CN116384255B - Park dangerous situation perception method and system based on multi-source data fusion - Google Patents
Park dangerous situation perception method and system based on multi-source data fusion Download PDFInfo
- Publication number
- CN116384255B CN116384255B CN202310524192.5A CN202310524192A CN116384255B CN 116384255 B CN116384255 B CN 116384255B CN 202310524192 A CN202310524192 A CN 202310524192A CN 116384255 B CN116384255 B CN 116384255B
- Authority
- CN
- China
- Prior art keywords
- park
- dangerous
- data
- model
- virtual
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000004927 fusion Effects 0.000 title claims abstract description 23
- 230000008447 perception Effects 0.000 title claims abstract description 11
- 238000004088 simulation Methods 0.000 claims abstract description 62
- 238000003860 storage Methods 0.000 claims description 22
- 238000005259 measurement Methods 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 239000000126 substance Substances 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 238000007637 random forest analysis Methods 0.000 claims description 4
- 238000009434 installation Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 abstract description 17
- 238000004458 analytical method Methods 0.000 abstract description 4
- 230000000875 corresponding effect Effects 0.000 description 39
- 230000008569 process Effects 0.000 description 12
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 239000010865 sewage Substances 0.000 description 7
- 238000012545 processing Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 208000025274 Lightning injury Diseases 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000002079 cooperative effect Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 230000005180 public health Effects 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
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
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Geometry (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The application provides a park dangerous situation awareness method and system based on multi-source data fusion, and relates to the technical field of intelligent analysis. The method comprises the following steps: and constructing a virtual park model according to the building environment information of the target park and the historical operation conditions. And performing simulation operation by using the virtual park model to obtain simulation data. And analyzing the simulated simulation data through the trained AI model to obtain the dangerous index and the index weight. And acquiring heterogeneous data according to the dangerous index. And according to the ground surface feature information, the real scene map data is divided, and a node uniform network is obtained. And weighting and superposing the measured data of any grid with the index weight to obtain a dangerous situation value, and combining the dangerous situation values of all grids to obtain the dangerous situation data of the target park. The dangerous situation of each area is determined according to the types of the parks and the distribution of the ground surface features, and the comprehensive perception of the dangerous situation of the parks is realized.
Description
Technical Field
The application relates to the technical field of intelligent analysis, in particular to a park dangerous situation perception method and system based on multi-source data fusion.
Background
A campus refers to a standard building or building group that is generally planned and constructed by government (civil enterprises and government cooperatives), has complete water supply, power supply, air supply, communication, road, storage and other supporting facilities, is reasonably arranged, and can meet the requirements of production and scientific experiments in a specific industry, including industrial parks, logistical parks, metropolitan industrial parks, scientific parks, creative parks, and the like.
Existing campus hazard situation analysis generally relies on sensors to monitor whether there is a hazard in the campus to ensure safety in the campus. However, this method can only detect the potential safety hazard exceeding the safety standard when the potential safety hazard factor in the park has exceeded the safety standard, and cannot comprehensively obtain the dangerous situation of the park. And the prior art can only sense the dangerous situation for a certain type of park. In fact, due to the different types of parks, the applicable scenarios are different, as are the hazards presented by the parks. If the dangerous situation of each area cannot be determined according to the types of the parks and the distribution of the ground surface features, the comprehensive perception of the dangerous situation of the parks cannot be realized.
Disclosure of Invention
The application aims to provide a park dangerous situation sensing method and system based on multi-source data fusion, which are used for solving the problem that the overall sensing of the park dangerous situation cannot be realized in the prior art if the dangerous situation of each area cannot be determined according to the park type and the ground surface feature distribution.
In order to solve the technical problems, the application adopts the following technical scheme:
in a first aspect, an embodiment of the present application provides a method for sensing a park dangerous situation based on multi-source data fusion, including the following steps:
acquiring building environment information and a plurality of historical operating conditions of a target park;
constructing a virtual park model corresponding to the target park according to the building environment information and the plurality of historical operation conditions;
obtaining park operation simulation data matched with the park type of the target park, inputting the park operation simulation data into a virtual park model for simulation operation, and obtaining simulation data;
inputting the simulated data into a trained AI model to obtain corresponding dangerous indexes and index weights;
according to the dangerous index, acquiring various heterogeneous data of a target park, wherein the various heterogeneous data comprise surface ground feature information, live-action map data and actual measurement data;
uniformly dividing live-action map data into a plurality of grids, calculating the number of nodes in each grid according to ground surface feature information, and iteratively dividing grids with the number of nodes being greater than a preset node number threshold value based on the number of nodes of each grid until the number of nodes in all grids is smaller than the preset node number threshold value to obtain a node uniform network;
aiming at any grid in the node uniform network, weighting and superposing measured data corresponding to the grid and index weights to obtain a dangerous situation value of the grid;
and combining the dangerous situation values of all grids to obtain dangerous situation data of the target park.
In the present application, further, the step of constructing the virtual park model corresponding to the target park according to the building environment information and the plurality of historical operation conditions includes:
calling a plurality of virtual models according to building environment information, wherein the plurality of virtual models at least comprise a building model, a road model and a scene model;
introducing various virtual models into the UE system, and constructing a virtual park initial model by utilizing the various virtual models according to building environment information;
establishing a virtual park neural network model based on the virtual park initial model by combining a neural network algorithm;
all the historical operating conditions are input into the virtual park neural network model for training, and the virtual park model matched with the target park is obtained.
In the present application, further, the step of importing a plurality of virtual models into the UE system and constructing the virtual park initial model using the plurality of virtual models according to the building environment information includes:
according to the building environment information, arranging a building model and a road model in a scene model;
and splicing the multiple virtual models to obtain the initial model of the virtual park.
In the present application, further, the step of inputting the simulated simulation data into the trained AI model to obtain the corresponding risk index and the index weight includes:
if the type of the target park is a chemical industry park, the obtained dangerous index is the path track of the dangerous goods transport vehicle and the configuration information of the dangerous goods sensor;
acquiring real-time monitoring data of the dangerous article sensor according to configuration information of the dangerous article sensor;
according to real-time monitoring data of the dangerous goods sensor, calculating the weight of the sensor by combining the geographic position of the dangerous goods sensor;
and calculating the weight of the dangerous goods transportation vehicle according to the path track of the dangerous goods transportation vehicle.
In the present application, further, before the step of obtaining the dangerous index as the path track of the dangerous goods transport vehicle and the configuration information of the dangerous goods sensor if the type of the target park is the chemical park, the method further includes:
acquiring basic information of all dangerous goods storage tanks, wherein the basic information of any dangerous goods storage tank at least comprises one or more of enterprises to which the dangerous goods storage tank belongs, the types of stored dangerous goods and installation positions;
the dangerous article sensor is configured for any dangerous article storage tank, and the configuration information of the dangerous article sensor is input according to the basic information of the dangerous article storage tank, wherein the configuration information comprises a sensor number, a sensor leakage alarm threshold value and the basic information of the corresponding dangerous article storage tank.
In the present application, before the step of inputting the simulated simulation data into the trained AI model, the method further includes:
establishing an AI model;
acquiring a plurality of samples, wherein the plurality of samples comprise historical dangerous index data of various parks;
the AI model is trained using a plurality of samples to obtain a trained AI model.
In the present application, further, the step of establishing an AI model includes:
and constructing an AI model through a random forest algorithm and a convolutional neural network algorithm.
In a second aspect, an embodiment of the present application provides a system for sensing a campus hazard situation based on multi-source data fusion, including:
the target park acquisition module is used for acquiring building environment information and a plurality of historical operating conditions of the target park;
the virtual park model building module is used for building a virtual park model corresponding to the target park according to the building environment information and the historical operation conditions;
the simulation operation module is used for acquiring park operation simulation data matched with the park type of the target park, inputting the park operation simulation data into the virtual park model for simulation operation, and obtaining simulation data;
the dangerous index obtaining module is used for inputting the simulated simulation data into the trained AI model to obtain corresponding dangerous indexes and index weights;
the heterogeneous data acquisition module is used for acquiring various heterogeneous data of the target park according to the dangerous indexes, wherein the various heterogeneous data comprise ground surface feature information, live-action map data and actual measurement data;
the grid dividing module is used for uniformly dividing the live-action map data into a plurality of grids, calculating the number of nodes in each grid according to the ground surface feature information, and iteratively dividing the grids with the number of nodes being larger than a preset node number threshold value based on the number of nodes of each grid until the number of nodes in all the grids is smaller than the preset node number threshold value to obtain a node uniform network;
the weighting module is used for carrying out weighted superposition on the measured data corresponding to the grids and the index weight aiming at any grid in the node uniform network to obtain a dangerous situation value of the grid;
and the dangerous situation value combination module is used for combining the dangerous situation values of all grids to obtain dangerous situation data of the target park.
Compared with the prior art, the embodiment of the application has at least the following advantages or beneficial effects:
the application provides a park dangerous situation awareness method and system based on multi-source data fusion, comprising the following steps: and re-engraving the target park according to the building environment information of the target park and a plurality of historical operation conditions to construct a virtual park model matched with the target park. The operation process of the target park can be simulated through the virtual park model. Then, the virtual park model is utilized to perform simulation operation through park operation simulation data matched with the park type of the target park, and simulation data are obtained. And analyzing the simulated simulation data through the trained AI model to obtain the dangerous indexes corresponding to the target park and the index weights corresponding to the dangerous indexes. And acquiring various heterogeneous data of the target park according to the dangerous index, wherein the various heterogeneous data comprise ground surface feature information, live-action map data and measured data. Thereby avoiding acquiring excessive invalid data while reducing the acquired data volume. And then uniformly dividing the live-action map into a plurality of grids, calculating the number of nodes in each grid according to the ground surface feature information in the process of carrying out iterative division on the live-action map, and if the number of nodes in any grid is larger than a preset node number threshold value, continuing iterative division on the grids until the number of nodes in all grids is smaller than the preset node number threshold value, thereby completing division on the live-action map and obtaining a node uniform network. The situation that the number of nodes of each grid is unbalanced due to the difference of the surface ground object spatial distribution and the difference of the space object graph complexity is avoided, and the problem that the grid can not be uniformly divided due to the unbalanced number of the nodes of each grid is avoided. And finally, aiming at any grid in the node uniform network, carrying out weighted superposition on measured data corresponding to the grid and index weights to obtain dangerous situation values of the grid, and combining the dangerous situation values of all the grids to obtain the dangerous situation data of the target park. Therefore, the dangerous situation of each area is determined according to the types of the parks and the distribution of the ground surface features, and the comprehensive perception of the dangerous situation of the parks is realized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a park dangerous situation awareness method based on multi-source data fusion provided by an embodiment of the application;
fig. 2 is a block diagram of a park dangerous situation awareness system based on multi-source data fusion according to an embodiment of the present application;
fig. 3 is a schematic block diagram of an electronic device according to an embodiment of the present application.
Icon: 110-a target park acquisition module; 120-a virtual park model building module; 130-a simulation run module; 140, a dangerous index obtaining module; 150-a heterogeneous data acquisition module; 160-meshing module; 170-a weighting module; 180-a dangerous situation value combination module; 101-memory; 102-a processor; 103-communication interface.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Referring to fig. 1, fig. 1 is a flowchart of a method for sensing a dangerous situation of a park based on multi-source data fusion according to an embodiment of the present application. The embodiment of the application provides a park dangerous situation perception method based on multi-source data fusion, which comprises the following steps:
s110: acquiring building environment information and a plurality of historical operating conditions of a target park;
illustratively, the building environment information includes building structures, road facilities, and the like.
Wherein any historical operating condition of the target campus characterizes asset management operating data, people flow data, traffic flow data, power equipment operating condition of each building, etc. of the target campus in a certain time period.
S120: constructing a virtual park model corresponding to the target park according to the building environment information and the plurality of historical operation conditions;
in some implementations of this embodiment, the step of constructing the virtual campus model corresponding to the target campus according to the building environment information and the plurality of historical operating conditions includes: and calling a plurality of virtual models according to the building environment information, wherein the plurality of virtual models at least comprise a building model, a road model and a scene model. And importing the multiple virtual models into the UE system, and constructing a virtual park initial model by utilizing the multiple virtual models according to the building environment information. And combining a neural network algorithm, and establishing a virtual park neural network model based on the virtual park initial model. All the historical operating conditions are input into the virtual park neural network model for training, and the virtual park model matched with the target park is obtained. The UE system is a virtual engine (UE), and is formed by combining software such as a mature blueprint technology, materials, modeling and the like.
Specifically, a virtual model consistent with building environment information is invoked from modeling software. And importing all the virtual models into the UE system, and laying and splicing all the virtual models according to the building environment information so as to revise the actual building environment of the target park, thereby constructing the initial model of the virtual park. And then inputting the neural network algorithm into the virtual park initial model to obtain the virtual park neural network model. And finally, training the virtual park neural network model by utilizing all the historical operating conditions to obtain the virtual park model matched with the target park. The operation process of the target park can be simulated through the virtual park model.
S130: obtaining park operation simulation data matched with the park type of the target park, inputting the park operation simulation data into a virtual park model for simulation operation, and obtaining simulation data;
among other types of parks, industrial parks, chemical parks, industrial parks, logistic parks, and metropolitan industrial parks may be included.
For example, if the campus type of the target campus is an industrial campus, historical operating conditions of a plurality of industrial parks are acquired as campus operation simulation data, and a virtual campus model is simulated by using the campus operation simulation data to obtain simulation data consistent with the target campus. The historical operating conditions may include asset management operating data, people flow data, traffic flow data, power equipment operating conditions for each building, and the like, among others.
S140: inputting the simulated data into a trained AI model to obtain corresponding dangerous indexes and index weights;
specifically, the simulated simulation data is analyzed through the trained AI model, and the dangerous indexes corresponding to the target park and the index weights corresponding to the dangerous indexes are obtained.
S150: according to the dangerous indexes, acquiring various heterogeneous data of the target park, wherein the various heterogeneous data comprise surface ground feature information, live-action map data and actual measurement data corresponding to the dangerous indexes;
specifically, according to the dangerous index, multiple heterogeneous data of the target park are acquired in a targeted mode. Thereby avoiding acquiring excessive invalid data while reducing the acquired data volume.
For example, if the target park is an industrial park, the simulated data obtained in S130 characterizes the operation of the target park, and the trained AI model is used to analyze the simulated data to obtain a risk index of the target park, where the risk index may include various on-park project risks (such as risks in terms of construction electricity, branch project processes, construction machinery operation processes, temporary construction facilities, material stacking and handling, foundation treatment, etc.) under various natural risks (such as safety risks in terms of earthquake, lightning strike, atmospheric temperature, precipitation, public health, etc.), various public supporting facility risks (such as risks in terms of sewage treatment systems, power supply and distribution systems, communication systems, heating systems, gas supply systems, logistics transportation systems, public piping lane systems, greenbelt systems, sanitation systems, fire protection systems, etc.), various dangerous chemical risks, and the like. If the AI model analysis obtains that the dangerous indexes comprise a sewage treatment system under rainfall weather and a power supply and distribution system under lightning stroke disaster, namely the sewage treatment system under rainfall weather or the power supply and distribution system under lightning stroke disaster can cause the safety risk of a target park. The actual measurement data corresponding to the hazard index is 1 if the sewage treatment system under rainfall weather exists when the hazard situation data of the target park is calculated, and the event is true. Or if the situation of the power supply and distribution system under the lightning stroke disaster exists when the dangerous situation data of the target park is calculated, the actual measurement data is 1, and the event is true. Otherwise, if no event corresponding to the danger index occurs when the danger situation data of the target park is calculated, the actual measurement data corresponding to the danger index is 0.
S160: uniformly dividing live-action map data into a plurality of grids, calculating the number of nodes in each grid according to ground surface feature information, and iteratively dividing grids with the number of nodes being greater than a preset node number threshold value based on the number of nodes of each grid until the number of nodes in all grids is smaller than the preset node number threshold value to obtain a node uniform network;
specifically, in the process of carrying out iterative division on the live-action map data, according to the ground surface feature information, the number of nodes in each grid is calculated, if the number of nodes in any grid is larger than a preset node number threshold value, the grid continues to carry out iterative division until the number of nodes in all grids is smaller than the preset node number threshold value, and therefore division on the live-action map is completed, and a node uniform network is obtained. The situation that the number of nodes of each grid is unbalanced due to the difference of the surface ground object spatial distribution and the difference of the space object graph complexity is avoided, and the problem that the grid can not be uniformly divided due to the unbalanced number of the nodes of each grid is avoided.
S170: aiming at any grid in the node uniform network, weighting and superposing measured data corresponding to the grid and index weights to obtain a dangerous situation value of the grid;
by way of example, suppose that the hazard index is a sewage treatment system in rainfall weather and a power supply and distribution system in lightning strike disaster, and that the index weight of the sewage treatment system in rainfall weather is 3% and the index weight of the power supply and distribution system in lightning strike disaster is 4%. Then, for any grid, if the grid is in the presence of a sewage treatment system in rainfall weather and is not in the presence of a power supply and distribution system in lightning strike disaster, the risk situation value of the grid is 1x3% +0x4% = 0.03.
S180: and combining the dangerous situation values of all grids to obtain dangerous situation data of the target park.
In the implementation process, the method first carries out resculpting on the target park according to the building environment information and a plurality of historical operation conditions of the target park so as to construct a virtual park model matched with the target park. The operation process of the target park can be simulated through the virtual park model. Then, the virtual park model is utilized to perform simulation operation through park operation simulation data matched with the park type of the target park, and simulation data are obtained. And analyzing the simulated simulation data through the trained AI model to obtain the dangerous indexes corresponding to the target park and the index weights corresponding to the dangerous indexes. According to the dangerous indexes, multiple heterogeneous data of the target park are acquired in a targeted mode, wherein the multiple heterogeneous data comprise surface ground feature information, live-action map data and actual measurement data corresponding to the dangerous indexes. Thereby avoiding acquiring excessive invalid data while reducing the acquired data volume. And then uniformly dividing the live-action map into a plurality of grids, calculating the number of nodes in each grid according to the ground surface feature information in the process of carrying out iterative division on the live-action map, and if the number of nodes in any grid is larger than a preset node number threshold value, continuing iterative division on the grids until the number of nodes in all grids is smaller than the preset node number threshold value, thereby completing division on the live-action map and obtaining a node uniform network. The situation that the number of nodes of each grid is unbalanced due to the difference of the surface ground object spatial distribution and the difference of the space object graph complexity is avoided, and the problem that the grid can not be uniformly divided due to the unbalanced number of the nodes of each grid is avoided. And finally, aiming at any grid in the node uniform network, carrying out weighted superposition on measured data corresponding to the grid and index weights to obtain dangerous situation values of the grid, and combining the dangerous situation values of all the grids to obtain the dangerous situation data of the target park. Therefore, the dangerous situation of each area is determined according to the types of the parks and the distribution of the ground surface features, and the comprehensive perception of the dangerous situation of the parks is realized.
In some implementations of this embodiment, the step of importing a plurality of virtual models into the UE system, and constructing the virtual park initial model using the plurality of virtual models according to the building environment information includes:
according to the building environment information, arranging a building model and a road model in a scene model;
and splicing the multiple virtual models to obtain the initial model of the virtual park. Thereby realizing the purpose of resculpting the building environment information of the target park.
In some implementations of this embodiment, the step of inputting the simulated simulation data into the trained AI model to obtain the corresponding risk indicator and the indicator weight includes:
if the type of the target park is a chemical industry park, the obtained dangerous index is the path track of the dangerous goods transport vehicle and the configuration information of the dangerous goods sensor;
acquiring real-time monitoring data of the dangerous article sensor according to configuration information of the dangerous article sensor;
according to real-time monitoring data of the dangerous goods sensor, calculating the weight of the sensor by combining the geographic position of the dangerous goods sensor;
and calculating the weight of the dangerous goods transportation vehicle according to the path track of the dangerous goods transportation vehicle.
Specifically, if the type of the target park is a chemical park, the park operation simulation data of the chemical park are acquired and input into the virtual park model to perform simulation operation, and simulation data are obtained. And analyzing the simulated simulation data by using the trained AI model to obtain the corresponding dangerous index which is the path track of the dangerous goods transport vehicle and the configuration information of the dangerous goods sensor. And the sensor weight is calculated according to the real-time monitoring data of the dangerous goods sensor and the geographical position of the dangerous goods sensor, and the dangerous goods transport vehicle weight is calculated according to the path track of the dangerous goods transport vehicle. And further obtaining the corresponding dangerous indexes and the index weights of the dangerous indexes of the target park.
In some implementations of this embodiment, if the type of the target campus is a chemical industry campus, before the step of obtaining the dangerous index that is the path track of the dangerous goods transport vehicle and the configuration information of the dangerous goods sensor, the method further includes:
acquiring basic information of all dangerous goods storage tanks, wherein the basic information of any dangerous goods storage tank at least comprises one or more of enterprises to which the dangerous goods storage tank belongs, the types of stored dangerous goods and installation positions;
the dangerous article sensor is configured for any dangerous article storage tank, and the configuration information of the dangerous article sensor is input according to the basic information of the dangerous article storage tank, wherein the configuration information comprises a sensor number, a sensor leakage alarm threshold value and the basic information of the corresponding dangerous article storage tank. Thereby completing the information configuration of the dangerous goods sensor.
In some implementations of this embodiment, before the step of inputting the simulated simulation data into the trained AI model, the method further includes:
establishing an AI model;
acquiring a plurality of samples, wherein the plurality of samples comprise historical dangerous index data of various parks;
the AI model is trained using a plurality of samples to obtain a trained AI model.
In some implementations of this embodiment, the step of establishing the AI model includes:
and constructing an AI model through a random forest algorithm and a convolutional neural network algorithm. Thus, the AI model can analyze various simulation data.
Wherein the random forest algorithm is a classifier comprising a plurality of decision trees and the class of the output is a mode of the class output by the individual trees. The convolutional neural network algorithm is a feedforward neural network which comprises convolutional calculation and has a depth structure, and is one of representative algorithms of deep learning. The convolutional neural network has characteristic learning capability and can carry out translation invariant classification on input information according to a hierarchical structure of the convolutional neural network.
Referring to fig. 2, fig. 2 is a block diagram illustrating a system for sensing a dangerous situation of a campus based on multi-source data fusion according to an embodiment of the present application. The embodiment of the application provides a park dangerous situation perception system based on multi-source data fusion, which comprises the following steps:
a target park obtaining module 110, configured to obtain building environment information and a plurality of historical operating conditions of the target park;
the virtual park model building module 120 is configured to build a virtual park model corresponding to the target park according to the building environment information and the plurality of historical operation conditions;
the simulation running module 130 is configured to obtain campus running simulation data matched with a type of a target campus, input the campus running simulation data into the virtual campus model for performing simulation running, and obtain simulation data;
the dangerous index obtaining module 140 is configured to input the simulated simulation data into the trained AI model, so as to obtain a corresponding dangerous index and index weight;
the heterogeneous data acquisition module 150 is configured to acquire multiple heterogeneous data of the target park according to the danger index, where the multiple heterogeneous data includes surface feature information, live-action map data, and actual measurement data corresponding to the danger index;
the grid dividing module 160 is configured to divide the live-action map data into a plurality of grids uniformly, calculate the number of nodes in each grid according to the surface feature information, and iteratively divide the grids with the number of nodes greater than the preset node number threshold value based on the number of nodes in each grid until the number of nodes in all the grids is less than the preset node number threshold value, so as to obtain a node uniform network;
the weighting module 170 is configured to perform weighted superposition on the actually measured data corresponding to the grid and the index weight for any grid in the node uniform network, so as to obtain a dangerous situation value of the grid;
and the dangerous situation value combining module 180 is used for combining the dangerous situation values of all grids to obtain dangerous situation data of the target park.
In the implementation process, the system first re-nicks the target park according to the building environment information and a plurality of historical operation conditions of the target park so as to construct a virtual park model matched with the target park. The operation process of the target park can be simulated through the virtual park model. Then, the virtual park model is utilized to perform simulation operation through park operation simulation data matched with the park type of the target park, and simulation data are obtained. And analyzing the simulated simulation data through the trained AI model to obtain the dangerous indexes corresponding to the target park and the index weights corresponding to the dangerous indexes. According to the dangerous indexes, multiple heterogeneous data of the target park are acquired in a targeted mode, wherein the multiple heterogeneous data comprise surface ground feature information, live-action map data and actual measurement data corresponding to the dangerous indexes. Thereby avoiding acquiring excessive invalid data while reducing the acquired data volume. And then uniformly dividing the live-action map into a plurality of grids, calculating the number of nodes in each grid according to the ground surface feature information in the process of carrying out iterative division on the live-action map, and if the number of nodes in any grid is larger than a preset node number threshold value, continuing iterative division on the grids until the number of nodes in all grids is smaller than the preset node number threshold value, thereby completing division on the live-action map and obtaining a node uniform network. The situation that the number of nodes of each grid is unbalanced due to the difference of the surface ground object spatial distribution and the difference of the space object graph complexity is avoided, and the problem that the grid can not be uniformly divided due to the unbalanced number of the nodes of each grid is avoided. And finally, aiming at any grid in the node uniform network, carrying out weighted superposition on measured data corresponding to the grid and index weights to obtain dangerous situation values of the grid, and combining the dangerous situation values of all the grids to obtain the dangerous situation data of the target park. Therefore, the dangerous situation of each area is determined according to the types of the parks and the distribution of the ground surface features, and the comprehensive perception of the dangerous situation of the parks is realized.
Referring to fig. 3, fig. 3 is a schematic block diagram of an electronic device according to an embodiment of the application. The electronic device comprises a memory 101, a processor 102 and a communication interface 103, wherein the memory 101, the processor 102 and the communication interface 103 are electrically connected with each other directly or indirectly to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The memory 101 may be configured to store software programs and modules, such as program instructions/modules corresponding to a system for sensing a campus hazard situation based on multi-source data fusion according to the embodiments of the present application, and the processor 102 executes the software programs and modules stored in the memory 101, thereby performing various functional applications and data processing. The communication interface 103 may be used for communication of signaling or data with other node devices.
The Memory 101 may be, but is not limited to, a random access Memory (Random Access Memory, RAM), a Read Only Memory (ROM), a programmable read Only Memory (Programmable Read-Only Memory, PROM), an erasable read Only Memory (Erasable Programmable Read-Only Memory, EPROM), an electrically erasable read Only Memory (ElectricErasable Programmable Read-Only Memory, EEPROM), etc.
The processor 102 may be an integrated circuit chip with signal processing capabilities. The processor 102 may be a general purpose processor including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processing, DSP), application specific integrated circuits (ApplicationSpecific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
It will be appreciated that the configuration shown in fig. 3 is merely illustrative, and that the electronic device may also include more or fewer components than shown in fig. 3, or have a different configuration than shown in fig. 3. The components shown in fig. 3 may be implemented in hardware, software, or a combination thereof.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The apparatus embodiments described above are merely illustrative, for example, of the flowcharts and block diagrams in the figures that illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form a single part, or each module may exist alone, or two or more modules may be integrated to form a single part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
It will be evident to those skilled in the art that the application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Claims (8)
1. A park dangerous situation perception method based on multi-source data fusion is characterized by comprising the following steps:
acquiring building environment information and a plurality of historical operating conditions of a target park;
constructing a virtual park model corresponding to the target park according to the building environment information and the historical operation conditions;
obtaining park operation simulation data matched with the park type of the target park, and inputting the park operation simulation data into the virtual park model for simulation operation to obtain simulation data;
inputting the simulated simulation data into a trained AI model to obtain corresponding dangerous indexes and index weights;
acquiring various heterogeneous data of the target park according to the dangerous index, wherein the various heterogeneous data comprise surface ground feature information, live-action map data and actual measurement data, and the actual measurement data are data obtained based on the dangerous index and actual measurement occurrence conditions;
uniformly dividing the live-action map data into a plurality of grids, calculating the number of nodes in each grid according to the surface feature information, and iteratively dividing grids with the number of nodes being greater than a preset node number threshold value based on the number of nodes of each grid until the number of nodes in all grids is smaller than the preset node number threshold value, so as to obtain a node uniform network;
aiming at any grid in the node uniform network, weighting and superposing measured data corresponding to the grid and the index weight to obtain a dangerous situation value of the grid;
and combining all the dangerous situation values of the grids to obtain dangerous situation data of the target park.
2. The method for sensing a campus hazard situation based on multi-source data fusion according to claim 1, wherein the step of constructing a virtual campus model corresponding to the target campus according to the building environment information and the plurality of historical operating conditions comprises:
calling a plurality of virtual models according to the building environment information, wherein the plurality of virtual models at least comprise a building model, a road model and a scene model;
the multiple virtual models are imported into a UE system, and a virtual park initial model is built by utilizing the multiple virtual models according to the building environment information;
establishing a virtual park neural network model based on the virtual park initial model by combining a neural network algorithm;
and inputting all the historical operation conditions into the virtual park neural network model for training to obtain a virtual park model matched with the target park.
3. The method for sensing a campus dangerous situation based on multi-source data fusion according to claim 2, wherein the step of importing the plurality of virtual models into a UE system and constructing a virtual campus initial model using the plurality of virtual models according to the building environment information comprises:
according to the building environment information, the building model and the road model are arranged in the scene model;
and splicing the multiple virtual models to obtain an initial model of the virtual park.
4. The method for sensing a campus hazard situation based on multi-source data fusion according to claim 1, wherein the step of inputting the simulated simulation data into a trained AI model to obtain corresponding hazard indexes and index weights comprises:
if the park type of the target park is a chemical industry park, the obtained dangerous index is the path track of the dangerous goods transport vehicle and the configuration information of the dangerous goods sensor;
acquiring real-time monitoring data of the dangerous article sensor according to the configuration information of the dangerous article sensor;
according to the real-time monitoring data of the dangerous article sensor, calculating the sensor weight by combining the geographic position of the dangerous article sensor;
and calculating the weight of the dangerous goods transportation vehicle according to the path track of the dangerous goods transportation vehicle.
5. The method for sensing a park hazard situation based on multi-source data fusion according to claim 4, wherein if the park type of the target park is a chemical park, the step of obtaining the hazard index as the path track of the hazard transport vehicle and the configuration information of the hazard sensor further comprises:
acquiring basic information of all dangerous goods storage tanks, wherein the basic information of any dangerous goods storage tank at least comprises one or more of enterprises to which the dangerous goods storage tank belongs, the types of stored dangerous goods and installation positions;
the dangerous goods storage tank is provided with a dangerous goods sensor, and the configuration information of the dangerous goods sensor is input according to the basic information of the dangerous goods storage tank, wherein the configuration information comprises a sensor number, a sensor leakage alarm threshold value and the basic information of the corresponding dangerous goods storage tank.
6. The method of multi-source data fusion based campus hazard situation awareness according to claim 1, further comprising, prior to the step of inputting the simulated simulation data into the trained AI model:
establishing an AI model;
acquiring a plurality of samples, wherein the plurality of samples comprise historical hazard index data of various parks;
and training the AI model by using the plurality of samples to obtain a trained AI model.
7. The method of multi-source data fusion based campus hazard situation awareness according to claim 6, wherein the step of creating AI models comprises:
and constructing an AI model through a random forest algorithm and a convolutional neural network algorithm.
8. Park dangerous situation awareness system based on multisource data fusion, which is characterized by comprising:
the target park acquisition module is used for acquiring building environment information and a plurality of historical operating conditions of the target park;
the virtual park model building module is used for building a virtual park model corresponding to the target park according to the building environment information and the historical operation conditions;
the simulation operation module is used for acquiring park operation simulation data matched with the park type of the target park, inputting the park operation simulation data into the virtual park model for simulation operation, and obtaining simulation data;
the dangerous index obtaining module is used for inputting the simulated simulation data into a trained AI model to obtain corresponding dangerous indexes and index weights;
the heterogeneous data acquisition module is used for acquiring various heterogeneous data of the target park according to the dangerous indexes, wherein the various heterogeneous data comprise surface ground feature information, live-action map data and actual measurement data, and the actual measurement data are data obtained based on the dangerous indexes and actual measurement occurrence conditions;
the grid dividing module is used for uniformly dividing the live-action map data into a plurality of grids, calculating the number of nodes in each grid according to the surface feature information, and carrying out iterative division on grids with the number of nodes being larger than a preset node number threshold value based on the number of nodes of each grid until the number of nodes in all grids is smaller than the preset node number threshold value, so as to obtain a node uniform network;
the weighting module is used for weighting and superposing measured data corresponding to any grid in the node uniform network and the index weight to obtain a dangerous situation value of the grid;
and the dangerous situation value combination module is used for combining all the dangerous situation values of the grids to obtain dangerous situation data of the target park.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310524192.5A CN116384255B (en) | 2023-05-11 | 2023-05-11 | Park dangerous situation perception method and system based on multi-source data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310524192.5A CN116384255B (en) | 2023-05-11 | 2023-05-11 | Park dangerous situation perception method and system based on multi-source data fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116384255A CN116384255A (en) | 2023-07-04 |
| CN116384255B true CN116384255B (en) | 2023-08-11 |
Family
ID=86964204
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310524192.5A Active CN116384255B (en) | 2023-05-11 | 2023-05-11 | Park dangerous situation perception method and system based on multi-source data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116384255B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110070026A (en) * | 2019-04-17 | 2019-07-30 | 中国科学院自动化研究所 | Video image danger Situation Awareness method, system, device based on pattern-recognition |
| WO2019213980A1 (en) * | 2018-05-08 | 2019-11-14 | 清华大学 | Intelligent vehicle safety decision-making method employing driving safety field |
| CN110544402A (en) * | 2019-08-08 | 2019-12-06 | 南京莱斯信息技术股份有限公司 | Chemical industry park virtual simulation emergency deduction system based on big data fusion |
| CN110705842A (en) * | 2019-09-16 | 2020-01-17 | 嘉兴恒云数据科技有限公司 | Chemical industry park dangerous situation sensing method and system based on multi-source data fusion |
| CN111913406A (en) * | 2020-08-11 | 2020-11-10 | 闽江学院 | Ship-shore collaborative simulation system for intelligent navigation and safety of ship |
| CN115496398A (en) * | 2022-10-12 | 2022-12-20 | 国网四川省电力公司电力科学研究院 | Electric power operation safety control method and system |
| CN115689281A (en) * | 2022-10-24 | 2023-02-03 | 新特能源股份有限公司 | Factory hazard risk management and control method and device |
| CN115795697A (en) * | 2023-01-31 | 2023-03-14 | 中国电子科技集团公司第二十八研究所 | Method for generating field maneuvering channel in complex environment based on space grid |
| CN115937288A (en) * | 2022-10-12 | 2023-04-07 | 国网四川省电力公司电力科学研究院 | Three-dimensional scene model construction method for transformer substation |
-
2023
- 2023-05-11 CN CN202310524192.5A patent/CN116384255B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019213980A1 (en) * | 2018-05-08 | 2019-11-14 | 清华大学 | Intelligent vehicle safety decision-making method employing driving safety field |
| CN110070026A (en) * | 2019-04-17 | 2019-07-30 | 中国科学院自动化研究所 | Video image danger Situation Awareness method, system, device based on pattern-recognition |
| CN110544402A (en) * | 2019-08-08 | 2019-12-06 | 南京莱斯信息技术股份有限公司 | Chemical industry park virtual simulation emergency deduction system based on big data fusion |
| CN110705842A (en) * | 2019-09-16 | 2020-01-17 | 嘉兴恒云数据科技有限公司 | Chemical industry park dangerous situation sensing method and system based on multi-source data fusion |
| CN111913406A (en) * | 2020-08-11 | 2020-11-10 | 闽江学院 | Ship-shore collaborative simulation system for intelligent navigation and safety of ship |
| CN115496398A (en) * | 2022-10-12 | 2022-12-20 | 国网四川省电力公司电力科学研究院 | Electric power operation safety control method and system |
| CN115937288A (en) * | 2022-10-12 | 2023-04-07 | 国网四川省电力公司电力科学研究院 | Three-dimensional scene model construction method for transformer substation |
| CN115689281A (en) * | 2022-10-24 | 2023-02-03 | 新特能源股份有限公司 | Factory hazard risk management and control method and device |
| CN115795697A (en) * | 2023-01-31 | 2023-03-14 | 中国电子科技集团公司第二十八研究所 | Method for generating field maneuvering channel in complex environment based on space grid |
Non-Patent Citations (1)
| Title |
|---|
| 城市消防安全风险态势计算方法研究;隗乔;《北京工业大学学报》;第37卷(第05期);第729-733页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116384255A (en) | 2023-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103914622B (en) | A kind of chemical leakage fast prediction alarm emergency response decision-making method | |
| Skakun et al. | Flood hazard and flood risk assessment using a time series of satellite images: A case study in Namibia | |
| Li et al. | A household-level approach to staging wildfire evacuation warnings using trigger modeling | |
| RU2680652C2 (en) | Method of integrated environmental monitoring | |
| Liao et al. | Environmental emergency decision support system based on Artificial Neural Network | |
| CN116109462B (en) | Pollution monitoring and early warning method and system for drinking water source area after natural disaster | |
| Yoshida et al. | Optimal sampling placement in a Gaussian random field based on value of information | |
| CN110503805A (en) | Pipeline Geological Hazards Monitoring method for early warning and device | |
| CN115170980B (en) | Gas pipeline high consequence area intelligent identification method based on satellite image identification | |
| Mangukiya et al. | Flood risk mapping for the lower Narmada basin in India: A machine learning and IoT-based framework | |
| Koshimura | Establishing the advanced disaster reduction management system by fusion of real-time disaster simulation and big data assimilation | |
| Luo et al. | A fast PM2. 5 forecast approach based on time-series data analysis, regression and regularization | |
| CN115688032A (en) | Power grid risk early warning method and system under typhoon disaster based on multi-source data fusion | |
| Diao et al. | Uncertainty aware wildfire management | |
| Finogeev et al. | Collection and consolidation of big data for proactive monitoring of critical events at infrastructure facilities in an urban environment | |
| Raihan | A comprehensive review of the recent advancement in integrating deep learning with geographic information systems | |
| Bolt et al. | A spatio-temporal neural network forecasting approach for emulation of firefront models | |
| CN116384255B (en) | Park dangerous situation perception method and system based on multi-source data fusion | |
| Liu et al. | Exploiting a real-time self-correcting digital twin model for the middle route of the south-to-north water diversion project of china | |
| CN110399537B (en) | Artificial intelligence technology-based warning situation space-time prediction method | |
| CN114925517A (en) | Urban multi-disaster coupling analysis method | |
| Bruzzone et al. | Simulation of hazardous material fallout for emergency management during accidents | |
| CN112884310A (en) | Computer-aided assessment method, system and device for pollutant diffusion rule | |
| CN113408984B (en) | Dangerous chemical transportation tracking system and method | |
| Mukhacheva et al. | Digital Twins of the Urban Ecosystem to Ensure the Quality of Life of the Population |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 610042 floor 6, unit 1, building 4, Fengde International Plaza, No. 6, hanghang Road, Wuhou District, Chengdu, Sichuan Patentee after: New Yingshun Information Technology Co.,Ltd. Address before: 610042 floor 6, unit 1, building 4, Fengde International Plaza, No. 6, hanghang Road, Wuhou District, Chengdu, Sichuan Patentee before: SICHUAN XINYINGSHUN INFORMATION TECHNOLOGY CO.,LTD. |
|
| CP01 | Change in the name or title of a patent holder |