CN112115641A - Intelligent city information infrastructure planning system - Google Patents
Intelligent city information infrastructure planning system Download PDFInfo
- Publication number
- CN112115641A CN112115641A CN202010953101.6A CN202010953101A CN112115641A CN 112115641 A CN112115641 A CN 112115641A CN 202010953101 A CN202010953101 A CN 202010953101A CN 112115641 A CN112115641 A CN 112115641A
- Authority
- CN
- China
- Prior art keywords
- planning
- infrastructure
- target
- evaluation parameters
- optimization
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to an intelligent city information infrastructure planning system, which comprises a geographic information identification module, a data processing module and a data processing module, wherein the geographic information identification module is used for acquiring an image of an area to be planned and identifying various construction lands and infrastructures in the image through a convolutional neural network; the parameter calculation module is used for acquiring candidate planning point coordinates of the target infrastructure and calculating a planning score of the target infrastructure according to the candidate planning point coordinates and preset infrastructure evaluation parameters; and the planning point position optimization module is used for acquiring candidate planning point coordinates of all target infrastructures in the area to be planned, calculating planning scores of all the target infrastructures, optimizing for multiple times according to an optimization algorithm, calculating the sum of the planning scores after each optimization, and taking the candidate planning point coordinates corresponding to all the target infrastructures when the sum of the planning scores is the highest as final infrastructure planning coordinates. Compared with the prior art, the method has the advantages of improving the efficiency of target infrastructure point selection in urban planning, reducing the labor cost in urban planning and the like.
Description
Technical Field
The invention relates to the technical field of urban planning, in particular to an intelligent urban information infrastructure planning system.
Background
At present, the demand of 5G communication is increasing, the convergence service with various industries such as transportation, industry and medical treatment is developing gradually, and the construction demand of various facilities in cities, such as communication base stations, pipelines and machine rooms, is increasing. However, such planning still requires a lot of manual operations, and the planning is relatively independent from each other and time-consuming and labor-consuming.
Machine learning is a method for mining the rules of input data and extracting a model for prediction by using various algorithms through data training. The convolutional neural network is a novel machine learning method particularly suitable for mining information hidden in image data, and is widely applied to various fields at present.
The optimization algorithm is divided into a local optimization algorithm and a global optimization algorithm, and common global optimization algorithms include a genetic algorithm, a simulated annealing method, a particle swarm optimization algorithm and the like. The prior art discloses a 5G base station site selection planning method based on an immune genetic algorithm, combines the search characteristic of the genetic algorithm and the self-adaptive characteristic of the immune algorithm, adopts the immune genetic algorithm to carry out mathematical modeling and solving on the 5G base station site selection problem, but is easy to fall into a local optimal solution when the target population is large, and the convergence speed is also limited. The prior art also discloses a communication base station addressing algorithm based on an improved particle swarm optimization, and introduces an inertia weight method, an external diffusion method, an internal collision method and a designated coordinate method, so that the communication base station addressing algorithm is easy to trap into the local optimum early although the searching speed is high.
The simulated annealing algorithm comprises two parts, namely a Metropolis algorithm and an annealing process, has asymptotic convergence, but the solution with higher performance needs to pay the cost of low convergence rate; otherwise, a global optimal solution is likely not obtained.
Besides the global algorithm, a series of search methods based on gradient and derivative exist, and the convergence rate is high, such as Newton's method. The Newton method has the advantages of second-order convergence, high convergence speed and capability of highly approaching an optimal value; however, the inverse matrix of the Hessian matrix of the objective function needs to be solved in each step, and the calculation is complex; meanwhile, newton's method is locally convergent, and when the initial point is not properly selected, the result is often not convergent.
The gradient descent method is an optimization algorithm based on a convex function, and minimizes a given function to its local minimum value by iteratively adjusting parameters. When the objective function is a convex function, the solution of the gradient descent method is a global solution. But in general, the solution is not guaranteed to be the global optimal solution, and the speed is not necessarily the fastest; close to the minimum, convergence slows down, requiring multiple iterations, and may even fall "zig-zag".
The Levenberg-Marquardt method, also called a damped least squares method, combines the characteristics of the two methods (the newton method has the characteristic of fast convergence but is sensitive to the initial point position, and the gradient descent rule is opposite), the LM algorithm has fast convergence speed but needs to calculate the partial derivative for each parameter to be estimated, and when the fitting function f is too complex or the parameters to be estimated are too many, the LM algorithm is not suitable for solving the algorithm.
In summary, although the global optimization algorithm, such as a genetic algorithm, a particle swarm algorithm, and a simulated annealing algorithm, can obtain a global optimal solution meeting the requirements, a relatively slow convergence rate corresponds to the global optimal solution, and if the initial value and the parameter setting are not appropriate, a local optimal solution may be obtained; although the convergence rate of algorithms such as the Newton method, the gradient descent method, the Levenberg-Marquardt method and the like is high, the global optimal solution is not necessarily obtained.
Disclosure of Invention
The invention aims to provide an intelligent city information infrastructure planning system for overcoming the defect that the convergence rate of an algorithm and a global optimal solution in city planning in the prior art cannot be met simultaneously.
The purpose of the invention can be realized by the following technical scheme:
an intelligent city information infrastructure planning system comprises a geographic information identification module, a parameter calculation module and a planning point position optimization module, wherein:
the geographic information identification module is used for acquiring an image of the area to be planned and identifying various construction lands and corresponding infrastructure in the area to be planned through a convolutional neural network;
the parameter calculation module is used for acquiring candidate planning point coordinates of the target infrastructure in the construction land and calculating a planning score of the target infrastructure according to the candidate planning point coordinates and preset infrastructure evaluation parameters;
and the planning point position optimization module is used for acquiring candidate planning point coordinates of all target infrastructures in the area to be planned, calculating planning scores of all the target infrastructures, optimizing the candidate planning point coordinates for multiple times according to an optimization algorithm, continuously improving the planning scores of the target infrastructures, calculating the sum of the planning scores of all the target infrastructures after each optimization, and taking the candidate planning point coordinates corresponding to all the target infrastructures when the sum of the planning scores is the highest as final infrastructure planning coordinates.
The image of the area to be planned comprises a map, a photo and a satellite map.
Further, the map is specifically a land use planning map, and the photo is specifically an aerial photo.
The optimization algorithms include gradient descent methods, newton methods, Levenberg-Marquardt methods, simulated annealing methods, and genetic algorithms.
The infrastructure includes a machine room and a base station.
The geographic information identification module marks the edge and the center of the construction land according to the identified construction land.
The infrastructure evaluation parameters comprise technical evaluation parameters, safety evaluation parameters and economic evaluation parameters.
Further, the technical evaluation parameters comprise a coverage range and a radiation measurement, the safety evaluation parameters comprise a distance between a target infrastructure and a high-risk building in an area to be planned, and the economic evaluation parameters comprise an existing infrastructure and construction cost.
The high-risk building comprises a gas station and a high-voltage line.
Further, the technical evaluation parameter, the safety evaluation parameter and the economic evaluation parameter are determined according to an analytic hierarchy process, and corresponding weight coefficients in the planning scores are determined according to the analytic hierarchy process.
And the optimization algorithm optimizes the candidate planning point coordinates for multiple times in a perturbation optimization mode.
Compared with the prior art, the invention has the following beneficial effects:
according to the method, evaluation parameters are set to score the candidate planning point coordinates of the target infrastructure to obtain the planning scores, the candidate planning point coordinates are optimized through an optimization algorithm, the global optimal solution is judged according to the scores of the planning scores, the problem that convergence speed is low due to the fact that the coordinates are directly optimized is avoided, the efficiency of target infrastructure point selection in urban planning is improved, and the labor cost in urban planning is reduced.
Drawings
FIG. 1 is a schematic structural view of the present invention;
FIG. 2 is a schematic flow chart illustrating position optimization according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of gradient descent method optimization in an embodiment of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
As shown in fig. 1, an intelligent city information infrastructure planning system includes a geographic information identification module, a parameter calculation module, and a planning point location optimization module, wherein:
the geographic information identification module is used for acquiring an image of the area to be planned and identifying various construction lands and corresponding infrastructure in the area to be planned through a convolutional neural network;
the parameter calculation module is used for acquiring candidate planning point coordinates of the target infrastructure in the construction land and calculating a planning score of the target infrastructure according to the candidate planning point coordinates and preset infrastructure evaluation parameters;
and the planning point position optimization module is used for acquiring candidate planning point coordinates of all target infrastructures in the area to be planned, calculating planning scores of all the target infrastructures, optimizing the candidate planning point coordinates for multiple times according to an optimization algorithm, continuously improving the planning scores of the target infrastructures, calculating the sum of the planning scores of all the target infrastructures after each optimization, and taking the candidate planning point coordinates corresponding to all the target infrastructures when the sum of the planning scores is the highest as final infrastructure planning coordinates.
The images of the area to be planned include maps, photographs and satellite maps.
The map is a land use planning map, and the photo is an aerial photo.
Optimization algorithms include gradient descent methods, newton methods, Levenberg-Marquardt methods, simulated annealing methods, and genetic algorithms.
The infrastructure includes a machine room and a base station.
The geographic information identification module marks the edge and the center of the construction land according to the identified construction land.
The infrastructure evaluation parameters include technical evaluation parameters, safety evaluation parameters and economic evaluation parameters.
The technical evaluation parameters comprise coverage and radiometric values, the safety evaluation parameters comprise the distance between the target infrastructure and the high-risk buildings in the area to be planned, and the economic evaluation parameters comprise the existing infrastructure and construction cost.
High-risk buildings include gas stations and high-voltage lines.
And determining the corresponding weight coefficients of the technical evaluation parameters, the safety evaluation parameters and the economic evaluation parameters in the planning scores according to an analytic hierarchy process.
The mode of optimizing the candidate planning point coordinates for multiple times by the optimization algorithm is perturbation optimization.
Example one
Taking a base station as an example to carry out city planning, specifically executing the following steps:
step S11: judging whether a planning map exists, if yes, turning to step S12, and if not, identifying the drawing of the geographic information system through a Convolutional Neural Network (CNN) to generate a corresponding planning map;
step S12: geographic information identification is carried out on the area planning map, the area planning map comprises multiple construction lands, and each construction land is provided with corresponding RGB colors for distinguishing. In this embodiment, the commercial site color RGB value is (255, 0, 0), and the residential site of one type is (255, 255, 107);
step S13: obtaining the range, edge and area information of each construction land, dividing coordinate points of the construction land on a planning graph into a plurality of subsets, and marking the construction land;
step S14: as shown in fig. 2, selecting a target site in the planning graph, calculating a planning score of the target site, adjusting the position of the target site according to an optimization algorithm, calculating the planning score again, judging whether the planning score meets a preset score threshold, and if so, outputting the position of the target site corresponding to the current planning score;
step S15: acquiring a propagation model of a target station, and calculating the path loss of the target station according to the propagation model to obtain the radius of a coverage area of the target station;
step S16: acquiring the user density and the number of users in the coverage area of the target site, and calculating to obtain the total investment cost of the base station planning combination by combining the radius of the coverage area of the target site;
step S17: and calculating the planning score of each site in the planning map, and calculating the overall evaluation score of the base station combination according to the weight obtained by the hierarchical analysis.
In this embodiment, the propagation model is specifically a COST-231Hata model, and the calculation formula of the path loss L is specifically as follows:
L＝46.3+33.9logf-13.82loghb-a(hm)+[44.9-6.55loghb]logd+CM
wherein:
f is carrier working frequency, and the value range is 1500MHz or more and 2300MHz or less;
hbfor the effective height of the transmitting antenna, h is more than or equal to 30mb≤200m；
hmFor the effective height of the receiving antenna, h is more than or equal to 1mm≤10m；
d is the distance between the transmitter and the receiver, and d is more than or equal to 1km and less than or equal to 20 km;
a(hm) Modifying the factor for the mobile station antenna;
CMcorrection factor for large city center, in medium and suburbs, CM0dB, in the center of a large city, CM＝3dB。
In this embodiment, the optimization algorithm is optimized by a gradient descent method, as shown in fig. 3, the following steps are specifically implemented when the optimization algorithm is executed:
step S21: determining a propagation model and a corresponding target function and a loss function thereof;
step S22: acquiring step length, learning rate and initial position of a target station;
step S23: calculating the gradient of the loss function according to the learning rate;
step S24: moving the target station according to the step length, and calculating the gradient corresponding to the new position of the target station;
step S25: and judging whether the gradient vector is smaller than a preset gradient threshold value, if so, outputting the optimized position of the target station, and otherwise, turning to the step 24.
This embodiment minimizes the given function to its local minimum by iteratively adjusting the parameters. In this embodiment, the optimized site score reaches 126.5108, which is greatly improved compared to the initial value 0.
In addition, it should be noted that the specific implementation examples described in this specification may have different names, and the above contents described in this specification are only illustrations of the structures of the present invention. All equivalent or simple changes in the structure, characteristics and principles of the invention are included in the protection scope of the invention. Various modifications or additions may be made to the described embodiments or methods may be similarly employed by those skilled in the art without departing from the scope of the invention as defined in the appending claims.
Claims (10)
1. The utility model provides an intelligence city information infrastructure planning system which characterized in that, includes geographic information identification module, parameter calculation module and planning point position optimization module, wherein:
the geographic information identification module is used for acquiring an image of the area to be planned and identifying various construction lands and corresponding infrastructure in the area to be planned through a convolutional neural network;
the parameter calculation module is used for acquiring candidate planning point coordinates of the target infrastructure in the construction land and calculating a planning score of the target infrastructure according to the candidate planning point coordinates and preset infrastructure evaluation parameters;
and the planning point position optimization module is used for acquiring candidate planning point coordinates of all target infrastructures in the area to be planned, calculating planning scores of all the target infrastructures, optimizing the candidate planning point coordinates for multiple times according to an optimization algorithm, calculating the sum of the planning scores of all the target infrastructures after each optimization, and taking the candidate planning point coordinates corresponding to all the target infrastructures when the sum of the planning scores is the highest as final infrastructure planning coordinates.
2. The intelligent city information infrastructure planning system of claim 1 wherein the images of the area to be planned include maps, photographs and satellite maps.
3. An intelligent city information infrastructure planning system as claimed in claim 2 wherein the map is specifically a land use planning map and the photographs are specifically aerial photographs.
4. An intelligent city information infrastructure planning system according to claim 1, wherein the optimization algorithms include gradient descent method, newton method, Levenberg-Marquardt method, simulated annealing method and genetic algorithm.
5. An intelligent city information infrastructure planning system as claimed in claim 1 wherein the infrastructure includes a computer room and a base station.
6. The intelligent city information infrastructure planning system of claim 1, wherein the geographic information identification module marks the edge and center of the construction site according to the identified construction site.
7. The intelligent city information infrastructure planning system of claim 1, wherein the infrastructure evaluation parameters include technical evaluation parameters, security evaluation parameters, and economic evaluation parameters.
8. The intelligent city information infrastructure planning system of claim 7, wherein the technical evaluation parameters include coverage and radiometric, the safety evaluation parameters include distance of target infrastructure from high-risk buildings in the area to be planned, and the economic evaluation parameters include existing infrastructure and construction costs.
9. The intelligent city information infrastructure planning system of claim 7, wherein the technical evaluation parameters, the safety evaluation parameters and the economic evaluation parameters determine their corresponding weight coefficients in the planning score according to an analytic hierarchy process.
10. The intelligent city information infrastructure planning system of claim 1, wherein the optimization algorithm optimizes the candidate planning point coordinates multiple times in a perturbation optimization.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010953101.6A CN112115641A (en) | 2020-09-11 | 2020-09-11 | Intelligent city information infrastructure planning system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010953101.6A CN112115641A (en) | 2020-09-11 | 2020-09-11 | Intelligent city information infrastructure planning system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112115641A true CN112115641A (en) | 2020-12-22 |
Family
ID=73803630
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010953101.6A Pending CN112115641A (en) | 2020-09-11 | 2020-09-11 | Intelligent city information infrastructure planning system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112115641A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103049799A (en) * | 2012-12-10 | 2013-04-17 | 河海大学 | Multi-objective-optimization-based power grid planning and designing method |
| CN105512752A (en) * | 2015-11-30 | 2016-04-20 | 北京大学 | Urban public service facility site selection method |
| CN107918776A (en) * | 2017-11-01 | 2018-04-17 | 中国科学院深圳先进技术研究院 | A kind of plan for land method, system and electronic equipment based on machine vision |
| CN110675189A (en) * | 2019-09-20 | 2020-01-10 | 东北大学 | Unmanned shelf location method based on Wide & Deep model and genetic algorithm |
| WO2020151089A1 (en) * | 2019-01-25 | 2020-07-30 | 东南大学 | Automatic city land identification system integrating industrial big data and building form |
-
2020
- 2020-09-11 CN CN202010953101.6A patent/CN112115641A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103049799A (en) * | 2012-12-10 | 2013-04-17 | 河海大学 | Multi-objective-optimization-based power grid planning and designing method |
| CN105512752A (en) * | 2015-11-30 | 2016-04-20 | 北京大学 | Urban public service facility site selection method |
| CN107918776A (en) * | 2017-11-01 | 2018-04-17 | 中国科学院深圳先进技术研究院 | A kind of plan for land method, system and electronic equipment based on machine vision |
| WO2020151089A1 (en) * | 2019-01-25 | 2020-07-30 | 东南大学 | Automatic city land identification system integrating industrial big data and building form |
| CN110675189A (en) * | 2019-09-20 | 2020-01-10 | 东北大学 | Unmanned shelf location method based on Wide & Deep model and genetic algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jiang et al. | A Practical and Economical Ultra‐wideband Base Station Placement Approach for Indoor Autonomous Driving Systems | |
| CN109743683B (en) | Method for determining position of mobile phone user by adopting deep learning fusion network model | |
| CN108320323B (en) | Building three-dimensional modeling method and device | |
| CN105430664B (en) | It is a kind of to be fitted the method and apparatus that path loss is propagated in prediction based on classification | |
| WO2016067017A1 (en) | Method for predicting indoor three-dimensional space signal field strength using an outdoor-to-indoor propagation model | |
| CN103778477B (en) | Monitor station cloth station method and apparatus | |
| CN104410978A (en) | Method and device of evaluating site planning | |
| CN109041210B (en) | Wireless sensor network positioning method | |
| CN105512344B (en) | A kind of querying method of indoor moving object relative location | |
| US11145089B2 (en) | Method for measuring antenna downtilt based on multi-scale detection algorithm | |
| CN111027491A (en) | Flame detection method and device based on anchor-free mechanism | |
| CN110674975B (en) | Spatial layout method and device for reducing carbon emission in urban planning | |
| Wang et al. | Optimizing the ultra-dense 5G base stations in urban outdoor areas: Coupling GIS and heuristic optimization | |
| CN115828479B (en) | Planning method for construction channel of power transmission line | |
| CN106056577B (en) | SAR image change detection based on MDS-SRM Mixed cascading | |
| CN111461251A (en) | Indoor positioning method of WiFi fingerprint based on random forest and self-encoder | |
| Bharadwaj et al. | Determination of optimal location for setting up cell phone tower in city environment using lidar data | |
| CN109472416B (en) | Indoor path planning method and device based on automatic road network data extraction, and client | |
| CN114828026A (en) | Base station planning method, device, equipment, storage medium and program product | |
| KR20200132225A (en) | Prediction Method of Urban Land Use Change by Autonomous Vehicle | |
| US11877159B2 (en) | Computing system that is configured to assign wireless beacons to positions within a building | |
| CN112115641A (en) | Intelligent city information infrastructure planning system | |
| US11237008B2 (en) | System and method for controlling vehicular pollution concentration and providing maximum traffic flow throughput | |
| CN112839342B (en) | Disaster relief mobile emergency base station site selection method based on support vector machine | |
| Sun et al. | Study on safe evacuation routes based on crowd density map of shopping mall |
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: 20201222 |
|
| RJ01 | Rejection of invention patent application after publication |