CN115470707A - City scene simulation system - Google Patents
City scene simulation system Download PDFInfo
- Publication number
- CN115470707A CN115470707A CN202211156784.8A CN202211156784A CN115470707A CN 115470707 A CN115470707 A CN 115470707A CN 202211156784 A CN202211156784 A CN 202211156784A CN 115470707 A CN115470707 A CN 115470707A
- Authority
- CN
- China
- Prior art keywords
- scene
- library
- training
- neural network
- 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
- 238000004088 simulation Methods 0.000 title claims abstract description 56
- 238000005457 optimization Methods 0.000 claims abstract description 48
- 238000013135 deep learning Methods 0.000 claims abstract description 35
- 238000000034 method Methods 0.000 claims abstract description 30
- 229910052799 carbon Inorganic materials 0.000 claims abstract description 25
- 238000011161 development Methods 0.000 claims abstract description 19
- 238000013439 planning Methods 0.000 claims abstract description 9
- 230000009467 reduction Effects 0.000 claims abstract description 8
- 230000007613 environmental effect Effects 0.000 claims abstract description 6
- 230000009471 action Effects 0.000 claims abstract description 5
- 238000004519 manufacturing process Methods 0.000 claims abstract description 5
- 239000000463 material Substances 0.000 claims abstract description 3
- 230000006870 function Effects 0.000 claims description 54
- 238000012549 training Methods 0.000 claims description 40
- 238000013528 artificial neural network Methods 0.000 claims description 33
- 238000009826 distribution Methods 0.000 claims description 17
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 claims description 16
- 229910002092 carbon dioxide Inorganic materials 0.000 claims description 8
- 238000005265 energy consumption Methods 0.000 claims description 6
- 238000013515 script Methods 0.000 claims description 6
- 230000008520 organization Effects 0.000 claims description 4
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 claims description 3
- 239000001569 carbon dioxide Substances 0.000 claims description 3
- 238000013079 data visualisation Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000009877 rendering Methods 0.000 claims description 3
- 238000011534 incubation Methods 0.000 claims description 2
- 238000011084 recovery Methods 0.000 claims 1
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 abstract description 4
- 230000018109 developmental process Effects 0.000 description 14
- 238000003860 storage Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000005094 computer simulation Methods 0.000 description 4
- 238000013526 transfer learning Methods 0.000 description 4
- 230000004075 alteration Effects 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 238000005338 heat storage Methods 0.000 description 2
- 238000004806 packaging method and process Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000012447 hatching Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G06Q50/40—
Abstract
The invention discloses a city scene simulation system, which comprises: the system comprises a city scene development editor, a city scene library module and an optimization module; a scene development editor calls and self-defines a combined solution library, and creates an urban scene library module according to roles, vehicles and environmental systems to render and preview a three-dimensional scene in real time; the solution library comprises a scene resource library, a task action library and a product material library; the city scene library module comprises a traffic planning scene library, a manufacturing industry scene library, a business center scene library and a low-carbon life scene library; the optimization module comprises a traffic optimization submodule, an industry optimization submodule, a business optimization submodule and a low-carbon optimization submodule. The simulation prediction method based on the PINN principle and the deep learning can be used for dynamically simulating and predicting the human flow, the traffic condition, the carbon emission reduction, the commercial layout and the like in various scenes.
Description
Technical Field
The invention belongs to the technical field of simulation, and particularly relates to a city scene simulation system.
Background
Virtual simulation, or simulation technology, is a technology that uses one system to simulate another real system. Virtual simulation is actually a computer system that can create and experience a virtual world. Such a virtual world may be generated by a computer, may be a real world representation, or may be a world in which a user is in the conception, and may interact with the virtual world naturally by means of various sensing channels such as vision, hearing, and touch.
At present, a game engine, a teaching simulation system or a smart city and other simulation systems exist, wherein the smart city simulates the space at present, but various dynamic scenes exist for the city, and the space simulation system performs simulation on the space in an isolated manner and is still different from the real world. Current content creation and experience is more targeted at entertainment and brand marketing, and does not see a virtual platform that can interact with real business scenarios in advance, which can introduce greater productivity in industry development.
In the process of the development of the Yuanzhou industry, each technical company prepares technical capability and business layout in the fields of games and industry. But the method lacks tools which can be used in business universality and small and medium-sized enterprises and business teams, business scenes and business level simulation temporarily have no tools for supporting creation, and a platform which can integrate a plurality of business scenes and development requirements of cross-industry is lacking.
Disclosure of Invention
In view of this, the invention provides a simulation system capable of simulating various dynamic scenes in a city.
The invention discloses a city scene simulation system, which comprises: the system comprises a city scene development editor, a city scene library module and an optimization module;
the scene development editor calls and self-defines a combined solution library, creates an urban scene library module according to roles, vehicles and environmental systems, and carries out real-time rendering and previewing of a three-dimensional scene; the solution library comprises a scene resource library, a task action library and a product material library;
the city scene library module comprises a traffic planning scene library, a manufacturing industry scene library, a business center scene library and a low-carbon life scene library;
the optimization module comprises: the system comprises a traffic optimization submodule, an industry optimization submodule, a business optimization submodule and a low-carbon optimization submodule;
the traffic optimization submodule generates a new data set by using a GAN neural network according to the input road planning parameters and the unmanned vehicle action parameters, and performs deep learning on the data set to obtain the optimal driving path of the unmanned vehicle;
the industry optimization submodule generates a new data set by using a GAN neural network according to the input industry planning parameters, the area position parameters and the population structure parameters, and performs deep learning on the data set to output the optimal industry layout;
the business optimization submodule generates a new data set by using a GAN neural network according to the input business type parameters and business position parameters, and performs deep learning on the data set to output an optimal pedestrian flow model;
the low-carbon optimization submodule generates a new data set by using a GAN neural network according to the input spatial layout, the hot and humid environment, the light environment and the air flow organization parameters, deeply learns the data set and outputs CO2 emission reduction, and the spatial layout comprises the arrangement of an air conditioner terminal system, a shading system, a lighting system and doors and windows.
Further, the traffic planning module comprises an automatic driving loop scene and an underground logistics distribution network scene; the manufacturing industry scene library comprises an incubation space scene, an enterprise service space scene and an enterprise headquarter scene; the business center scene library comprises a business complex scene and an entertainment culture center scene; the low-carbon living scene library comprises an underground energy network scene, a green building scene, a garbage recycling scene and a rainwater collection scene.
Furthermore, the scene development engine builds scenes in a modularization mode, performs scene simulation and data visualization through scripts, and performs preset simulation on the environment based on geographic information.
Further, the traffic optimization submodule performs deep learning by using a PINN-based deep learning method, wherein an objective function is as follows:
wherein route is a control variable and represents a recommended route of a vehicle or a dredging signal of a crowd or real-time control of a traffic signal lamp; u is a state variable concerned by the system, and comprises road traffic flow, vehicle congestion condition, crowd distribution condition, real-time base station signal, scene energy consumption and transaction information; l is u And L f Calculating to obtain a loss function according to state measured data and a system operation rule, wherein the loss function is used for training a neural network used in the PINN, and W is a neural network parameter and is obtained through training; x and t are space and time variables of system operation,andfor the specific time-space node of interest of the state loss function after the system discretization,andspecific time space nodes concerned by the loss function of the operation rule after the system is dispersed; u. of i To collect status data for training, N u And N f Is the number of data samples sampled.
Further, the industry optimization submodule performs deep learning by using a PINN-based deep learning method, wherein an objective function is as follows:
wherein, the site is a control variable and represents the setting distribution of each industry; u is a state variable concerned by the system, including urban capacity, population number and industrial chain distribution; l is u And L f Calculating to obtain a loss function according to state measured data and a system operation rule, wherein the loss function is used for training a neural network used in the PINN, and W is a neural network parameter and is obtained through training; x and t are space and time variables of system operation,andfor the particular spatio-temporal node of interest to the state loss function after system discretization,andspecific time space nodes concerned by the loss function of the operation rule after the system is dispersed; u. of i To collect status data for training, N u And N f For samplingThe number of data samples.
Further, the business optimization sub-module performs deep learning by using a PINN-based deep learning method, wherein the objective function is as follows:
wherein the flowrate is a control variable and represents the human flow; u is a state variable concerned by the system, including commercial network distribution, parking lot position and public transport station position; l is a radical of an alcohol u And L f Calculating to obtain a loss function according to state measured data and a system operation rule, wherein the loss function is used for training a neural network used in the PINN, and W is a neural network parameter and is obtained through training; x and t are space and time variables of system operation,andfor the particular spatio-temporal node of interest to the state loss function after system discretization,andspecific time space concerned by running rule loss function after system dispersionAn inter-node; u. u i To collect status data for training, N u And N f Is the number of data samples sampled.
Further, the low-carbon optimization submodule performs deep learning by using a PINN-based deep learning method, wherein an objective function is as follows:
wherein CO2 is carbon dioxide emission reduction amount; u is a state variable concerned by the system, and comprises low-carbon monitoring equipment distribution and energy consumption coefficient; l is u And L f Calculating to obtain a loss function according to state measured data and a system operation rule, wherein the loss function is used for training a neural network used in the PINN, and W is a neural network parameter and is obtained through training; x and t are space and time variables for system operation,andfor the particular spatio-temporal node of interest to the state loss function after system discretization,andspecific time space nodes concerned by the loss function of the operation rule after the system is dispersed; u. of i To collect status data for training, N u And N f Is the number of data samples sampled.
The invention has the following beneficial effects:
by the method, a user can create a future business scene and simulate the data-driven business value on the meta-universe platform, and the created business content has experience of being placed in the meta-universe platform;
the simulation prediction method based on the PINN principle and the deep learning can be used for dynamically simulating and predicting the human flow, the traffic condition, the carbon emission reduction, the commercial layout and the like in various scenes.
Drawings
FIG. 1 is a block diagram of the system of the present invention.
Detailed Description
The invention is further described with reference to the accompanying drawings, but the invention is not limited in any way, and any alterations or substitutions based on the teaching of the invention are within the scope of the invention.
The urban scene dynamic simulation system comprises a dynamic simulation 3D real-time rendering and a complex system simulation prediction technology based on the PINN principle and deep learning. Wherein, the complex system simulation prediction technology based on the PINN principle and deep learning takes the screened characteristics as input, carries out simulation model construction, then carries out adjustment and optimization on the simulation result, carries out training and transfer learning on the model, after training is finished, the simulation scene is adapted, after the generalization capability of the model is improved, if the generalization capability of the model does not meet the requirement, the steps of simulation data structure adjustment and optimization, model training and transfer learning are continuously carried out. And after the generalization capability of the model meets the requirement, constructing and packaging the simulation framework with the model, and performing simulation prediction on the system scene.
Specifically, the scene dynamic simulation development system disclosed by the invention comprises a scene development editor, a scene simulator and a scene optimization module;
the scene development editor comprises a scene creating module and a scene developing module; a scene editor in the scene creation module acquires roles, vehicles and environmental resources of a scene module library and a local asset library in the Unity3D, creates a city application scene, and then renders and previews a three-dimensional scene in real time; the scene development module sets and runs a script to obtain a script prefabricated part, then the script prefabricated part is fused with an application scene, and VR, video, images and programs after data fusion are exported; in some embodiments, the scenario development editor supports 3D file importation in multiple formats. In some embodiments, the scene development editor builds scenes through modularization, performs scene simulation and data visualization through scripts, and performs preset simulation on the environment based on geographic information.
The scene simulator imports a program output by the scene development editor, and carries out system simulation display, scene effect display and real-time data display after parameterization adjustment to obtain a scene simulation result and then stores the scene simulation result; in some embodiments, the scene simulator comprises a scene dynamics simulation module, a simulation parameter control module, a real-time scene data display module, and a digital twinning simulator; the digital twin simulator is based on data driving and AI driving.
The scene optimization module generates high-quality data in virtual simulation by using a GAN neural network according to the stored scene simulation result; selecting an optimal feature combination from the data through feature engineering; inputting the screened features into a simulation prediction system based on a PINN principle and deep learning, constructing a simulation model, adjusting and optimizing a simulation result, training and transfer learning the model, adapting a simulation scene after the training is finished, and continuing to perform steps of adjusting and optimizing a simulation data structure, training the model and transfer learning if the generalization capability of the model does not meet the requirement after the generalization capability of the model is improved; and after the generalization capability of the model meets the requirement, constructing and packaging a simulation framework, and performing system scene simulation prediction based on a data-driven prediction control method to optimize the scene in virtual simulation.
A Generative Adaptive Networks (GAN) is a deep learning model, and is one of the most promising methods for unsupervised learning in complex distribution in recent years. The model passes through (at least) two modules in the framework: mutual game learning of the Generative Model and the Discriminative Model produces a reasonably good output. Excellent GAN application requires good training methods, otherwise the output may be unsatisfactory due to the freedom of the neural network model. In the invention, the generator and the discriminator of the GAN are utilized to generate the prototype of the role, the vehicle, the environmental resource, the building space, the road plan, the business center layout and the like of the scene so as to perfect the role, the vehicle, the environmental resource, the building space, the road plan, the business center layout and the like of the scene module library and the local asset library.
The PINN (neural network based on physical information) can restore all flow fields through observed partial velocity field data, so that information such as lift force, resistance, displacement change and the like of a flow-around object is obtained, and the method has important significance for researching a resonance phenomenon induced by eddy current. The change of the scene, such as the flow of people, the traffic condition and the like, is also a fluctuation in physics in a certain sense, so that the method creatively applies the PINN to the dynamic scene simulation, and obtains the future value information of the dynamic scene by using the observed value characteristic data, which has important significance for researching the dynamic scene simulation.
In one embodiment, in a smart unmanned vehicle scene, information and communication technology is fully used for sensing, analyzing and integrating various key information of a city operation core system, so that intelligent response is made to various demands including smart travel. A multidimensional and three-dimensional intelligent transportation scheme is designed to provide efficient and convenient travel infrastructure to improve the open and transparent digital city management capacity, a traffic optimization submodule performs deep learning by using a PINN-based deep learning method, and an objective function is as follows:
wherein route is a control variable and represents a recommended route of a vehicle or a dredging signal of a crowd or real-time control of a traffic signal lamp; u is a state variable concerned by the system, and comprises road traffic flow, vehicle congestion condition, crowd distribution condition, real-time base station signal, scene energy consumption and transaction information; l is u And L f Calculating to obtain a loss function according to state measured data and a system operation rule, wherein the loss function is used for training a neural network used in the PINN, and W is a neural network parameter and is obtained through training; x and t are space and time variables of system operation,andfor the particular spatio-temporal node of interest to the state loss function after system discretization,andspecific time space nodes concerned by the operation rule loss function after the system is dispersed; u. of i To collect status data for training, N u And N f Is the number of data samples sampled.
In some embodiments, the method type of the intelligent unmanned vehicle scene and the invention can support a community unmanned logistics scene, an intelligent connection scene, a customized travel scene, an underground logistics scene, an intelligent unmanned vehicle scene, intelligent underground parking and the like, and the invention is not repeated herein.
In one embodiment, the industrial infrastructure can support the continuous development of urban innovation entrepreneurship enthusiasm, including the intelligent promotion of the creation, application, protection and management capabilities of all-regional intellectual property rights, and the building of a harmonious industrial park from multiple angles such as natural and human fusion, occupation and living fusion and the like. The intelligent and convenient industrial service space of the shared innovative hatching service system is provided by fusing the established intelligent advanced research and development environment. The industry optimization submodule carries out deep learning by using a PINN-based deep learning method, wherein an objective function is as follows:
wherein the site is a control variable and represents the setting distribution of each industry; u is a state variable concerned by the system, including urban capacity, population number and industrial chain distribution; l is a radical of an alcohol u And L f Calculating according to state measured data and a system operation rule to obtain a loss function for training a neural network used in the PINN, wherein W is a neural network parameter and is obtained through training; x and t are space and time variables for system operation,andspecific times of interest for state loss function after system dispersionThe nodes of the space are connected with each other,andspecific time space nodes concerned by the loss function of the operation rule after the system is dispersed; u. of i To collect status data for training, N u And N f Is the number of data samples sampled.
In one embodiment, the business optimization sub-module performs deep learning using a PINN-based deep learning method, wherein the objective function is:
wherein flowrate is human traffic, t is, u is, x is, L u Is as follows, L f W is, N u In order to realize the purpose,
in order to realize the purpose of the method,in order to realize the purpose,u is i Is N f Is as follows.
In some embodiments, similar to the business optimization embodiment, the method can support a civil service scenario, a smart city display scenario, and the like, and the method is not repeated herein.
In some embodiments, the low-carbon optimization submodule establishes a disaster-resistant reasonable resource allocation capability for the city, has a low-carbon economic and industrial structure, is environment-friendly, and has a benign and sustainable energy ecosystem. The low-carbon optimization submodule carries out deep learning by using a PINN-based deep learning method, wherein an objective function is as follows:
wherein CO2 is carbon dioxide emission reduction amount; u is a state variable concerned by the system, and comprises low-carbon monitoring equipment distribution and energy consumption coefficient; l is u And L f Calculating to obtain a loss function according to state measured data and a system operation rule, wherein the loss function is used for training a neural network used in the PINN, and W is a neural network parameter and is obtained through training; x and t are space and time variables of system operation,andfor the specific time-space node of interest of the state loss function after the system discretization,andspecific time space nodes concerned by the loss function of the operation rule after the system is dispersed; u. u i To collect status data for training, N u And N f Is the number of data samples sampled.
The parameters received by the low-carbon optimization submodule comprise building three-dimensional modeling, spatial layout, building form, an envelope structure and the like. The building energy system can be modeled: the system comprises an electric power system (a power grid and a photovoltaic system), a cold and heat source system (a heat pump, a water chilling unit, a full air system and a CHP system), a heat storage system (a heat storage tank, a cold storage tank and an ice storage tank), a storage battery system and the like.
In some embodiments, the parameters received by the low carbon optimization sub-module further include a build space three-dimensional modeling: spatial layout, hot and humid environment, light environment, air current organization and the like, and also comprises a built spatial environment control system modeling: air conditioner end system, solar shading system, lighting system, door and window, etc.
In some embodiments, the parameters received by the low-carbon optimization submodule further include a built-up space photothermal partitioning module, an environment control system intelligent joint control module, a target person (crowd) environment preference description and identification module, a target person (crowd) regulation and control module, and a habit description and identification module.
In some embodiments, a large number of outdoor microclimate (external disturbance) parameters are generated through simulation based on a Monte Carlo random simulation or a GAN network, or a large number of personnel indoor situation (internal disturbance) parameters are generated through simulation based on the Monte Carlo random simulation or the GAN network, and the parameters are optimized through a PINN neural network of a low-carbon optimization submodule, so that parameter combinations and parameter values when the optimal reduced displacement is output are obtained.
The invention has the following beneficial effects:
by the method, a user can create a future business scene and simulate the data-driven business value on the meta-space platform, and the created business content has experience in the meta-space platform;
the simulation prediction method based on the PINN principle and the deep learning can be used for dynamically simulating and predicting the human flow, the traffic condition, the carbon emission reduction, the commercial layout and the like in various scenes.
The word "preferred" is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as "preferred" is not necessarily to be construed as advantageous over other aspects or designs. Rather, use of the word "preferred" is intended to present concepts in a concrete fashion. The term "or" as used in this application is intended to mean an inclusive "or" rather than an exclusive "or". That is, unless specified otherwise or clear from context, "X employs A or B" is intended to include either of the permutations as a matter of course. That is, if X employs A; b is used as X; or X employs both A and B, then "X employs A or B" is satisfied in any of the foregoing examples.
Also, although the disclosure has been shown and described with respect to one or an implementation, equivalent alterations and modifications will occur to others skilled in the art based upon a reading and understanding of this specification and the annexed drawings. The present disclosure includes all such modifications and alterations, and is limited only by the scope of the appended claims. In particular regard to the various functions performed by the above described components (e.g., elements, etc.), the terms used to describe such components are intended to correspond, unless otherwise indicated, to any component which performs the specified function of the described component (e.g., that is functionally equivalent), even though not structurally equivalent to the disclosed structure which performs the function in the herein illustrated exemplary implementations of the disclosure. In addition, while a particular feature of the disclosure may have been disclosed with respect to only one of several implementations, such feature may be combined with one or other features of the other implementations as may be desired and advantageous for a given or particular application. Furthermore, to the extent that the terms "includes," has, "" contains, "or variants thereof are used in either the detailed description or the claims, such terms are intended to be inclusive in a manner similar to the term" comprising.
Each functional unit in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or a plurality of or more than one unit are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium. The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Each apparatus or system described above may execute the storage method in the corresponding method embodiment.
In summary, the above-mentioned embodiment is an implementation manner of the present invention, but the implementation manner of the present invention is not limited by the above-mentioned embodiment, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be regarded as equivalent replacements within the protection scope of the present invention.
Claims (7)
1. A simulation system for urban scene, comprising: the system comprises a city scene development editor, a city scene library module and an optimization module;
the scene development editor calls and self-defines a combined solution library, creates an urban scene library module according to roles, vehicles and environmental systems, and carries out real-time rendering and previewing of a three-dimensional scene; the solution library comprises a scene resource library, a task action library and a product material library;
the city scene library module comprises a traffic planning scene library, a manufacturing industry scene library, a business center scene library and a low-carbon life scene library;
the optimization module comprises: the system comprises a traffic optimization submodule, an industrial optimization submodule, a commercial optimization submodule and a low-carbon optimization submodule;
the traffic optimization submodule generates a new data set by using a GAN neural network according to the input road planning parameters and the unmanned vehicle action parameters, and performs deep learning on the data set to obtain the optimal driving path of the unmanned vehicle;
the industry optimization submodule generates a new data set by using a GAN neural network according to the input industry planning parameter, the area position parameter and the population structure parameter, performs deep learning on the data set and outputs the optimal industry layout;
the business optimization submodule generates a new data set by using a GAN neural network according to the input business type parameters and the business position parameters, and deeply learns the data set to output an optimal pedestrian flow model;
the low-carbon optimization submodule generates a new data set by using a GAN neural network according to the input spatial layout, the hot and humid environment, the light environment and the air flow organization parameters, deeply learns the data set and outputs CO2 emission reduction, and the spatial layout comprises the arrangement of an air conditioner terminal system, a shading system, a lighting system and doors and windows.
2. The city scene simulation system of claim 1, wherein the traffic planning module comprises an automatic driving loop scene and an underground logistics distribution network scene; the manufacturing industry scene library comprises an incubation space scene, an enterprise service space scene and an enterprise headquarter scene; the business center scene library comprises a business complex scene and an entertainment culture center scene; the low-carbon life scene library comprises an underground energy network scene, a green building scene, a garbage recovery scene and a rainwater collection scene.
3. The city scene simulation system according to claim 1, wherein the scene development engine builds scenes through modularization, performs scene simulation and data visualization through scripts, and performs preset simulation on the environment based on geographic information.
4. The urban scene simulation system according to claim 1, wherein the traffic optimization sub-module performs deep learning using a PINN-based deep learning method, wherein the objective function is:
wherein route is a control variable and represents a recommended route of a vehicle or a dredging signal of a crowd or real-time control of a traffic signal lamp; u is a state variable concerned by the system, and comprises road traffic flow, vehicle congestion condition, crowd distribution condition, real-time base station signal, scene energy consumption and transaction information; l is u And L f Calculating according to state measured data and a system operation rule to obtain a loss function for training a neural network used in the PINN, wherein W is a neural network parameter and is obtained through training; x and t are space and time variables for system operation,andfor the specific time-space node of interest of the state loss function after the system discretization,andspecific time space nodes concerned by the operation rule loss function after the system is dispersed; u. of i To collect status data for training, N u And N f To sampleThe number of data samples.
5. The urban scene simulation system according to claim 1, wherein the industry optimization submodule performs deep learning by using a PINN-based deep learning method, wherein the objective function is as follows:
wherein, the site is a control variable and represents the setting distribution of each industry; u is a state variable concerned by the system, including urban capacity, population number and industrial chain distribution; l is a radical of an alcohol u And L f Calculating according to state measured data and a system operation rule to obtain a loss function for training a neural network used in the PINN, wherein W is a neural network parameter and is obtained through training; x and t are space and time variables of system operation,andfor the specific time-space node of interest of the state loss function after the system discretization,andspecific time space nodes concerned by the operation rule loss function after the system is dispersed; u. u i To collect status data for training, N u And N f Is the number of data samples sampled.
6. The urban scene simulation system according to claim 1, wherein the business optimization sub-module performs deep learning using a PINN-based deep learning method, wherein the objective function is:
wherein flowrate is a control variable and represents the flow of people; u is a state variable concerned by the system, including commercial network distribution, parking lot position and public transport station position; l is u And L f Calculating according to state measured data and a system operation rule to obtain a loss function for training a neural network used in the PINN, wherein W is a neural network parameter and is obtained through training; x and t are space and time variables of system operation,andfor the specific time-space node of interest of the state loss function after the system discretization,andspecific time space nodes concerned by the loss function of the operation rule after the system is dispersed; u. of i To collect status data for training, N u And N f Is the number of data samples sampled.
7. The urban scene simulation system according to claim 1, wherein the low-carbon optimization submodule performs deep learning using a PINN-based deep learning method, wherein the objective function is:
wherein CO2 is carbon dioxide emission reduction amount; u is a state variable concerned by the system, and comprises low-carbon monitoring equipment distribution, an energy consumption coefficient, a spatial layout, a hot and humid environment, a light environment and an airflow organization; l is u And L f Is according to the shapeCalculating the state measured data and the system operation rule to obtain a loss function for training a neural network used in the PINN, wherein W is a neural network parameter and is obtained through training; x and t are space and time variables of system operation,andfor the particular spatio-temporal node of interest to the state loss function after system discretization,andspecific time space nodes concerned by the operation rule loss function after the system is dispersed; u. u i To collect status data for training, N u And N f Is the number of data samples sampled.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211156784.8A CN115470707A (en) | 2022-09-22 | 2022-09-22 | City scene simulation system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211156784.8A CN115470707A (en) | 2022-09-22 | 2022-09-22 | City scene simulation system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115470707A true CN115470707A (en) | 2022-12-13 |
Family
ID=84335814
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211156784.8A Pending CN115470707A (en) | 2022-09-22 | 2022-09-22 | City scene simulation system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115470707A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116415756A (en) * | 2023-05-29 | 2023-07-11 | 深圳市友昆标识制造有限公司 | Urban virtual scene experience management system based on VR technology |
CN116611582A (en) * | 2023-07-14 | 2023-08-18 | 联想新视界(北京)科技有限公司 | Meta universe system of subway station |
CN117172352A (en) * | 2023-07-20 | 2023-12-05 | 南京电力设计研究院有限公司 | Carbon emission optimization design method for digital park building |
CN117475115A (en) * | 2023-11-11 | 2024-01-30 | 华中师范大学 | Path guiding system in virtual-real fusion environment and working method thereof |
CN117557300A (en) * | 2024-01-12 | 2024-02-13 | 湖南大学 | Method and system for deducing business liveness based on energy consumption data of main equipment |
-
2022
- 2022-09-22 CN CN202211156784.8A patent/CN115470707A/en active Pending
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116415756A (en) * | 2023-05-29 | 2023-07-11 | 深圳市友昆标识制造有限公司 | Urban virtual scene experience management system based on VR technology |
CN116415756B (en) * | 2023-05-29 | 2023-10-03 | 深圳市友昆标识制造有限公司 | Urban virtual scene experience management system based on VR technology |
CN116611582A (en) * | 2023-07-14 | 2023-08-18 | 联想新视界(北京)科技有限公司 | Meta universe system of subway station |
CN117172352A (en) * | 2023-07-20 | 2023-12-05 | 南京电力设计研究院有限公司 | Carbon emission optimization design method for digital park building |
CN117172352B (en) * | 2023-07-20 | 2024-04-02 | 南京电力设计研究院有限公司 | Carbon emission optimization design method for digital park building |
CN117475115A (en) * | 2023-11-11 | 2024-01-30 | 华中师范大学 | Path guiding system in virtual-real fusion environment and working method thereof |
CN117557300A (en) * | 2024-01-12 | 2024-02-13 | 湖南大学 | Method and system for deducing business liveness based on energy consumption data of main equipment |
CN117557300B (en) * | 2024-01-12 | 2024-04-05 | 湖南大学 | Method and system for deducing business liveness based on energy consumption data of main equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115470707A (en) | City scene simulation system | |
Zhang | Design and application of fog computing and Internet of Things service platform for smart city | |
CN110083119A (en) | A kind of the visual power system machine room monitoring system and method twin based on number | |
Motieyan et al. | An agent-based modeling approach for sustainable urban planning from land use and public transit perspectives | |
CN106599332A (en) | Three-dimensional digital scheme aided design and display method | |
Gann et al. | Physical–digital integration in city infrastructure | |
CN113704956A (en) | Urban road online microscopic simulation method and system based on digital twin technology | |
CN109472390A (en) | Programme intelligent generation method and system based on machine learning | |
CN105043379A (en) | Scenic spot visiting path planning method and device based on space-time constraint | |
CN108985516B (en) | Indoor path planning method based on cellular automaton | |
Luo et al. | A rule-based city modeling method for supporting district protective planning | |
CN114519932B (en) | Regional traffic condition integrated prediction method based on space-time relation extraction | |
KR102369358B1 (en) | Method, apparatus and system of managing lifecycle of 3 dimensional digital space and contents based on customer needs | |
CN114065348A (en) | Crowd emergency evacuation method, system, terminal and storage medium | |
Li et al. | Construction of smart city street landscape big data-driven intelligent system based on industry 4.0 | |
Lu et al. | Exploring spatial and environmental heterogeneity affecting energy consumption in commercial buildings using machine learning | |
Saad et al. | Role of Cyber-Physical Systems in Smart Cities | |
CN116612633A (en) | Self-adaptive dynamic path planning method based on vehicle-road cooperative sensing | |
CN113554221B (en) | Method for simulating and predicting town development boundary under view angle of' flow space | |
CN108199895A (en) | A kind of intelligent bayonet optimization placement method and device | |
CN114626294A (en) | Planning layout hierarchical generation method for university campus | |
Chmielewski et al. | Hexagonal Zones in Transport Demand Models | |
Yu | Simulation and Application of Urban Road Landscape Based on Geographic Information Data | |
ElBanhawy et al. | Real-time electric mobility simulation in metropolitan areas | |
Liu et al. | [Retracted] Collaboration and Management of Heterogeneous Robotic Systems for Road Network Construction, Management, and Maintenance under the Vision of “BIM+ GIS” Technology |
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 |