital twinning-based smart city grid object construction method and system | |
| CN117079124B (en) | Urban and rural landscape image quantification and promotion method based on community differentiation | |
| CN116911055B (en) | Digital twinning-based intelligent city planning management system |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |