CN115131498A - Method for quickly constructing intelligent water conservancy digital twin model of reservoir - Google Patents
Method for quickly constructing intelligent water conservancy digital twin model of reservoir Download PDFInfo
- Publication number
- CN115131498A CN115131498A CN202210643023.9A CN202210643023A CN115131498A CN 115131498 A CN115131498 A CN 115131498A CN 202210643023 A CN202210643023 A CN 202210643023A CN 115131498 A CN115131498 A CN 115131498A
- Authority
- CN
- China
- Prior art keywords
- model
- reservoir
- data
- digital twin
- water conservancy
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Computer Graphics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Geometry (AREA)
- Architecture (AREA)
- Computer Hardware Design (AREA)
- Processing Or Creating Images (AREA)
Abstract
A method for quickly constructing a reservoir intelligent water conservancy digital twin model comprises the following steps: 1) preparing photographic equipment on a reservoir site and preparing corresponding data; 2) sending the obtained data into a deep learning model for training to obtain a rough three-dimensional model of the reservoir site; 3) designing and modifying the three-dimensional model according to actual reservoir site production information, rendering a map, and guiding the map into a digital twin construction tool to construct a reservoir site model; 4) according to the attributes and functions of different equipment models in a scene, modifying the model attributes and events in the corresponding digital twin construction tool to enable the models to receive external signals or send external instructions to control the operation condition of the models; 5) the database is used as an interaction center of virtual and real signals and a storage center for storing user information and historical data of various devices, so that communication interaction of various signals is realized; 6) according to the actual condition of the reservoir site, a reservoir intelligent water conservancy digital twin model is constructed, and the virtual-real combined production management work is completed.
Description
Technical Field
The invention designs a rapid construction method of a reservoir intelligent water conservancy digital twin model.
Background
The water resource safety management system has the advantages of wide China regions, numerous water systems, more points, wide areas, large quantity and complex types of hydraulic engineering, and the rapid development of the economic society puts higher requirements on the water resource safety and the safe and efficient operation of the hydraulic engineering. The water conservancy department has highly paid attention to water conservancy informatization construction and has provided the overall requirement of driving water conservancy modernization with water conservancy informatization.
At present, most of reservoirs depend on a visual management platform in the aspect of water conservancy visual monitoring and are managed based on simple two-dimensional plane maps or character data, the display of water conservancy business data information is not visual enough, areas cannot be checked visually and three-dimensionally, monitoring personnel cannot grasp the actual conditions of the sites in time, and management measures are easy to fail; in addition, with the continuous development of water conservancy informatization, informatization data is exponentially increased, the water conservancy intelligent monitoring effect is not remarkably improved, and the traditional means such as big data mining analysis and the like are more and more difficult to support the processing of water conservancy multi-source heterogeneous data, so that the application value of the data is not fully exerted, and the deep insight requirement of mass data cannot be met; in short, the hydraulic industry is still in its infancy as a whole.
At present, the digitization and the informatization of the water conservancy industry in China are still in the exploration stage, the used technical means are old, and the research and development aim is single. For example, wu yun et al discloses a water conservancy monitoring system and equipment based on the internet of things, which includes a monitoring processing unit, a user side, a big data retrieval unit and a monitoring instrument unit (wu yun, a water conservancy monitoring system and equipment [ M ] based on the internet of things, jiangsu, Nanjing Ninggao information technology Co., Ltd., 2021, CN 113589868A). For another example, thin zhong-wai discloses a water conservancy monitoring method and system based on digital twin, after obtaining the sensing data of the water conservancy station collected by the current sensor, the sensing data is processed by a digital twin reasoning model, so as to obtain the state data of the water conservancy station; after the state data of the water conservancy site is obtained, the running state of the water conservancy site is further displayed in a visual mode through a digital twin display model (in the thin, digital twin-based water conservancy monitoring method and system [ M ], Beijing, Zhike cloud corporation (Beijing) science and technology limited, 2021, CN 113139659A). The technologies such as AI artificial intelligence, AR augmented reality and the like are not widely applied, the water conservancy monitoring intelligent function is not fully displayed, the operation is still manually operated and risks are judged in the business links such as reservoir safety inspection, water conservancy monitoring, event detection and the like, the problems of repeated operation, easiness in being influenced by subjective factors and the like exist, the working efficiency is low, risk points of illegal operation in a working place can only be checked and treated after the fact, the event cannot be effectively and actively early warned in advance, the potential safety hazard of production is reduced, the standard production safety function aiming at the working site needs to be perfected, and the intelligent identification application level needs to be further improved.
Disclosure of Invention
The invention aims to solve the technical problems in the prior art and provides a method for quickly constructing an intelligent water conservancy digital twin model of a reservoir.
The intelligent water conservancy digital twin model of the reservoir is established based on a three-dimensional model tool and deep learning, and signal interaction of a physical world and a virtual world is realized through a database. The invention discloses a rapid construction method of a reservoir intelligent water conservancy digital twin model, which is realized based on a deep learning model capable of rapidly obtaining a model result, wherein the model is rapidly established under the condition of only utilizing image information, a three-dimensional model tool is utilized to refine and perfect the model, and simultaneously, the virtual and real combined work is carried out to provide the functions of real-time monitoring, water level alarming, personnel management and the like, and the rapid construction method is mainly divided into four parts of data preparation, rapid modeling, model optimization, database use and the like.
In order to solve the technical problem, the invention provides a method for quickly constructing an intelligent water conservancy digital twin model of a reservoir, which comprises the following steps of:
step 2, further processing the obtained data, wherein the data needs to be filtered, cut and sequenced, and the data is sent into a deep learning model which can quickly obtain a model result for training so as to obtain a rough three-dimensional model of the reservoir site;
step 3, according to the relevant information of actual reservoir field production, modifying and designing the three-dimensional model by using a three-dimensional model tool, performing operations such as map rendering and the like, and introducing the operations into a digital twin construction tool, so as to construct a vivid reservoir field model;
step 4, designing corresponding model attributes and events in the digital twin construction tool according to attributes and functions of different equipment models in a scene to enable the models to receive external signals or send external instructions, and therefore controlling the operation condition of the models;
step 5, designing a database function as an interaction center of virtual and real signals and a storage center for storing user information and historical data of various devices, realizing communication interaction of various signals and meeting the storage requirement of complex data of a system;
and 6, based on the steps, according to the actual condition of the reservoir site, after a certain picture is taken and the required information is obtained, the site model can be quickly obtained by using the deep learning model, and the intelligent water conservancy digital twin model of the reservoir can be quickly constructed only by certain modification, so that the corresponding virtual and real combined production management work is completed.
Wherein, the step 1 specifically comprises:
preparing corresponding photographic equipment, fixing a camera at one point, and carrying out 360-degree rotation shooting around a scene to be modeled to obtain an original image, depth information and a camera posture;
the photographic equipment used in the method is known as a depth camera. According to different shooting modes, three types of the method can be classified, such as Structured-light (Structured-light), binocular vision (Stereo), and time of flight (TOF). The method needs to acquire three kinds of information, such as an original image, depth information and camera attitude, in the shooting process. The raw data is obtained by a camera and processed in a computer.
Wherein, the step 2 specifically comprises:
2.1. and (3) performing 360-degree rotation shooting on the scene to be modeled in the step 1 to obtain data obtained in behaviors such as original images, depth information and camera postures, and further processing is required at this time. For original images, depth images and camera posture information shot in a certain direction, filtering, cutting and sequencing are needed, and naming is carried out according to a certain naming method;
2.2. after finishing the data set, setting various parameters required by deep learning, and then starting training the model according to preset steps. A coordinate-based scene representation network is first optimized with color frames and corresponding aligned depth frames, while using a hybrid scene representation consisting of implicit surface representation (SDF) and volumetric radiation field. And then extracting triangular meshes from the optimized implicit scene representation when generating the model, and finally obtaining the rough three-dimensional model of the reservoir site.
Wherein, the step 3 specifically comprises:
3.1. the model obtained in the step 2 has defects and errors, and at this time, the model needs to be processed by using a three-dimensional model tool, redundant parts are removed, and meanwhile, several mutually independent parts are separated, so that the subsequent design processing is facilitated.
3.2. And introducing the modified model into a digital twin construction tool, and optimally designing the appearance of the model by using the chartlet rendering function of the tool to make the model more attractive. And meanwhile, the divided models are arranged and designed to meet the actual situation of the site.
Wherein, the step 4 specifically comprises:
and (4) designing the model attributes of the production equipment in reality corresponding to each individual model in the step (3) according to the functions and attributes in reality, and then designing various events, such as click events, creation events and the like, of the model in the digital twin construction tool according to the attributes and the functions of the model.
Aiming at the actual condition of a reservoir site, the above models can be divided into two types of receiving external signals and sending external instructions. The former mainly includes monitoring probe, various sensors, etc., and the latter mainly includes gate, various electronic equipment, etc.
Wherein, the step 5 specifically comprises:
5.1. designing a database function as a reservoir production data storage center, wherein the database table comprises a factory information table, an equipment data table, a log table and a user record table, the factory information table is a main table, and the relations with other tables are all one-to-many correlation; designing a database function as an interaction center of physical signals and virtual signals, storing data such as video signal addresses of monitoring equipment, sensor detection values and the like in data acquisition software in a key-value form in the database, and refreshing data values in real time according to the running condition of a control program;
5.2. the reservoir digital twin system can read data values stored in the database in real time, modify corresponding data through actual signal values, display the data in a scene, and store the running state of the digital twin system into the database in a key-value mode; and the data acquisition software synchronously acquires the running state of the digital twin system from the interaction center and feeds the running state back to the actual control equipment, so that the signal interaction of the physical world and the virtual environment is realized.
Wherein, the step 6 specifically comprises:
the reservoir can be modeled as required only by preparing a set of shooting device, a set of computing device and the algorithm provided by the method. After the field data are shot according to the requirements, the shot field data are sorted and sent into a computing device to be processed through an algorithm to generate a rough model, then the rough model is further processed in a digital twin construction tool to generate a fine model scene, and the mutual exchange of virtual and real signals and the operation and processing of corresponding events are well designed. And finally, generating a digital twin model of the reservoir by combining the actual situation of the production field.
The invention discloses a rapid construction method of a digital twin model for intelligent water conservancy of a reservoir, which is realized based on a deep learning model capable of rapidly obtaining a model result, wherein the model is rapidly established under the condition of only utilizing image information, a digital twin construction tool is utilized to refine and perfect the model, and simultaneously, the virtual and real combined work is carried out to provide the functions of real-time monitoring, water level alarming, personnel management and the like, and the rapid construction method is mainly divided into four parts of data preparation, rapid modeling, model optimization, database use and the like. Firstly, preparing corresponding photographic equipment, fixing a camera at one point, carrying out 360-degree rotation shooting around a scene to be modeled, and obtaining detailed information such as an original image, depth information, camera posture and the like after subsequent processing of a computer. Further processing is then required at this point on the data obtained. For original images, depth images and camera posture information shot in a certain direction, filtering, cutting and sequencing are needed, naming is conducted according to a certain naming method, after a data set is sorted, all parameters needed by deep learning need to be set, and then training of a model is started according to preset steps. A coordinate-based scene representation network is first optimized with color frames and corresponding aligned depth frames, while using a hybrid scene representation consisting of implicit surface representation (SDF) and volumetric radiation field. And then when the model is generated, extracting triangular meshes from the optimized implicit scene representation, and finally obtaining the rough three-dimensional model of the reservoir site. Then, because the obtained model has defects and errors, the model needs to be processed by a three-dimensional model tool at the moment, redundant parts are removed, and meanwhile, several mutually independent parts are separated, so that the subsequent processing is facilitated. And (3) introducing the modified model into a digital twin construction tool, and optimizing the appearance of the model by using the mapping rendering function of the tool so as to make the model more attractive. And meanwhile, the segmented models are arranged to meet the actual situation of the site. In addition, the actual production equipment corresponding to each individual model designs the model attribute according to the actual function and attribute, and then designs various events, such as click events, creation events and the like, of the model in the digital twin construction tool according to the attribute and the function thereof. Aiming at the actual condition of a reservoir site, the above models can be divided into two types of receiving external signals and sending external instructions. The former mainly includes monitoring probe, various sensors, etc., and the latter mainly includes gate, various electronic equipment, etc. Two databases with different functions are combined and designed to store and manage data of the system, and a script is compiled to realize information interaction between the digital twin model and the databases. In conclusion, the method provides a simple and convenient reservoir intelligent water conservancy digital twin model rapid construction method, realizes rapid reservoir construction site based on a digital twin construction tool and deep learning capable of rapidly obtaining a model result, and can facilitate enterprises to perform safety management and production on the reservoir.
The invention has the advantages that: only simple image data is needed in the data preparation part, and other complicated steps are not needed; after the data set is simply processed, the data set is directly sent to a deep model learning training, and a relatively perfect reservoir model can be obtained; after the models are simply processed in a three-dimensional model tool and a digital twin construction tool respectively, the practical application of virtual-real combination can be supported; two databases with different functions are designed and used simultaneously, so that user information and historical data of reservoir equipment can be stored, interaction of virtual and real signals is realized, and the storage requirement of complex data of the intelligent water conservancy digital twin model system of the reservoir is met.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is within the scope of the present invention for those skilled in the art to obtain other drawings based on the drawings without inventive exercise.
Fig. 1 is a structural block diagram of a method for rapidly modeling a reservoir site according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a data set in the method for rapidly modeling a reservoir site according to the embodiment of the present invention, and the diagram is an original image captured by a camera.
Fig. 3 is a deep learning algorithm diagram in the method for rapidly modeling the reservoir site according to the embodiment of the present invention.
Fig. 4 is a schematic diagram of a model modeled from example data in the method for rapidly modeling a reservoir site according to the embodiment of the present invention, which is processed in a digital twin construction tool.
Fig. 5 is a relational diagram of an actual scene connected by a database and a digital twin model in the method for rapidly modeling a reservoir site according to the embodiment of the present invention.
Detailed Description
The present invention is further described in detail below with reference to the accompanying drawings.
As shown in fig. 1 to 4, for the embodiment of the present invention, a method for quickly constructing a digital twin model for intelligent water conservancy of a reservoir is provided, which is implemented based on a deep learning model capable of quickly obtaining a model result, and is implemented by quickly establishing a model only using image information, designing and refining the model by using a three-dimensional model tool and a digital twin construction tool, and performing virtual-real combination work to provide functions such as real-time monitoring, water level alarm, personnel management, and the like.
The data preparation mainly includes that corresponding photographic equipment is prepared, a camera is fixed at one point, 360-degree rotation shooting is carried out around a scene needing modeling, and detailed information such as an original image, depth information and a camera posture are obtained after subsequent processing of a computer. Further processing is then required at this point on the data obtained. For original images, depth images and camera posture information shot in a certain direction, filtering, cutting and sequencing are needed, and naming is carried out according to a certain naming method.
The rapid modeling mainly refers to setting various parameters required by deep learning after finishing sorting the data set, and then starting training the model according to preset steps. A coordinate-based scene representation network is first optimized with color frames and corresponding aligned depth frames, while using a hybrid scene representation consisting of implicit surface representation (SDF) and volumetric radiation field. And then when the model is generated, extracting triangular meshes from the optimized implicit scene representation, and finally obtaining the rough three-dimensional model of the reservoir site.
The optimization model mainly refers to the fact that a three-dimensional model tool is used for processing the model, redundant parts are removed, and meanwhile, the independent parts are separated, so that subsequent design processing is facilitated. And introducing the modified model into a digital twin construction tool, and optimally designing the appearance of the model by using the chartlet rendering function of the tool to make the model more attractive. And meanwhile, the divided models are arranged and designed to meet the actual situation of the site. In addition, the actual production equipment corresponding to each individual model designs the model attribute according to the actual function and attribute, and then designs various events, such as click events, creation events and the like, of the model in the digital twin construction tool according to the attribute and the function thereof. Aiming at the actual condition of a reservoir site, the above models can be divided into two types of receiving external signals and sending external instructions. The former mainly includes monitoring probe, various sensors, etc., and the latter mainly includes gate, various electronic equipment, etc.
The database use mainly refers to the design and use of two databases with different functions to store and manage data of the system, and meanwhile, scripts are written to realize information interaction between the digital twin model and the databases. Designing a database function as a reservoir production data storage center, wherein the database table comprises a factory information table, an equipment data table, a log table and a user record table, the factory information table is a main table, and the relations with other tables are all one-to-many correlation; the function of a database is designed to serve as an interaction center of physical signals and virtual signals, data such as video signal addresses of monitoring equipment, sensor detection values and the like in data acquisition software are stored in the database in a key-value mode, and data values are refreshed in real time according to the running condition of a control program. The reservoir digital twin system reads data values stored in the database in real time, modifies corresponding data through actual signal values, displays the data in a scene, and stores the running state of the digital twin system in the database in a key-value mode; and the data acquisition software synchronously acquires the running state of the digital twin system from the interaction center and feeds the running state back to the actual control equipment, so that the signal interaction of the physical world and the virtual environment is realized.
Finally, all functional modules are combined to form a reliable, concise and easy-to-use rapid construction method for the intelligent water conservancy digital twin model of the reservoir, a user only needs to prepare a set of shooting device, a set of computing device and an algorithm provided by the method, after field data is shot according to requirements, the field data is sorted and sent into the computing device to be processed through the algorithm to generate a rough model, then the rough model is further processed in a digital twin construction tool to generate a fine model scene, and the interchange of virtual and real signals and the operation and processing of corresponding events are well designed. And finally, generating a digital twin model of the reservoir by combining the actual situation of the production field.
As shown in fig. 3, a deep learning algorithm in the method for rapidly modeling a reservoir site according to the embodiment of the present invention is further described:
the method employs an optimization-based model generation method for geometric reconstruction from the RGB-D sequence of a consumer-grade depth camera. The method utilizes N color frames and corresponding aligned depth frames to optimize a coordinate-based scene representation network while utilizing a volume rendering of NeRF, using a hybrid scene representation consisting of an implicit surface representation (SDF) and a volumetric radiation field. The method incorporates depth measurements therein to enable robust and metric 3D reconstruction. The method uses marching cubes to extract triangular meshes from the optimized implicit scene representation at evaluation.
Specifically, the method comprises the steps of firstly importing parameter setting and a prepared reservoir scene data set, and then initializing a deep learning model and an optimizer. After the above steps are completed, the method firstly generates RGBD ray data to prepare a more complete front-end step for the following model training. Then, the method randomly divides the data set and then trains the deep learning model. And after the training reaches a certain number of times, a rough reservoir site 3D model is generated firstly, if the use requirement is met or no progress is made compared with the previously generated model, the method completes the model generation, otherwise, the deep learning model is trained continuously.
The above is the case of the whole invention, the construction of the model in the digital twin construction tool has highly realistic visualization; the requirement of the deep learning model for the data set, which can quickly obtain the model result, simplifies the preparation difficulty; the use of the three-dimensional model tool and the digital twinning construction tool facilitates rapid processing of the missing parts of the obtained model; the use of two databases with different functions meets the requirement of a system on complex data storage, and realizes the interaction of virtual and real signals and the persistence of service data.
The embodiment of the invention has the following beneficial effects:
according to the method for rapidly modeling on the reservoir site, only simple image data is needed in the data preparation part, and other complex steps are not needed; after the data set is simply processed, the data set is directly sent into a deep model learning training device capable of quickly obtaining a model result, and a relatively perfect reservoir model can be obtained; after the model is simply processed in a three-dimensional model tool and a digital twin construction tool, the practical application of virtual-real combination can be supported; meanwhile, two databases with different functions are adopted, so that the interaction of user information, historical data of reservoir equipment and virtual and real signals can be stored, and the storage requirement of complex data of the intelligent water conservancy digital twin model system of the reservoir is met.
The embodiments described in this specification are merely illustrative of implementation forms of the inventive concept, and the scope of the present invention should not be considered limited to the specific forms set forth in the embodiments, but also equivalent technical means that can be conceived by one skilled in the art based on the inventive concept.
Claims (7)
1. A method for quickly constructing a reservoir intelligent water conservancy digital twin model comprises the following steps:
step 1, preparing photographic equipment on a reservoir site, carrying out 360-degree rotary shooting around a scene to be modeled according to the requirements of a required data set, and preparing corresponding data;
step 2, filtering, cutting and sequencing the obtained data, and sending the data into a deep learning model which can quickly obtain a model result for training so as to obtain a rough three-dimensional model of the reservoir site;
step 3, modifying and designing the three-dimensional model by using a three-dimensional model tool according to relevant information of actual reservoir field production, performing chartlet rendering, and introducing the chartlet rendering into a digital twin construction tool, so as to construct a vivid reservoir field model;
step 4, designing model attributes and events in corresponding digital twin construction tools according to attributes and functions of different equipment models in a scene to enable the models to receive external signals or send external instructions, and accordingly controlling the operation conditions of the models;
step 5, designing a database function as an interaction center of virtual and real signals and a storage center for storing user information and historical data of various devices, realizing communication interaction of various signals and meeting the storage requirement of complex data of a system;
and 6, based on the steps, according to the actual condition of the reservoir site, after the picture is shot and the required information is obtained, the deep learning model is used for quickly obtaining the site model, the intelligent water conservancy digital twin model of the reservoir is quickly constructed through modification, and the corresponding virtual and real combined production management work is completed.
2. The method for quickly constructing the intelligent water conservancy digital twin model of the reservoir as claimed in claim 1, wherein the step 2 specifically comprises:
2.1. performing 360-degree rotation shooting on the scene to be modeled in the step 1 to obtain data obtained in behaviors such as an original image, depth information and a camera posture, and further processing is required at the moment; for original images, depth images and camera posture information shot in a certain direction, filtering, cutting and sequencing are needed, and naming is carried out according to a certain naming method;
2.2. after finishing the data set, setting various parameters required by deep learning, and then starting to train the model according to preset steps; first optimizing a coordinate-based scene representation network with color frames and corresponding aligned depth frames while using a hybrid scene representation consisting of implicit surface representation (SDF) and volumetric radiation field; and then when the model is generated, extracting triangular meshes from the optimized implicit scene representation, and finally obtaining the rough three-dimensional model of the reservoir site.
3. The method for quickly constructing the intelligent water conservancy digital twin model of the reservoir as claimed in claim 1, wherein the step 3 specifically comprises:
3.1. the model obtained in the step 2 has defects and errors, and at the moment, the model needs to be processed by a three-dimensional model tool, redundant parts are removed, and meanwhile, several mutually independent parts are separated, so that the subsequent design processing is facilitated;
3.2. importing the modified model into a digital twin construction tool, and optimally designing the appearance of the model by using the chartlet rendering function of the tool to make the model more attractive; and meanwhile, the divided models are arranged and designed to meet the actual situation of the site.
4. The method for rapidly constructing the intelligent water conservancy digital twin model of the reservoir as claimed in claim 1, wherein the step 4 comprises:
aiming at the actual production equipment corresponding to each individual model in the step 3, designing the model attribute according to the actual function and attribute, and then designing various events of the model in the digital twin construction tool according to the attribute and the function;
aiming at the actual condition of a reservoir site, the models can be divided into two types of receiving external signals and sending external instructions; the former includes monitoring probe, various sensors, etc. and the latter includes gate and various electronic equipment.
5. The method as claimed in claim 1, wherein the step 5 comprises:
5.1. designing a database function as a reservoir production data storage center, wherein the database table comprises a factory information table, an equipment data table, a log table and a user record table, the factory information table is a main table, and the relations with other tables are all one-to-many correlation; designing a database function as an interaction center of physical signals and virtual signals, storing data such as video signal addresses of monitoring equipment, sensor detection values and the like in data acquisition software in a key-value form in the database, and refreshing data values in real time according to the running condition of a control program;
5.2. the reservoir digital twin system can read data values stored in the database in real time, modify corresponding data through actual signal values, display the data in a scene, and store the running state of the digital twin system into the database in a key-value mode; and the data acquisition software synchronously acquires the running state of the digital twin system from the interaction center and feeds the running state back to the actual control equipment, so that the signal interaction of the physical world and the virtual environment is realized.
6. The method for rapidly constructing the intelligent water conservancy digital twin model of the reservoir as claimed in claim 1, wherein the step 6 specifically comprises:
the reservoir can be modeled as required only by preparing a set of shooting device, a set of computing device and the algorithm provided by the method; after the field data is shot according to the requirements, the field data is sorted and sent into a computing device to generate a rough model through algorithm processing, then the rough model is further processed in a digital twin construction tool to generate a fine model scene, and the mutual exchange of virtual and real signals and the operation and processing of corresponding events are well designed; and finally, generating a digital twin model of the reservoir by combining the actual situation of the production field.
7. The method for rapidly constructing the intelligent water conservancy digital twin model of the reservoir as claimed in claim 3, wherein the step 2.2 specifically comprises:
designing an optimization-based model generation method for performing geometric reconstruction from an RGB-D sequence of a consumer-grade depth camera; optimizing a coordinate-based scene representation network with N color frames and corresponding aligned depth frames while volume rendering with NeRF, using a hybrid scene representation consisting of implicit surface representation (SDF) and volumetric radiation field; depth measurements are included to enable robust and metric 3D reconstruction; extracting triangular meshes from the optimized implicit scene representation using a marching cube when generating the model;
specifically, firstly, parameter setting and a prepared reservoir scene data set are imported, and then a deep learning model and an optimizer are initialized; after the steps are completed, firstly generating RGBD ray data to prepare a more complete pre-step for the following model training; then, randomly dividing the data set, and then starting training a deep learning model; and after the training reaches a certain number of times, firstly generating a rough reservoir site 3D model, if the use requirement is met or no progress is made compared with the previously generated model, completing the model generation, and otherwise, returning to continuously train the deep learning model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210643023.9A CN115131498A (en) | 2022-06-08 | 2022-06-08 | Method for quickly constructing intelligent water conservancy digital twin model of reservoir |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210643023.9A CN115131498A (en) | 2022-06-08 | 2022-06-08 | Method for quickly constructing intelligent water conservancy digital twin model of reservoir |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115131498A true CN115131498A (en) | 2022-09-30 |
Family
ID=83377253
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210643023.9A Pending CN115131498A (en) | 2022-06-08 | 2022-06-08 | Method for quickly constructing intelligent water conservancy digital twin model of reservoir |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115131498A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115688227A (en) * | 2022-10-13 | 2023-02-03 | 长江空间信息技术工程有限公司(武汉) | Digital twin hydraulic engineering operation safety monitoring system and operation method |
| CN115937482A (en) * | 2022-11-24 | 2023-04-07 | 西南交通大学 | Holographic scene dynamic construction method and system capable of adapting to screen size |
| CN117372629A (en) * | 2023-12-07 | 2024-01-09 | 山东圣瑞信息技术有限公司 | Reservoir visual data supervision control system and method based on digital twinning |
-
2022
- 2022-06-08 CN CN202210643023.9A patent/CN115131498A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115688227A (en) * | 2022-10-13 | 2023-02-03 | 长江空间信息技术工程有限公司(武汉) | Digital twin hydraulic engineering operation safety monitoring system and operation method |
| CN115937482A (en) * | 2022-11-24 | 2023-04-07 | 西南交通大学 | Holographic scene dynamic construction method and system capable of adapting to screen size |
| CN115937482B (en) * | 2022-11-24 | 2023-09-15 | 西南交通大学 | Holographic scene dynamic construction method and system for self-adapting screen size |
| CN117372629A (en) * | 2023-12-07 | 2024-01-09 | 山东圣瑞信息技术有限公司 | Reservoir visual data supervision control system and method based on digital twinning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115131498A (en) | Method for quickly constructing intelligent water conservancy digital twin model of reservoir | |
| Shan et al. | Research on landscape design system based on 3D virtual reality and image processing technology | |
| CN104765280B (en) | The three-dimensional aobvious control comprehensive training system of no-manned plane three-dimensional | |
| CN112084556A (en) | Lightweight display method and system for mass data BIM model | |
| CN108919944A (en) | A kind of virtual roaming method carrying out data lossless interaction in display end based on digital city model realization | |
| Rhyne | Going virtual with geographic information and scientific visualization | |
| CN109255743A (en) | Conserve total management system | |
| CN115222792B (en) | Digital twin modeling method for railway bridge | |
| Zhong et al. | The application of virtual reality technology in the digital preservation of cultural heritage | |
| CN116229019A (en) | Digital twinning-oriented large-scene fusion three-dimensional reconstruction method and system | |
| CN114372107A (en) | GIS-based method and system for visualizing homeland improvement and ecological restoration data | |
| CN109584668A (en) | A kind of rock tunnel(ling) machine training platform based on virtual reality and big data | |
| CN104268941A (en) | Hot spot forming method and device for simulated three-dimensional map | |
| CN109615700A (en) | Three-dimension virtual reality tunnel monitoring system and its operating method | |
| CN108805978A (en) | A kind of automatically generating device and method based on deep learning threedimensional model | |
| CN105574930B (en) | A kind of Chinese character three-dimensional data Collection ＆ Processing System | |
| CN115035225B (en) | Battlefield threat assessment warning method based on OSG | |
| Zang et al. | Virtual reality and the application in virtual experiment for agricultural equipment | |
| CN115546416A (en) | Web-based lightweight 3D visualization method and system | |
| CN114529689A (en) | Ceramic cup defect sample amplification method and system based on antagonistic neural network | |
| DE102021119725A1 (en) | OVERLAYING METADATA ONTO VIDEO STREAM ON-DEMAND FOR INTELLIGENT VIDEO ANALYTICS | |
| CN208225113U (en) | A kind of data acquisition device for the simulation of VR device emulation | |
| CN113848750A (en) | Two-wheeled robot simulation system and robot system | |
| CN113194140A (en) | Integrated remote monitoring system based on fire-fighting robot | |
| Kaspar et al. | Holographic mixed reality: an enhanced technology for visualizing and evaluating complex 3D geologic data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |