CN110929923A - Urban safety risk management and control system based on digital twin technology - Google Patents
Urban safety risk management and control system based on digital twin technology Download PDFInfo
- Publication number
- CN110929923A CN110929923A CN201911092036.6A CN201911092036A CN110929923A CN 110929923 A CN110929923 A CN 110929923A CN 201911092036 A CN201911092036 A CN 201911092036A CN 110929923 A CN110929923 A CN 110929923A
- Authority
- CN
- China
- Prior art keywords
- risk
- data
- information acquisition
- model
- monitoring
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Abstract
The invention relates to a city safety risk management and control system based on a digital twin technology, which comprises an information acquisition module, a safety risk big database for processing and storing data of the information acquisition module, and a risk display platform for processing and applying the safety risk big database, wherein the safety risk big database also comprises basic information data, the information acquisition module comprises a ground information acquisition unit and an aerial information acquisition unit, the ground information acquisition unit comprises mobile phone signaling in an area, an RFID label of a target object, monitoring video in the area, and various sensors arranged in a key monitoring area, the aerial information acquisition unit comprises unmanned aerial vehicle real-time shooting data and a remote sensing satellite image map, the risk display platform comprises geographic information data display analysis, risk thematic data display and risk thematic data retrieval, risk thematic data summarization and internet of things sensing equipment real-time monitoring.
Description
Technical Field
The invention relates to a digital twin technology, in particular to a function for realizing city safety monitoring and risk early warning by utilizing the technology.
Background
With the rapid promotion of urbanization in China, high concentration of urban population, industry, logistics and energy is brought, and further urban safety risks are increased day by day. Therefore, how to establish a city risk management and control system is an urgent problem to be solved.
Disclosure of Invention
In order to overcome the defects of the technology, the invention provides a safety risk management and control system.
In order to solve the technical problems, the system comprises an information acquisition module, a large safety risk database for processing and storing data of the information acquisition module, a risk display platform for processing and applying the large safety risk database, the large safety risk database also comprises basic information data, the information acquisition module comprises a ground information acquisition unit and an air information acquisition unit, the ground information acquisition unit comprises mobile phone signaling in an area, an RFID label of a target object, monitoring video in the area and various sensors installed in a key monitoring area, the air information acquisition unit comprises unmanned aerial vehicle real-time shooting data and a remote sensing satellite image map, the risk display platform comprises geographic information data display analysis, risk thematic data display and risk thematic data retrieval, risk thematic data summarization and internet of things sensing equipment real-time monitoring.
After the method is adopted, the safety risk management and control system adopts a mode of collecting information and arranging the information into a safety risk big database, and finally, the safety risk management and control system is applied to the public or a supervision department in a mode of a display platform, so that the monitoring capability of an application on a risk target can be effectively improved. The collection mode mainly comprises ground collection and air collection, the average personnel number, the personnel density thermodynamic distribution and the space-time operation trend condition in different periods (early, middle and late) in the near future in a specific area of a major hazard source, and the information such as the sex and the native proportion of personnel in the area are obtained through deep mining and application of mobile phone signaling big data of a communication operator, the threat degree of various risks to the personnel safety can be predicted, and powerful support is provided for the formulation of an emergency plan, the configuration of emergency resources and the decision and command of safety event disposal. The system comprises a ground mobile signaling, a monitoring video and an aerial information acquisition unit, wherein the ground mobile signaling can acquire a signaling state to monitor a local communication condition or inform corresponding people of emergency evacuation by using the mobile signaling, an RFID (radio frequency identification) label of a target object can be quickly positioned to a corresponding building or a target object with high safety risk, the monitoring video can sample the safety condition in an area in real time, once the risk condition occurs, a risk source can be quickly and accurately found and searched, various sensors can return corresponding sensor data, each risk value in the accurate monitoring area can be accurately monitored, the aerial information acquisition unit comprises GPS (global positioning system) positioning satellite remote sensing and unmanned aerial vehicle shooting, the monitoring breadth of aerial monitoring can be increased and the high-altitude risk monitoring can be increased by unmanned aerial vehicle shooting, the monitoring area can be used for capturing monitoring areas which are not acquired by the monitoring video, and the GPS positioning satellite remote sensing is an emergency vehicle for transporting high-risk goods, such as, The real-time position information, the track and the like of vehicles such as passenger and cargo transport vehicles, ambulances, ships and the like and equipment can monitor the transportation routes in real time, and an application can monitor the type of target objects to avoid corresponding dangerous situations. And the risk display platform can visually display the acquired data and correspondingly analyze the acquired data to obtain various display systems for users to use.
As a further improvement of the invention, the geographic information data display analysis comprises the steps of carrying out data acquisition on a remote sensing satellite image map and forming a historical remote sensing image, a three-dimensional model scene based on a digital twin technology and a virtual reality technology, and the display of a place name and an address. The digital twinning technology and the virtual reality technology are based on a three-dimensional GIS system, a city region model is established by performing three-dimensional modeling on the existing region, and the digital twinning module acquires relevant data in the ground information acquisition unit on the basis of the virtual and reality technology, performs simulation and real-time monitoring on the environment.
After the method is adopted, the digital twin technology fully utilizes data such as a physical model, sensor updating, operation history and the like, integrates simulation processes of multiple disciplines, multiple physical quantities, multiple scales and multiple probabilities, and finishes mapping in a virtual space so as to reflect the whole life cycle process of corresponding entity equipment. Therefore, the entity objects and the environment are simulated and simulated in the virtual digital space, and the results are used for actual planning, construction, management, monitoring, early warning, prediction and the like. The risk thematic data display comprises the steps of displaying risk data information in various industry fields in a superposition mode according to spatial positions, and meanwhile marking differently according to risk grades, risk areas, risk types and risk historical conditions to form a risk thematic data map. And (3) overlapping and displaying each item of risk data in space by taking the space as a reference, carrying out different labels, realizing real-time online monitoring and analysis on the urban safety risk sources in the Wenzhou city, and carrying out red, orange, yellow and blue risk early warning classification.
As a further improvement of the invention, the urban regional model comprises an urban flooding model, a chemical plant danger early warning model and a forest park fire prevention susceptibility model, the forest park fire prevention susceptibility model comprises a first step of extracting climate characteristics by using sensor technologies respectively, extracting vegetation related information by using forestry bureau basic big data, extracting a slope, a slope direction and a terrain humidity index by using a GIS technology, a second step of establishing a GIS database by using the data, establishing a machine learning model by using nuclear logic regression for the data in the GIS database, and a third step of optimizing the obtained result by using a BFGS algorithm.
After the method is adopted, an urban water flooding model, a chemical hazard early warning model and a forest park fire susceptibility model are all frequent and serious risk accidents in urban safety risks, the accident initiation probability of a region is researched through establishment of the models, the forest fire susceptibility model is a machine learning model which is important and easy to implement in the models, relevant data are stored in a forest Geographic Information System (GIS) database through information acquired by a sensor in a digital twin technology, basic big data and a GIS technology, and the data are subjected to correlation analysis by the machine learning model to finally obtain the fire occurrence probability of a forest in a certain park of a certain city.
As a further improvement of the invention, the kernel logistic regression is used for establishing the machine learning model to solve the fire occurrence probability p (x) mainly by setting a logistic function algorithm, and a training data set is considered
The first step, which can be obtained according to a kernel logistic regression formula,
wherein xiFor input variables, including relevant variables such as grade, slope, terrain moisture index, land cover, surface temperature, distance to road, distance to densely populated areas, wind speed and rainfall, obtained from sensors or GIS technology, yiE {0,1}, corresponding to a label of whether the relevant variable is fire, wherein αiAnd b are model weights, and in a second step, parameters α are solved and obtained using a minimized negative log-likelihood function and a gradient descent method, respectivelyiB, wherein the formula for minimizing the negative log-likelihood function is as follows,
c is a regularization parameter, and in the third step, a radial basis function k (x, x ') is used as one of the kernel logic functions, the kernel function is added on the basis of the logic function, and the correlation among the attribute characteristics of the used variables is calculated, wherein the algorithm is as follows, k (x, x ') is exp ((| | x-x ' | Y)2)/2δ2Where δ is a hyper-parameter for manual input and the optimization model can be adjusted manually.
After the method is adopted, a model of the fire disaster initiation probability is trained through machine learning. Kernel Logistic Regression (KLR) is a powerful machine learning classification method in which probabilistic outcomes are estimated based on optimizing a minimized negative log-likelihood function using a Broyden-Fletcher-Goldfarb-Shanno (BFGS). The method is characterized in that a kernel function is added on the basis of logistic regression, and the correlation among selected attribute features can be calculated. Consider a training data set, in which there is input data with N variables and N data samples. In the context of this study, the input variables are grade, slope, terrain moisture index, land cover, surface temperature, distance to road, distance to populated areas, wind speed and rainfall. y isiAre the corresponding labels indicating that a forest is on fire and that a forest is not on fire. KLR canEstablishing a non-linear decision boundary separates two classes in feature space by the following equation: where δ is the hyper-parameter, controlling the sensitivity of the kernel function. The parameters are obtained by minimizing negative log-likelihood functions, where C is a regularization parameter for which the model is not over-trained.
As a further improvement of the present invention, the machine learning model further comprises a model verification evaluation algorithm, the algorithm comprises evaluation indexes, such as overall accuracy, specificity, sensitivity, Positive Predictive Value (PPV) and Negative Predictive Value (NPV), and the five evaluation indexes are obtained by setting TP (positive sample predicted to be positive by the model); TN (negative samples predicted negative by the model); FP (negative sample predicted positive by the model); FN (positive samples predicted by the model to be negative); and the four parameters are respectively mapped to TP and TN and are the sample numbers of the training data set or the verification data set which are correctly classified into the forest fire-prevention grade and the non-forest fire-prevention grade, and FP and FN are the sample numbers of the training data set or the verification data set which are wrongly classified, so that the conclusion can be drawn that
After the method is adopted, the performance of the forest fire model is evaluated by using five statistical evaluation indexes, such as overall accuracy, specificity, sensitivity, Positive Predictive Value (PPV) and Negative Predictive Value (NPV). The overall accuracy is the proportion of training (or validation) samples that are correctly classified; the sensitivity is the proportion of correct classification of forest fires; specificity is the proportion of correctly classified non-forest fires. PPV is the probability of training (or validating) a sample in the data set being classified as a forest fire rating, while NPV is the probability of training (or validating) a sample in the data set being correctly classified as a non-forest fire rating: wherein TP and TN are the number of samples in the training data set or the validation data set that are correctly classified as forest fire rating and non-forest fire rating, respectively. FP and FN are the number of misclassified samples in the training data set or the validation data set. A Receiver Operating Characteristic (ROC) curve and an area under the ROC curve (AUC) can be used to evaluate a global measure of model performance. ROC curves are descriptive graphs constructed based on sensitivity and specificity. If AUC equals 1, a perfect model is obtained, whereas if AUC is 0, the model is not informative.
As a further improvement of the invention, the risk thematic data retrieval comprises comprehensive attribute keyword retrieval and image retrieval, and the risk level, the risk area, the risk type and the risk possible consequences are subjected to positioning tracking and detail review by the two retrieval modes.
After the method is adopted, the risk thematic data retrieval can help an application person to quickly locate corresponding risk data so as to track risks and inquire risk details.
As a further improvement of the invention, the risk topic data summarization classifies and statistically compares risk sources according to risk source industry, level, region, time, type, measure, and the like.
After the method is adopted, the intuitive overall grasp of all risk sources in the whole city is realized through classification, statistics, comparison and analogy of the risk sources.
As a further improvement of the invention, the Internet of things sensing equipment monitors Internet of things sensing data such as facility equipment monitoring, environment monitoring, personnel activity monitoring, hazard source monitoring, video monitoring and the like of various industries and departments in real time and dynamically in real time through an Internet of things equipment access interface in a multi-source format, combines a GIS technology, visually displays the Internet of things sensing data on a risk map, and can comprehensively compare and analyze single-station data and multi-station data through a plurality of data forms such as a broken line graph, a bar graph, a table and the like.
By adopting the method, the Internet of things perception data of the risk map is visually displayed, and meanwhile, the comprehensive comparative analysis of single-station and multi-station data can be carried out through various data forms such as a line graph, a bar graph, a table and the like, so that the operation monitoring condition of the Wenzhou global risk is displayed in real time from coarse granularity to fine granularity and from the whole to the local.
Detailed Description
Comprises an information acquisition module, a security risk big database for processing and storing the data of the information acquisition module, and a risk display platform for processing and applying a security risk big database, wherein the security risk big database also comprises basic information data, the information acquisition module comprises a ground information acquisition unit and an air information acquisition unit, the ground information acquisition unit comprises a mobile phone signaling in an area, an RFID label of a target object and a monitoring video in the area, and various sensors installed in the key monitoring area, the aerial information acquisition unit comprises real-time shooting data of the unmanned aerial vehicle and a remote sensing satellite image map, the risk display platform comprises geographic information data display analysis, risk thematic data display, risk thematic data retrieval, risk thematic data summarization and Internet of things sensing equipment real-time monitoring.
After the method is adopted, the safety risk management and control system adopts a mode of collecting information and arranging the information into a safety risk big database, and finally, the safety risk management and control system is applied to the public or a supervision department in a mode of a display platform, so that the monitoring capability of an application on a risk target can be effectively improved. The collection mode mainly comprises ground collection and air collection, the average personnel number, the personnel density thermodynamic distribution and the space-time operation trend condition in different periods (early, middle and late) in the near future in a specific area of a major hazard source, and the information such as the sex and the native proportion of personnel in the area are obtained through deep mining and application of mobile phone signaling big data of a communication operator, the threat degree of various risks to the personnel safety can be predicted, and powerful support is provided for the formulation of an emergency plan, the configuration of emergency resources and the decision and command of safety event disposal. The system comprises a ground mobile signaling, a monitoring video and an aerial information acquisition unit, wherein the ground mobile signaling can acquire a signaling state to monitor a local communication condition or inform corresponding people of emergency evacuation by using the mobile signaling, an RFID (radio frequency identification) label of a target object can be quickly positioned to a corresponding building or a target object with high safety risk, the monitoring video can sample the safety condition in an area in real time, once the risk condition occurs, a risk source can be quickly and accurately found and searched, various sensors can return corresponding sensor data, each risk value in the accurate monitoring area can be accurately monitored, the aerial information acquisition unit comprises GPS (global positioning system) positioning satellite remote sensing and unmanned aerial vehicle shooting, the monitoring breadth of aerial monitoring can be increased and the high-altitude risk monitoring can be increased by unmanned aerial vehicle shooting, the monitoring area can be used for capturing monitoring areas which are not acquired by the monitoring video, and the GPS positioning satellite remote sensing is an emergency vehicle for transporting high-risk goods, such as, The real-time position information, the track and the like of vehicles such as passenger and cargo transport vehicles, ambulances, ships and the like and equipment can monitor the transportation routes in real time, and an application can monitor the type of target objects to avoid corresponding dangerous situations. And the risk display platform can visually display the acquired data and correspondingly analyze the acquired data to obtain various display systems for users to use.
As a further improvement of the invention, the geographic information data display analysis comprises the steps of carrying out data acquisition on a remote sensing satellite image map and forming a historical remote sensing image, a three-dimensional model scene based on a digital twin technology and a virtual reality technology, and the display of a place name and an address. The digital twinning technology and the virtual reality technology are based on a three-dimensional GIS system, a city region model is established by performing three-dimensional modeling on the existing region, and the digital twinning module acquires relevant data in the ground information acquisition unit on the basis of the virtual and reality technology, performs simulation and real-time monitoring on the environment.
After the method is adopted, the digital twin technology fully utilizes data such as a physical model, sensor updating, operation history and the like, integrates simulation processes of multiple disciplines, multiple physical quantities, multiple scales and multiple probabilities, and finishes mapping in a virtual space so as to reflect the whole life cycle process of corresponding entity equipment. Therefore, the entity objects and the environment are simulated and simulated in the virtual digital space, and the results are used for actual planning, construction, management, monitoring, early warning, prediction and the like. The risk thematic data display comprises the steps of displaying risk data information in various industry fields in a superposition mode according to spatial positions, and meanwhile marking differently according to risk grades, risk areas, risk types and risk historical conditions to form a risk thematic data map. And (3) overlapping and displaying each item of risk data in space by taking the space as a reference, carrying out different labels, realizing real-time online monitoring and analysis on the urban safety risk sources in the Wenzhou city, and carrying out red, orange, yellow and blue risk early warning classification.
As a further improvement of the invention, the urban regional model comprises an urban flooding model, a chemical plant danger early warning model and a forest park fire prevention susceptibility model, the forest park fire prevention susceptibility model comprises a first step of extracting climate characteristics by using sensor technologies respectively, extracting vegetation related information by using forestry bureau basic big data, extracting a slope, a slope direction and a terrain humidity index by using a GIS technology, a second step of establishing a GIS database by using the data, establishing a machine learning model by using nuclear logic regression for the data in the GIS database, and a third step of optimizing the obtained result by using a BFGS algorithm.
After the method is adopted, an urban water flooding model, a chemical hazard early warning model and a forest park fire susceptibility model are all frequent and serious risk accidents in urban safety risks, the accident initiation probability of a region is researched through establishment of the models, the forest fire susceptibility model is a machine learning model which is important and easy to implement in the models, relevant data are stored in a GIS database through information acquired by a sensor in a digital twin technology, forestry basic big data and a GIS technology, and the machine learning model is used for carrying out correlation analysis on the data to finally obtain the fire occurrence probability of a certain forest park in a certain city. The kernel logistic regression is used for establishing a machine learning model, solving the fire occurrence probability p (x) mainly by setting a logistic function algorithm, and considering a training data set
The first step, according to the kernel logistic regression formula, the corresponding formula and algorithm for solving p (x) can be obtained, and the variables include the slope, the direction of the slope, the topographic moisture index, the land cover, the surface temperature and the height obtained from the sensor or the GIS technologyDistance to road, distance to densely populated area, wind speed and rainfall, and like variables, yiE {0,1}, corresponding to a label of whether the relevant variable is fire, wherein αiAnd b are model weights, and in the second step, parameters α are solved and obtained separately using a minimized negative log-likelihood function and a gradient descent methodiB, wherein the formula for minimizing the negative logarithm likelihood function is shown as follows, wherein C is a regularization parameter, and the third step is to calculate the correlation between the attribute characteristics of the variables used by using a radial basis function k (x, x ') as one of the kernel logic functions and adding the kernel function to the logic functions, and the algorithm is as follows, k (x, x ') -exp ((| | x-x ' | i) m2)/2δ2Where δ is a hyper-parameter used for manual input and the optimization model can be adjusted manually. A model of the probability of fire initiation is trained through machine learning. Nuclear logic regression (KLR) is a powerful machine learning classification method in which probabilistic results are estimated based on optimizing a minimized negative log-likelihood function using a Broyden-Fletcher-Goldfarb-shanno (bfgs). The method is characterized in that a kernel function is added on the basis of logistic regression, and the correlation among selected attribute features can be calculated. Consider a training data set, in which there is input data with N variables and N data samples. In the context of this study, the input variables were grade, slope, terrain moisture index, land cover, surface temperature, distance to road, distance to densely populated areas, wind speed and rainfall. y isiAre the corresponding labels indicating that a forest is on fire and that a forest is not on fire. The KLR can establish a non-linear decision boundary that separates two classes in feature space by the following equation: where δ is the hyper-parameter, controlling the sensitivity of the kernel function. The parameters are obtained by minimizing a negative log-likelihood function, where C is a regularization parameter for which the model is not trained to over-fit.
The machine learning model further comprises a model validation evaluation algorithm comprising evaluation metrics such as overall accuracy, specificity, sensitivity, Positive Predictive Value (PPV) andnegative Predictive Value (NPV), and the five evaluation indexes are determined by setting TP (positive sample predicted to be positive by the model); TN (negative samples predicted negative by the model); FP (negative sample predicted positive by the model); FN (positive samples predicted by the model to be negative); and the four parameters are respectively mapped into TP and TN samples which are correctly classified into forest fire prevention grade and non-forest fire prevention grade in the training data set or the verification data set, and FP and FN samples which are wrongly classified in the training data set or the verification data set, so that the method can be deduced
The performance of the forest fire model was evaluated using five statistical evaluation indicators, such as overall accuracy, specificity, sensitivity, Positive Predictive Value (PPV) and Negative Predictive Value (NPV). The overall accuracy is the proportion of training (or validation) samples that are correctly classified; the sensitivity is the proportion of correct classification of forest fires; specificity is the proportion of correctly classified non-forest fires. PPV is the probability of training (or validating) a sample in the data set being classified as a forest fire rating, while NPV is the probability of training (or validating) a sample in the data set being correctly classified to a non-forest fire rating: wherein TP and TN are the number of samples in the training data set or the validation data set that are correctly classified as forest fire rating and non-forest fire rating, respectively. FP and FN are the number of misclassified samples in the training data set or the validation data set. A Receiver Operating Characteristic (ROC) curve and an area under the ROC curve (AUC) can be used to evaluate a global measure of model performance. ROC curves are descriptive graphs constructed based on sensitivity and specificity. If AUC equals 1, a perfect model is obtained, whereas if AUC is 0, the model is not informative.
The risk thematic data retrieval comprises comprehensive attribute keyword retrieval and image retrieval, and the risk level, the risk area, the risk type and the risk occurring consequence are positioned, tracked and referred for details through the two retrieval modes.
The retrieval of risk profile data can help the user to quickly locate the corresponding risk data so as to track the risk and inquire the risk details. And summarizing the risk thematic data to classify the risk sources according to the risk source industry, level, area, time, type, measure and the like and to perform statistical comparison. Through classifying, counting, comparing and analogy of the risk sources, the intuitive global grasp on all the risk sources in the whole city is realized.
The internet of things sensing equipment monitors the internet of things sensing data such as facility equipment monitoring, environment monitoring, personnel activity monitoring, hazard source monitoring and video monitoring of various industries and departments in real time and dynamically in real time through an internet of things equipment access interface in a multi-source format, combines a GIS technology, visually displays the internet of things sensing data on a risk map, and can comprehensively compare and analyze single-station data and multi-station data through various data forms such as a line graph, a column graph and a table. The internet of things perception data of the risk map is visually displayed, meanwhile, comprehensive comparison and analysis of single-station data and multi-station data can be carried out through a plurality of data forms such as a line graph, a bar graph and a table, and the operation monitoring condition of the Wenzhou global risk is displayed in real time from coarse granularity to fine granularity and from the whole to the local.
Claims (10)
1. City safety risk management and control system based on digit twin technique, its characterized in that: comprises an information acquisition module, a security risk big database for processing and storing the data of the information acquisition module, and a risk display platform for processing and applying a security risk big database, wherein the security risk big database also comprises basic information data, the information acquisition module comprises a ground information acquisition unit and an air information acquisition unit, the ground information acquisition unit comprises a mobile phone signaling in an area, an RFID label of a target object and a monitoring video in the area, and various sensors installed in the key monitoring area, the aerial information acquisition unit comprises real-time shooting data of the unmanned aerial vehicle and a remote sensing satellite image map, the risk display platform comprises geographic information data display analysis, risk thematic data display, risk thematic data retrieval, risk thematic data summarization and Internet of things sensing equipment real-time monitoring.
2. The digital twin-based urban security risk management and control system according to claim 1, wherein: the geographic information data display analysis comprises the steps of obtaining data of a remote sensing satellite image map, forming a historical remote sensing image, displaying a three-dimensional model scene based on a digital twin technology and a virtual reality technology, and displaying a place name and an address.
3. The digital twin-based urban security risk management and control system according to claim 2, wherein: the digital twinning technology and the virtual reality technology are based on a three-dimensional GIS system, a city region model is established by performing three-dimensional modeling on the existing region, and the digital twinning module acquires relevant data in the ground information acquisition unit on the basis of the virtual and real technologies, performs simulation and real-time monitoring on the environment.
4. The digital twin-based urban security risk management and control system according to claim 3, wherein: the method comprises the following steps of establishing a GIS database by using the data, establishing a machine learning model by using nuclear logistic regression for the data in the GIS database, and optimizing the obtained result by using a BFGS algorithm.
5. The digital twin-based urban security risk management and control system according to claim 4, wherein: the kernel logistic regression is used for establishing a machine learning model mainly by setting a logistic functionAn algorithm to solve the probability of fire occurrence p (x) taking into account the training data set
The first step, which can be obtained according to a kernel logistic regression formula,
p(x)∈[0,1]wherein x isiFor input variables, including relevant variables such as grade, slope, terrain moisture index, land cover, surface temperature, distance to road, distance to densely populated areas, wind speed and rainfall, obtained from sensors or GIS technology, yiE {0,1}, corresponding to a label of whether the relevant variable is fire, wherein αiAnd b are model weights, and in the second step, parameters α are solved and obtained separately using a minimized negative log-likelihood function and a gradient descent methodiB, wherein the formula for minimizing the negative log-likelihood function is as follows,
c is a regularization parameter, and in the third step, a radial basis function k (x, x ') is used as one of the kernel logic functions, the kernel function is added on the basis of the logic function, and the correlation among the attribute characteristics of the used variables is calculated, wherein the algorithm is as follows, k (x, x ') is exp ((| | x-x ' |)2)/2δ2Where δ is a hyper-parameter used for manual input and the optimization model can be adjusted manually.
6. The digital twin-based urban security risk management and control system according to claim 5, wherein: the machine learning model further comprises a model verification evaluation algorithm, wherein the algorithm comprises evaluation indexes including overall accuracy, specificity, sensitivity, a Positive Predictive Value (PPV) and a Negative Predictive Value (NPV), and the evaluation indexes are determined by setting TP (positive sample predicted to be positive by the model); TN (negative samples predicted negative by the model); FP (negative sample predicted positive by the model); FN (positive samples predicted by the model to be negative); and the four parameters are respectively mapped to TP and TN are the number of samples in the training data set or validation data set correctly classified as forest fire rating and non-forest fire rating, respectively, and FP and FN are the number of samples in the training data set or validation data set that were misclassified, then it can be concluded that
7. The digital twin-based urban security risk management and control system according to claim 1, wherein: the risk thematic data display comprises the steps of displaying risk data information in various industry fields in a superposition mode according to spatial positions, and meanwhile marking differently according to risk grades, risk areas, risk types and risk historical conditions to form a risk thematic data map.
8. The digital twin-based urban security risk management and control system according to claim 4, wherein: the risk thematic data retrieval comprises comprehensive attribute keyword retrieval and image retrieval, and the risk level, the risk area, the risk type and the risk occurring consequence are positioned, tracked and referred for details through the two retrieval modes.
9. The digital twin-based urban security risk management and control system according to claim 5, wherein: and summarizing the risk thematic data to classify the risk sources according to the risk source industry, level, area, time, type, measure and the like and to perform statistical comparison.
10. The digital twin-based urban security risk management and control system according to claim 1, wherein: the internet of things sensing equipment monitors the internet of things sensing data such as facility equipment monitoring, environment monitoring, personnel activity monitoring, danger source monitoring and video monitoring of all industries and departments in real time and dynamically in real time through an internet of things equipment access interface in a multi-source format, combines a GIS technology, visually displays the internet of things sensing data on a risk map, and can perform comprehensive contrastive analysis on single-station data and multi-station data through various data forms such as a line graph, a column graph and a table.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911092036.6A CN110929923B (en) | 2019-11-08 | 2019-11-08 | Urban safety risk management and control system based on digital twin technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911092036.6A CN110929923B (en) | 2019-11-08 | 2019-11-08 | Urban safety risk management and control system based on digital twin technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110929923A true CN110929923A (en) | 2020-03-27 |
| CN110929923B CN110929923B (en) | 2023-04-07 |
Family
ID=69853681
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911092036.6A Active CN110929923B (en) | 2019-11-08 | 2019-11-08 | Urban safety risk management and control system based on digital twin technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110929923B (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111445168A (en) * | 2020-04-21 | 2020-07-24 | 深圳民声第三方监管有限公司 | Quality safety third-party supervision system and method |
| CN111443609A (en) * | 2020-04-16 | 2020-07-24 | 成都三业嘉豪科技有限公司 | Laboratory environment self-adaptive adjusting method based on Internet of things |
| CN111563446A (en) * | 2020-04-30 | 2020-08-21 | 郑州轻工业大学 | Human-machine interaction safety early warning and control method based on digital twin |
| CN111680897A (en) * | 2020-05-27 | 2020-09-18 | 公安部道路交通安全研究中心 | Service management and control method, device and system and computer readable storage medium |
| CN111723469A (en) * | 2020-05-25 | 2020-09-29 | 贵州华泰智远大数据服务有限公司 | Digital portrait modeling method and system based on twin data |
| CN111881495A (en) * | 2020-06-24 | 2020-11-03 | 北京工业大学 | Prestress steel structure safety assessment method based on digital twinning |
| CN111931260A (en) * | 2020-08-04 | 2020-11-13 | 灵然创智（天津）动画科技发展有限公司 | City digital twin visualization cooperation method |
| CN112802287A (en) * | 2020-12-16 | 2021-05-14 | 广东电网有限责任公司电力科学研究院 | Power transmission line forest fire monitoring, early warning and positioning system and method thereof |
| CN113051573A (en) * | 2021-02-19 | 2021-06-29 | 广州银汉科技有限公司 | Host safety real-time monitoring alarm system based on big data |
| CN113110221A (en) * | 2021-04-29 | 2021-07-13 | 上海智大电子有限公司 | Comprehensive intelligent monitoring method and system for pipe gallery system |
| CN113268867A (en) * | 2021-05-17 | 2021-08-17 | 中国科学院、水利部成都山地灾害与环境研究所 | Glacier debris flow area early warning method based on hydrothermal combination |
| CN113341893A (en) * | 2021-05-26 | 2021-09-03 | 苏州市给排水设计院有限责任公司 | Secondary water supply layered distributed control method, system and storage medium |
| CN113378700A (en) * | 2021-06-08 | 2021-09-10 | 四川农业大学 | Grassland rat damage dynamic monitoring method based on unmanned aerial vehicle aerial image |
| CN113792658A (en) * | 2021-09-15 | 2021-12-14 | 江苏迪赛司自动化工程有限公司 | Fire operation closed-loop automatic control method and system based on intelligent overall process monitoring |
| CN114548951A (en) * | 2022-04-22 | 2022-05-27 | 北京信义恒科技有限公司 | Power grid infrastructure ledger visualization system and method based on digital twinning technology |
| CN114995276A (en) * | 2022-06-09 | 2022-09-02 | 中冶南方城市建设工程技术有限公司 | Distributed intelligent sponge monitoring system |
| CN115114268A (en) * | 2022-06-30 | 2022-09-27 | 北京亚控科技发展有限公司 | Tissue future state twinning method, device and equipment |
| CN115412441A (en) * | 2021-05-26 | 2022-11-29 | 中国移动通信有限公司研究院 | Method and device for controlling bandwidth of packet transport network defined by software |
| CN115775092A (en) * | 2022-11-11 | 2023-03-10 | 中电建铁路建设投资集团有限公司 | Construction process safety risk management and control system based on digital twin technology |
| CN116308302A (en) * | 2023-05-19 | 2023-06-23 | 深圳抛物线科技有限公司 | Digital twinning-based advanced warning method and system for potential insulation hazards |
| CN116385690A (en) * | 2023-06-05 | 2023-07-04 | 四川云控交通科技有限责任公司 | BIM model-based three-dimensional operation and maintenance management and control platform and management and control method thereof |
| TWI821994B (en) * | 2022-04-15 | 2023-11-11 | 中華電信股份有限公司 | An area disaster prevention and control system, method and computer-readable medium thereof using digital twin |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070124335A1 (en) * | 2005-11-29 | 2007-05-31 | Park Gwang Woo | Method for quantifying plant resources using gis |
| US20180182029A1 (en) * | 2016-12-22 | 2018-06-28 | American Express Travel Related Services Company, Inc. | Systems and methods for custom ranking objectives for machine learning models applicable to fraud and credit risk assessments |
| CN108256172A (en) * | 2017-12-26 | 2018-07-06 | 同济大学 | Dangerous situation early alarming and forecasting method during existing box culvert is worn under a kind of push pipe |
| CN109657866A (en) * | 2017-12-25 | 2019-04-19 | 广州地理研究所 | The construction method of geological disaster Ecological security assessment model based on maximum entropy model |
-
2019
- 2019-11-08 CN CN201911092036.6A patent/CN110929923B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070124335A1 (en) * | 2005-11-29 | 2007-05-31 | Park Gwang Woo | Method for quantifying plant resources using gis |
| US20180182029A1 (en) * | 2016-12-22 | 2018-06-28 | American Express Travel Related Services Company, Inc. | Systems and methods for custom ranking objectives for machine learning models applicable to fraud and credit risk assessments |
| CN109657866A (en) * | 2017-12-25 | 2019-04-19 | 广州地理研究所 | The construction method of geological disaster Ecological security assessment model based on maximum entropy model |
| CN108256172A (en) * | 2017-12-26 | 2018-07-06 | 同济大学 | Dangerous situation early alarming and forecasting method during existing box culvert is worn under a kind of push pipe |
Non-Patent Citations (2)
| Title |
|---|
| 李欣 等: "数字孪生应用及安全发展综述" * |
| 皱亮: "遥感与GIS在消防规划火宅风险评估中的应用" * |
Cited By (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111443609A (en) * | 2020-04-16 | 2020-07-24 | 成都三业嘉豪科技有限公司 | Laboratory environment self-adaptive adjusting method based on Internet of things |
| CN111445168A (en) * | 2020-04-21 | 2020-07-24 | 深圳民声第三方监管有限公司 | Quality safety third-party supervision system and method |
| CN111563446A (en) * | 2020-04-30 | 2020-08-21 | 郑州轻工业大学 | Human-machine interaction safety early warning and control method based on digital twin |
| CN111723469A (en) * | 2020-05-25 | 2020-09-29 | 贵州华泰智远大数据服务有限公司 | Digital portrait modeling method and system based on twin data |
| CN111680897A (en) * | 2020-05-27 | 2020-09-18 | 公安部道路交通安全研究中心 | Service management and control method, device and system and computer readable storage medium |
| CN111881495A (en) * | 2020-06-24 | 2020-11-03 | 北京工业大学 | Prestress steel structure safety assessment method based on digital twinning |
| CN111931260A (en) * | 2020-08-04 | 2020-11-13 | 灵然创智（天津）动画科技发展有限公司 | City digital twin visualization cooperation method |
| CN112802287A (en) * | 2020-12-16 | 2021-05-14 | 广东电网有限责任公司电力科学研究院 | Power transmission line forest fire monitoring, early warning and positioning system and method thereof |
| CN113051573A (en) * | 2021-02-19 | 2021-06-29 | 广州银汉科技有限公司 | Host safety real-time monitoring alarm system based on big data |
| CN113051573B (en) * | 2021-02-19 | 2021-11-02 | 广州银汉科技有限公司 | Host safety real-time monitoring alarm system based on big data |
| CN113110221A (en) * | 2021-04-29 | 2021-07-13 | 上海智大电子有限公司 | Comprehensive intelligent monitoring method and system for pipe gallery system |
| CN113268867A (en) * | 2021-05-17 | 2021-08-17 | 中国科学院、水利部成都山地灾害与环境研究所 | Glacier debris flow area early warning method based on hydrothermal combination |
| CN113341893A (en) * | 2021-05-26 | 2021-09-03 | 苏州市给排水设计院有限责任公司 | Secondary water supply layered distributed control method, system and storage medium |
| CN115412441A (en) * | 2021-05-26 | 2022-11-29 | 中国移动通信有限公司研究院 | Method and device for controlling bandwidth of packet transport network defined by software |
| CN113378700A (en) * | 2021-06-08 | 2021-09-10 | 四川农业大学 | Grassland rat damage dynamic monitoring method based on unmanned aerial vehicle aerial image |
| CN113792658A (en) * | 2021-09-15 | 2021-12-14 | 江苏迪赛司自动化工程有限公司 | Fire operation closed-loop automatic control method and system based on intelligent overall process monitoring |
| CN113792658B (en) * | 2021-09-15 | 2023-09-01 | 江苏迪赛司自动化工程有限公司 | Automatic control method and system for closed loop of fire operation in whole process intelligent monitoring |
| TWI821994B (en) * | 2022-04-15 | 2023-11-11 | 中華電信股份有限公司 | An area disaster prevention and control system, method and computer-readable medium thereof using digital twin |
| CN114548951A (en) * | 2022-04-22 | 2022-05-27 | 北京信义恒科技有限公司 | Power grid infrastructure ledger visualization system and method based on digital twinning technology |
| CN114995276A (en) * | 2022-06-09 | 2022-09-02 | 中冶南方城市建设工程技术有限公司 | Distributed intelligent sponge monitoring system |
| CN115114268A (en) * | 2022-06-30 | 2022-09-27 | 北京亚控科技发展有限公司 | Tissue future state twinning method, device and equipment |
| CN115114268B (en) * | 2022-06-30 | 2023-08-11 | 北京亚控科技发展有限公司 | Method, device and equipment for organizing future state twinning |
| CN115775092A (en) * | 2022-11-11 | 2023-03-10 | 中电建铁路建设投资集团有限公司 | Construction process safety risk management and control system based on digital twin technology |
| CN115775092B (en) * | 2022-11-11 | 2023-12-22 | 中电建铁路建设投资集团有限公司 | Construction process safety risk management and control system based on digital twin technology |
| CN116308302A (en) * | 2023-05-19 | 2023-06-23 | 深圳抛物线科技有限公司 | Digital twinning-based advanced warning method and system for potential insulation hazards |
| CN116308302B (en) * | 2023-05-19 | 2023-08-08 | 深圳抛物线科技有限公司 | Digital twinning-based advanced warning method and system for potential insulation hazards |
| CN116385690A (en) * | 2023-06-05 | 2023-07-04 | 四川云控交通科技有限责任公司 | BIM model-based three-dimensional operation and maintenance management and control platform and management and control method thereof |
| CN116385690B (en) * | 2023-06-05 | 2023-09-26 | 四川云控交通科技有限责任公司 | BIM model-based three-dimensional operation and maintenance management and control platform and management and control method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110929923B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110929923B (en) | Urban safety risk management and control system based on digital twin technology | |
| CN109145954B (en) | Network taxi appointment travel safety evaluation method and system based on multi-source time-space data | |
| CN112766909A (en) | Intelligent management and control system and method suitable for ship fire safety and computer medium | |
| CN102567808B (en) | Method for forecasting and warning accident consequence of major hazard installation by combining with real-time meteorological information | |
| CN113379267B (en) | Urban fire event processing method, system and storage medium based on risk classification prediction | |
| CN106779185A (en) | A kind of emergency traffic command methods based on harmful influence diffusion model | |
| CN113115225B (en) | Electronic fence area generation system based on dangerous source monitoring and personnel positioning | |
| CN113988530A (en) | Non-coal mine safety risk assessment grading method and system | |
| Preeti et al. | Forest fire prediction using machine learning techniques | |
| Yang et al. | Identifying intercity freight trip ends of heavy trucks from GPS data | |
| US11170217B2 (en) | Method and system for prediction and mitigation of spontaneous combustion in coal stock piles | |
| CN111709661B (en) | Risk processing method, device, equipment and storage medium for business data | |
| CN116739245B (en) | Intelligent fire-fighting city alarm receiving and scheduling system | |
| CN116109462A (en) | Pollution monitoring and early warning method and system for drinking water source area after natural disaster | |
| CN116205504A (en) | Digital emergency plan management method and system | |
| Mi et al. | An integrated method for fire risk assessment in residential buildings | |
| CN115578850A (en) | Urban traffic intelligent control method and system based on traffic informatization | |
| You et al. | PANDA: predicting road risks after natural disasters leveraging heterogeneous urban data | |
| Peng et al. | Urban fire emergency management based on big data intelligent processing system and Internet of Things | |
| CN113887872A (en) | Traffic risk classification method based on situation assessment technology | |
| KR102213204B1 (en) | Method and apparatus for managing firefighting facilities and preventing fires based on gis using artificial neural networks | |
| CN112330129A (en) | Risk assessment and management and control method, system, terminal and storage medium | |
| CN111553199A (en) | Motor vehicle traffic violation automatic detection technology based on computer vision | |
| CN114664041B (en) | Multi-sensor fusion early warning prediction method based on deep learning | |
| CN114495421B (en) | Intelligent open type road construction operation monitoring and early warning method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |