CN114595764A - Method and system for acquiring influence degree of urban factors on inland inundation disaster loss - Google Patents
Method and system for acquiring influence degree of urban factors on inland inundation disaster loss Download PDFInfo
- Publication number
- CN114595764A CN114595764A CN202210221704.6A CN202210221704A CN114595764A CN 114595764 A CN114595764 A CN 114595764A CN 202210221704 A CN202210221704 A CN 202210221704A CN 114595764 A CN114595764 A CN 114595764A
- Authority
- CN
- China
- Prior art keywords
- urban
- network structure
- bayesian network
- influence
- node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000007246 mechanism Effects 0.000 claims abstract description 37
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 23
- 238000010206 sensitivity analysis Methods 0.000 claims abstract description 12
- 238000012549 training Methods 0.000 claims abstract description 9
- 230000001364 causal effect Effects 0.000 claims description 17
- 230000035945 sensitivity Effects 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 6
- 230000002123 temporal effect Effects 0.000 claims description 3
- 238000011158 quantitative evaluation Methods 0.000 abstract description 3
- 230000000875 corresponding effect Effects 0.000 description 15
- 238000004891 communication Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 5
- 238000011160 research Methods 0.000 description 4
- 239000000470 constituent Substances 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 238000007476 Maximum Likelihood Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003780 insertion Methods 0.000 description 2
- 230000037431 insertion Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000011273 social behavior Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/29—Graphical models, e.g. Bayesian networks
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Alarm Systems (AREA)
Abstract
The invention provides a method and a system for acquiring the influence degree of urban elements on waterlogging disaster loss, wherein the method comprises the following steps: constructing a hierarchical Bayesian network structure under the scale of a city domain and a block, and respectively corresponding the structure in the ontology model of the influence mechanism of the city elements on the waterlogging disaster to the hierarchical Bayesian network structure; performing parameter learning training on the hierarchical Bayesian network structure by adopting a parameter learning algorithm to obtain the conditional probability of each node; and carrying out sensitivity analysis on each node according to the conditional probability of each node in the hierarchical Bayesian network structure so as to quantify the influence degree of the urban elements on the waterlogging disaster loss. According to the technical scheme provided by the invention, the influence degree of the urban elements on the loss of the waterlogging disaster can be quantized, so that on one hand, the influence degree of each urban element on the loss of the waterlogging disaster can be visually observed by people, and on the other hand, a quantitative evaluation basis can be provided when the loss degree of the waterlogging disaster is evaluated.
Description
Technical Field
The invention belongs to the technical field of research on influence degree of urban elements on waterlogging disaster loss, and particularly relates to a method and a system for acquiring influence degree of urban elements on waterlogging disaster loss.
Background
Flood disasters are common natural disasters, the flood characteristics of urban flood disasters and drainage basin flood disasters are obviously different, and urban inland inundation causes greater loss compared with drainage basin flood disasters, which is because although dams built around cities can improve the flood resistance risk coefficient of the cities, the increase of urban roads and building density also reduces the water permeable area and the water permeable proportion of the cities, and in addition, due to historical reasons, a plurality of underground pipelines and drainage systems cannot meet the current flood drainage and flood discharge requirements of the cities, the duration of the flood disasters is prolonged, and the severity of the urban disasters is aggravated. And cities have large population, and buildings and various devices are high in quantity and dense, so that once disasters occur, a large amount of economic loss can be caused, and even casualties can be caused.
Relevant researches show that the conditions of the underlying surface, the pipe network distribution condition, the building distribution and the like in urban areas have obvious influence on production convergence, the loss of the waterlogging disaster is positively correlated with urban population and economic distribution density, and the characteristics can provide an optimized direction for risk management and control of the urban waterlogging. However, as a complex huge system coupling nature and human phenomena, the city has the characteristics of highly concentrated population and facility resources, complex spatial structure, high-speed dynamic operation evolution of physical objects and group social behaviors and the like, and the organization structure is the form and the mode of the interrelation and the interaction of all the components of the city. Therefore, the urban flood disaster process and the expression characteristics thereof are more complex, the related elements are wide in range and multiple in types, the influence degree among all the elements cannot be comprehensively compared only by selecting the relevant urban constituent elements from the physical mechanism formed by the urban flood disaster and analyzing the influence of the relevant urban constituent elements on the urban flood disaster, and an effective reference basis cannot be provided for urban risk management decision.
The urban elements have complex and various structures, and the research on the influence mechanism of the urban elements on waterlogging disasters needs to be carried out, so that the acquisition and the calculation of each element and disaster loss data are required. The theory and the method based on big data can effectively utilize the urban elements and the multi-source heterogeneous data of disaster loss, thereby deeply analyzing the influence degree of the urban elements on the flood disaster. Aiming at the problem of integration and utilization of multi-source heterogeneous data of urban elements and disaster losses, the ontology theory is based on semantic description thought, and can integrate and manage data of urban management departments related to urban constituent elements and disaster losses on the basis of establishing and combing complex action relations of the urban elements on waterlogging disasters, so that the accuracy of quantification of influence degree of the urban elements on the waterlogging disasters is effectively improved
Although relatively abundant results are obtained in the research on the influence of urban elements and flood disasters at present, some guidance and suggestions can be provided for the risk management of the urban flood disasters, the influence degree of the urban elements on the flood disasters cannot be specifically reflected, so that the guidance effect on urban flood control and disaster resistance is limited, and sufficient support cannot be provided for the decision problem of the priority management of the urban elements in the risk management process of the urban flood disasters.
Therefore, the relation between urban elements and the influence degree of waterlogging disaster loss obtained in the prior art cannot provide enough support for guiding flood control and disaster resistance of cities.
Disclosure of Invention
The invention aims to provide a method and a system for acquiring the influence degree of urban elements on waterlogging damage, so as to at least solve the problem that the relation between the urban elements and the influence degree of the waterlogging damage in the prior art cannot provide enough support for guiding flood control and disaster resistance of cities.
In order to achieve the above object, in one aspect, the present invention provides a method for obtaining the influence degree of an urban element on the flood damage loss, comprising: constructing a hierarchical Bayesian network structure under the scale of a city domain and a block, and respectively corresponding the structure in the ontology model of the influence mechanism of the city elements on the waterlogging disaster to the hierarchical Bayesian network structure; performing parameter learning training on the hierarchical Bayesian network structure by adopting a parameter learning algorithm to obtain the conditional probability of each node; and carrying out sensitivity analysis on each node according to the conditional probability of each node in the hierarchical Bayesian network structure so as to quantify the influence degree of the urban elements on the waterlogging disaster loss.
According to an embodiment of the present invention, respectively corresponding sets of the ontology model of the influence mechanism of the urban elements on the waterlogging disaster to the hierarchical bayesian network structure comprises: setting nodes in the hierarchical Bayesian network structure according to a concept set in the mechanism ontology model; obtaining a corresponding causal relationship according to the concept semantic relationship and the time relationship in the mechanism ontology model; and setting the edges of each node in the hierarchical Bayesian network structure according to the causal relationship.
Further in accordance with another embodiment of the present invention, the concept semantic relationships include whole/part relationships, positive impact relationships, negative impact relationships, both positive and negative impact relationships, and the temporal relationships include earlier and later.
According to yet another embodiment of the invention, the parameter learning algorithm is an expectation maximization algorithm, comprising: calculating an expectation of log-likelihood functions for nodes in the hierarchical bayesian network structure, and calculating parameters to maximize a full likelihood expectation.
According to another embodiment of the present invention, said analyzing the sensitivity of each node according to its conditional probability in the hierarchical bayesian network structure comprises: and acquiring the sensitivity of the waterlogging loss index node to the urban element index node, and acquiring the influence degree of the urban element index node on the waterlogging disaster loss according to the sensitivity.
In another aspect, the present invention further provides a system for obtaining the influence degree of an urban element on the inland inundation disaster damage, including a processor and a memory, where the memory stores computer program instructions for execution on the processor, and when the processor executes the computer city instructions, the following method for obtaining the influence degree of the urban element on the inland inundation damage is implemented: constructing a hierarchical Bayesian network structure under the scale of a city domain and a block, and respectively corresponding the structure in the ontology model of the influence mechanism of the city elements on the waterlogging disaster to the hierarchical Bayesian network structure; performing parameter learning training on the hierarchical Bayesian network structure by adopting a parameter learning algorithm to obtain the conditional probability of each node; and carrying out sensitivity analysis on each node according to the conditional probability of each node in the hierarchical Bayesian network structure so as to quantify the influence degree of the urban elements on the waterlogging disaster loss.
According to an embodiment of the present invention, respectively corresponding sets of the ontology model of the influence mechanism of the urban elements on the waterlogging disaster to the hierarchical bayesian network structure comprises: setting nodes in the hierarchical Bayesian network structure according to a concept set in the mechanism ontology model; obtaining a corresponding causal relationship according to the concept semantic relationship and the time relationship in the mechanism ontology model; and setting the edges of each node in the hierarchical Bayesian network structure according to the causal relationship.
Further in accordance with another embodiment of the present invention, the concept semantic relationships include whole/part relationships, positive impact relationships, negative impact relationships, both positive and negative impact relationships, and the temporal relationships include earlier and later.
According to yet another embodiment of the invention, the parameter learning algorithm is an expectation maximization algorithm, comprising: calculating an expectation of log-likelihood functions for nodes in the hierarchical bayesian network structure, and calculating parameters to maximize a full likelihood expectation.
According to another embodiment of the present invention, said analyzing the sensitivity of each node according to its conditional probability in the hierarchical bayesian network structure comprises: and acquiring the sensitivity of the waterlogging loss index node to the urban element index node, and acquiring the influence degree of the urban element index node on the waterlogging disaster loss according to the sensitivity.
The invention has the beneficial effects that: according to the technical scheme provided by the invention, the influence degree of the urban elements on the loss of the waterlogging disaster can be quantized, so that on one hand, the influence degree of each urban element on the loss of the waterlogging disaster can be visually observed by people, and on the other hand, a quantitative evaluation basis can be provided when the loss degree of the waterlogging disaster is evaluated. In addition, the technical scheme of the invention adopts ontology theory to better establish the relation between knowledge and data, is an effective method for integrating multi-source heterogeneous data of the urban elements and establishing the influence relation of the urban elements on the internal consumption disaster, and the hierarchical Bayesian network structure under the scale of the urban area and the block can be effectively integrated with an ontology model, so that the method is an effective method for quantifying the influence degree of the urban elements on the internal waterlogging disaster by using the multi-source heterogeneous data.
Drawings
The above and other objects, features and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the present invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar or corresponding parts and in which:
FIG. 1 is a flow chart of a method for obtaining a degree of influence of an urban element on an inlaying water disaster loss according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a system for obtaining the influence degree of urban elements on the loss of an inland water disaster according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present application are clearly and completely described below with reference to the drawings in the embodiments of the present application, and it should be understood by those skilled in the art that the embodiments described below are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Referring to fig. 1, fig. 1 shows a method for obtaining influence of an urban element on an inland inundation disaster loss according to the present application, which quantifies the influence of the urban element on the inland inundation disaster loss by using a bayesian network, so as to obtain the influence of the urban element on the inland inundation disaster loss. The method for obtaining the influence degree of the urban elements on the flood damage loss according to the present application is described in detail below with reference to the flow shown in fig. 1.
As shown in fig. 1, the method for obtaining the influence degree of the urban elements on the flood damage includes:
in step S1, a hierarchical bayesian network structure at the city and neighborhood scale is constructed. In this embodiment, when constructing the hierarchical bayesian network structure under the city domain and the street scale, a mechanism ontology model of the influence of the city elements on the waterlogging disaster may be obtained first, then the structure in the mechanism ontology model is respectively corresponding to the established hierarchical bayesian network structure, in the corresponding process, the loss of the waterlogging disaster in the city is taken as the root node of the hierarchical bayesian network structure under the city domain and the street scale, the city factors related to the waterlogging disaster in the city are taken as the leaf nodes of the hierarchical bayesian network structure under the city domain and the street scale, and the parent node and the child nodes in the hierarchical bayesian network structure under the city domain and the street scale are set in the causal relationship and the time relationship. The hierarchical Bayesian network structure constructed by the embodiment has Bayesian nodes under the city scale, Bayesian network structure nodes under the block scale and the same Bayesian nodes in the city domain and the block scale.
In this embodiment, let AcRepresenting a collection of Bayesian network structure nodes at the market scale, ASRepresenting a collection of Bayesian network structure nodes at the street level, ApRepresenting the same set of bayesian network structure nodes in the city domain and the neighborhood scale, the resulting bayesian network is:
A＝{Ac,As,Ap} (1)。
in addition, in step S1, the mechanism ontology model of the influence of the urban elements on the waterlogging disaster not only can fuse the data integration process and the influence relationship of the urban elements on the loss of the waterlogging disaster, so as to realize efficient utilization of data resources, but also can provide convenience for quantifying the influence relationship of the urban elements on the loss of the waterlogging disaster. In this embodiment, the method for constructing the mechanism ontology model of the influence of the urban elements on the waterlogging disaster includes:
(1) and (3) analyzing the demand: the method comprises the steps of establishing an influence relation of urban elements on the loss of the waterlogging disaster and integrating data, determining the construction requirement of an influence mechanism ontology model of the urban elements on the waterlogging disaster, and integrating the influence relation of the urban elements on the loss of the waterlogging disaster and the data relation of the influence relation on the loss of the waterlogging disaster on the basis of realizing the data integration management of the urban elements and the loss index of the waterlogging disaster.
(2) Data analysis and collection: and collecting urban elements of the dimensions of a city domain and a block, and data and related information of the waterlogging disaster loss indexes.
(3) Extracting information, concepts and attributes: according to an urban element identification result and an index system related to urban inland inundation disasters, analyzing concepts and data attributes (such as time, space and the like) of urban elements and inland inundation disaster losses, and defining a semantic concept system and concept attributes of an ontology model;
(4) judging the classification and relationship between concepts: constructing a hierarchical structure of an ontology model concept according to the composition condition of the urban elements and the influence mechanism content of the urban elements on the waterlogging disaster loss; establishing an influence relation of urban elements under the scale of a city domain and a block on waterlogging disaster loss, and determining a semantic relation of an ontology model; and judging the data relationship between the urban elements and the inland inundation disaster loss, and determining the time relationship and the space relationship of the ontology model data.
(5) And (3) inspecting the structure of the body: and (3) checking whether the influence mechanism ontology of the urban elements on the waterlogging disaster meets the standard by using an ontology construction criterion, wherein the influence mechanism ontology comprises five criteria of whether the concept is clear, whether the concept is consistent before and after, whether the concept has expandability, whether the principle meets the minimum coding preference degree principle and whether the minimum ontology agrees. If not, returning to the concept and attribute extraction stage until the requirements are met.
(6) Constructing an ontology model: and (3) constructing an influence mechanism ontology model of the urban elements on the waterlogging disaster by using model construction software (such as Prot g e).
(7) And (3) storing the ontology model: and the body is filed, so that subsequent reuse is facilitated.
In step S2, a parameter learning algorithm is used to perform parameter learning training on the constructed hierarchical bayesian network structure under the city domain and the neighborhood scale, so as to obtain the conditional probability of each node therein. Since each node in the hierarchical bayesian network structure under the city domain and the neighborhood scale constructed in step S1 only represents a causal relationship, but it has not been trained yet, and the parameters therein have not been solidified, in order to perform parameter learning training on the node in step S2 by using parameter learning algorithm, the parameters thereof are solidified, and then the conditional probability of each node therein is obtained.
In step S3, sensitivity analysis is performed on each node in the hierarchical bayesian network structure according to the conditional probability of the city domain and the block scale, so as to quantify the influence degree of the city elements on the inland disaster damage. The sensitivity analysis of the hierarchical Bayesian network structure under the city and block scales means that the influence on the output of the hierarchical Bayesian network structure under the city and block scales is analyzed after the parameters in the hierarchical Bayesian network structure are changed. In this embodiment, the influence degree of the urban element on the loss of the waterlogging disaster can be quantified by the sensitivity degree of the root node to the parent node in the urban area and block scale lower-level bayesian network structure, wherein the higher the sensitivity of a certain urban element on the loss of the waterlogging disaster is, the higher the influence degree of the urban element on the loss of the waterlogging disaster is. In the sensitivity analysis process of the internal and network models, the quantification of variable sensitivity can be realized by calculating and determining a sensitivity function, so that the influence degree of urban elements on the waterlogging disaster loss is quantified.
In summary, according to the technical scheme provided by the application, the influence degree of the urban elements on the loss of the waterlogging disaster can be quantified, so that on one hand, the influence degree of each urban element on the loss of the waterlogging disaster can be visually observed by people, and on the other hand, quantitative evaluation basis can be provided when the loss degree of the waterlogging disaster is evaluated. In addition, the ontology theory can be used for well establishing the relation between knowledge and data, the method is an effective method for integrating multi-source heterogeneous data of the urban elements and establishing the influence relation of the urban elements on the internal consumption disasters, and the hierarchical Bayesian network structure under the urban area and block scale can be effectively integrated with the ontology model, so that the method is an effective method for quantifying the influence degree of the urban elements on the internal waterlogging disasters by using the multi-source heterogeneous data.
The method for obtaining the influence degree of the urban elements on the flood damage loss is introduced in an integrated manner, and the method for respectively corresponding the concept sets in the mechanism ontology model to the established city domain and block scale lower-level bayesian network structure is described in the following in combination with specific application scenarios, and it can be understood that the description of the method in the following is exemplary and non-limiting, so that the description of the concept sets in the mechanism ontology model in the foregoing respectively corresponding to the established city domain and block scale lower-level bayesian network structure is also applicable to the description of the method in the following.
In one embodiment, in step S1, the method for corresponding the concept sets in the mechanism ontology model to the established hierarchical bayesian network structure under the city domain and the block scale includes: setting nodes in a hierarchical Bayesian network structure according to a concept set in the mechanism ontology model: the method comprises the steps of firstly obtaining causal relations between nodes in a hierarchical Bayesian network structure under a city domain and a block scale according to concept semantic relations and time relations in a mechanism ontology model, and then setting edges of the nodes in the hierarchical Bayesian network structure under the city domain and the block scale according to the causal relations between the nodes. The concept semantics refers to the relationship between concepts in the ontology model, such as the causal relationship between a city element and a waterlogging disaster, and the time relationship is the chronological order of the occurrence of time. When the nodes in the hierarchical Bayesian network structure in the city domain and the neighborhood scale are set according to the causal relationship among the nodes, if one node is the cause of the other node, the node is the father node of the other node. Through the setting mode of the embodiment, the mechanism ontology model of the influence of the urban elements on the waterlogging disaster can be corresponding to the hierarchical Bayesian network structure under the urban area and the street scale, so that the relation between the urban elements and the waterlogging disaster loss can be represented in the hierarchical Bayesian network structure under the urban area and the street scale.
Further, in another embodiment, in the ontology model of influence mechanism of the urban elements on the waterlogging disaster, the concept-semantic relationship system includes: whole/part relationships, positive influence relationships, negative influence relationships, both positive and negative influence relationships. In the established concept semantic relation system, one-way relations of urban elements to waterlogging disaster losses are constructed, and the one-way relations can be understood as causal relations taking the urban elements as reasons and taking the waterlogging disaster losses as effects, so that the concept semantics of the urban elements to the urban waterlogging disaster ontology model can be converted into the causal relations of corresponding nodes in the Bayesian network, for example, the nodes are partial father nodes as a whole, if one node is the positive influence relation of another node, the node is the father node of the other node, otherwise, if one node is the negative influence relation of the other node, the node is the child node of the other node. In addition, the influence mechanism ontology model index data time relationship of the urban elements on the waterlogging disaster in the influence mechanism ontology model comprises a causal relationship which is earlier than and later than and is obtained by obtaining corresponding nodes in the hierarchical bayesian network structure under the urban area and block scale, for example, if the time of one node is earlier than that of another node, the node is taken as a parent node of the other node, otherwise, if the time of the node is later than that of the other node, the node is taken as a child node of the other node.
The parameter learning algorithm in step S2 may be a maximum likelihood estimation algorithm, a gradient ascent algorithm, or an expectation maximization algorithm, and hereinafter, the expectation maximization algorithm is taken as an example, and a method for parameter learning of a city domain and block scale lower-level bayesian network structure constructed by training is described in detail, and it is understood that the description of the method is exemplary and not limiting.
In an embodiment, the method for performing parameter learning on the constructed hierarchical bayesian network structure under the urban area and the block scale by using the expectation-maximization algorithm in step S2 includes: calculating the expectation of the log-likelihood function of the nodes in the hierarchical Bayesian network structure under the city domain and the block scale, and calculating the parameter which maximizes the expectation of the likelihood completely. In this embodiment, assuming that the likelihood function of a node B sample in the hierarchical bayesian network structure under the city domain and the street scale is L (θ, D), the log likelihood function of the likelihood function is logL (θ, D), where θ represents the set of node B parameters, D represents the sample data of the node B, and then the expected Q (θ, θ) of the log likelihood function of the node B is calculatedk) The formula adoptedComprises the following steps:
Q(θ,θk)＝Eθ(k){logL(θ,D)} (2)
where θ ═ θ1，θ2，...，θk，...，θk+m}，θkFor the kth parameter in the node B, k and m are positive integers.
In the expectation maximization algorithm, when calculating a parameter that maximizes the complete likelihood expectation, θ is first assumed to existk+1C, the following formula can be made to
The following holds true:
Q(θk+1,θk)≥Q(θ,θk) (3)
wherein C is a set of settings, the steps are iterated as follows in equation (4) to produce a sequence of parameter values { θ }k}, to obtain θk+1Until convergence, the maximum likelihood function L (theta) of the value space theta of the parameters is obtainedk+1|D)：
θk+1＝arg maxQ(θ,θk) (4)
Through the steps, under a given sample D, the conditional probability of a city domain and a Bayesian node connected to a lower level of the size is as follows:
P(D|θk+1)＝L(θk+1|D) (5)。
according to the method, the probability of the child node under different states of the father node can be obtained, and the parameter learning training of the hierarchical Bayesian network structure of the city domain scale and the block scale is completed according to the probability of different disaster loss degrees under different city element index conditions.
In the above, a method for performing parameter learning on the constructed hierarchical bayesian network structure under the city domain and the neighborhood scale by using the expectation-maximization algorithm is introduced in detail, and a method for performing sensitivity analysis on each node in the hierarchical bayesian network structure under the city domain and the neighborhood scale according to the conditional probability of each node is described in detail below in combination with a specific application scenario.
In one embodiment, according to municipalitiesThe method for carrying out sensitivity analysis on each node in the hierarchical Bayesian network structure under the domain and block scale by the conditional probability comprises the following steps: firstly, the sensitivity of node parameters is obtained, and then the influence degree between the nodes is obtained according to the sensitivity of the node parameters. In the hierarchical bayesian network structure under the scale of urban area and block in the embodiment, the index node a of the flood disaster lossOAt city element index node AeUnder the condition of (1), the probability of taking a specific value is P (a | b), wherein a is an index node A of the inland inundation disaster damageOB is a city element index node AeValue of (2), city element index node AeHas a parameter of psi-P (c | pi) where c is an arbitrary city element index node ahIs a node A of the index of the city elementhThe sensitivity of the parameter psi is then:
wherein P (a | b) (ψ) is a function of P (a | b) and ψ ═ P (c | pi), and
m in the above formula1、m2、m3、m4Is a fixed coefficient.
With the change of the parameter psi, the city element index node AhThe value of the parameter σ ═ P (c '| pi) formed by other values of (a) is changed along with the parameter σ ═ P (c' | pi), so as to ensure that the sum of the probabilities of all the values is 1.
Let the index node A of city elementhLoss index node A relative to waterlogging disasterOHas an influence of IM (A)h) Namely, the node A for disaster damage index of waterloggingOFor city element index node AhSensitivity of (A) is IM (A)h) And then:
in the formula: r and s respectively represent index nodes A of waterlogging disaster lossOAnd city element index node AeThe value of (a) (#)ijIndex node A for inland inundation disaster lossOMiddle ith parameter and city element index node AeThe j-th parameter, IM (A)h) Index node A for representing waterlogging disaster lossOFor city element index node AhSensitivity of (2) is in the range of [0,1 ]]，IM(Ah) The larger the value is, the larger the city element index node AhInternal waterlogging disaster loss index node AOThe higher the degree of influence of (c).
According to another aspect of the present application, there is also provided a system for obtaining the degree of influence of urban elements on the loss of inland water disaster, as shown in fig. 2, the system includes a processor, an inserter, a communication interface and a communication bus, and the processor, the inserter and the communication interface complete mutual communication through the communication bus. The processor is used to provide computing and control capabilities. The interposer includes a nonvolatile interposer medium, an interposer. The non-volatile insertion medium has inserted therein an operating system and computer program instructions. The interposer provides an environment for the execution of operating system and computer program instructions in a non-volatile insertion medium. The communication interface of the device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The system for obtaining the influence degree of the urban element on the inland inundation disaster damage provided by the embodiment is provided, and the inserter is used for inserting computer program instructions, and the computer program instructions enable the processor to execute the method for obtaining the influence degree of the urban element on the inland inundation disaster damage and the plurality of embodiments thereof.
In light of the foregoing description of the present specification, those skilled in the art will also understand that terms used to indicate orientation or positional relationship, such as "front," "rear," "left," "right," "top," "bottom," etc., are based on the orientation or positional relationship shown in the drawings of the present specification, which are used for convenience in explaining aspects of the present invention and for simplicity in description, and do not explicitly or implicitly indicate that the device or element involved must have the particular orientation, be constructed and operated in the particular orientation, and thus should not be interpreted or construed as limiting the aspects of the present invention.
In addition, the terms "first" or "second", etc. used in this specification are used to refer to numbers or ordinal terms for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the feature. In the description of the present specification, "a plurality" means at least two, for example, two, three or more, and the like, unless specifically defined otherwise.
While various embodiments of the present invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, changes, and substitutions will occur to those skilled in the art without departing from the spirit and scope of the present invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention. It is intended that the following claims define the scope of the invention and that the module compositions, equivalents, or alternatives falling within the scope of these claims be covered thereby.
Claims (10)
1. A method for obtaining the influence degree of urban elements on waterlogging disaster loss is characterized by comprising the following steps:
constructing a hierarchical Bayesian network structure under the scale of a city domain and a block, and respectively corresponding the structure in the ontology model of the influence mechanism of the city elements on the waterlogging disaster to the hierarchical Bayesian network structure;
performing parameter learning training on the hierarchical Bayesian network structure by adopting a parameter learning algorithm to obtain the conditional probability of each node;
and carrying out sensitivity analysis on each node according to the conditional probability of each node in the hierarchical Bayesian network structure so as to quantify the influence degree of the urban elements on the waterlogging disaster loss.
2. The method for obtaining the influence degree of the urban elements on the waterlogging disaster damage according to claim 1, wherein the step of respectively corresponding the sets of the urban elements on the waterlogging disaster influence mechanism ontology model to the hierarchical bayesian network structure comprises:
setting nodes in the hierarchical Bayesian network structure according to a concept set in the mechanism ontology model;
obtaining a corresponding causal relationship according to the concept semantic relationship and the time relationship in the mechanism ontology model;
and setting the edges of each node in the hierarchical Bayesian network structure according to the causal relationship.
3. The method for obtaining influence of urban elements on waterlogging damage according to claim 2, wherein said concept-semantic relationships include whole/partial relationships, positive influence relationships, negative influence relationships, both positive and negative influence relationships, and said temporal relationships include earlier and later.
4. The method for obtaining the influence degree of the urban elements on the flood damage according to claim 1, wherein the parameter learning algorithm is an expectation maximization algorithm, and comprises: calculating an expectation of log-likelihood functions for nodes in the hierarchical bayesian network structure, and calculating parameters to maximize a full likelihood expectation.
5. The method for obtaining the influence degree of the urban elements on the flood damage loss according to claim 1, wherein the performing the sensitivity analysis on the nodes according to the conditional probability of each node in the hierarchical bayesian network structure comprises: and acquiring the sensitivity of the waterlogging loss index node to the urban element index node, and acquiring the influence degree of the urban element index node on the waterlogging disaster loss according to the sensitivity.
6. A system for obtaining a degree of influence of an urban element on waterlogging disaster damage, comprising a processor and a memory, the memory storing computer program instructions for execution on the processor, when executing the computer city instructions, implementing the following method for obtaining a degree of influence of an urban element on waterlogging damage:
constructing a hierarchical Bayesian network structure under the scale of a city domain and a block, and respectively corresponding the structure in the ontology model of the influence mechanism of the city elements on the waterlogging disaster to the hierarchical Bayesian network structure;
performing parameter learning training on the hierarchical Bayesian network structure by adopting a parameter learning algorithm to obtain the conditional probability of each node;
and carrying out sensitivity analysis on each node according to the conditional probability of each node in the hierarchical Bayesian network structure so as to quantify the influence degree of the urban elements on the waterlogging disaster loss.
7. The system for obtaining the influence degree of the urban elements on the waterlogging disaster damage according to claim 6, wherein the step of respectively corresponding the sets of the urban elements on the waterlogging disaster influence mechanism ontology model to the hierarchical Bayesian network structure comprises:
setting nodes in the hierarchical Bayesian network structure according to a concept set in the mechanism ontology model;
obtaining a corresponding causal relationship according to the concept semantic relationship and the time relationship in the mechanism ontology model;
and setting the edges of each node in the hierarchical Bayesian network structure according to the causal relationship.
8. The system for obtaining the influence degree of the urban elements on the flood disaster damage according to claim 7, wherein the concept semantic relations comprise whole/partial relations, positive influence relations, negative influence relations, both positive and negative influence relations, and the time relations comprise earlier and later.
9. The system for obtaining the influence degree of urban elements on waterlogging disaster damage according to claim 6, wherein the parameter learning algorithm is an expectation maximization algorithm comprising: calculating an expectation of log-likelihood functions for nodes in the hierarchical bayesian network structure, and calculating parameters to maximize a full likelihood expectation.
10. The system for obtaining influence degree of urban elements on waterlogging disaster damage according to claim 6, wherein the sensitivity analysis of the nodes in the hierarchical Bayesian network structure according to their conditional probabilities comprises: and acquiring the sensitivity of the waterlogging loss index node to the urban element index node, and acquiring the influence degree of the urban element index node on the waterlogging disaster loss according to the sensitivity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210221704.6A CN114595764A (en) | 2022-03-09 | 2022-03-09 | Method and system for acquiring influence degree of urban factors on inland inundation disaster loss |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210221704.6A CN114595764A (en) | 2022-03-09 | 2022-03-09 | Method and system for acquiring influence degree of urban factors on inland inundation disaster loss |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114595764A true CN114595764A (en) | 2022-06-07 |
Family
ID=81807751
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210221704.6A Pending CN114595764A (en) | 2022-03-09 | 2022-03-09 | Method and system for acquiring influence degree of urban factors on inland inundation disaster loss |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114595764A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118037062A (en) * | 2024-04-10 | 2024-05-14 | 暗物智能科技(广州)有限公司 | Waterlogging disaster risk assessment system based on causal sum or graph |
-
2022
- 2022-03-09 CN CN202210221704.6A patent/CN114595764A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118037062A (en) * | 2024-04-10 | 2024-05-14 | 暗物智能科技(广州)有限公司 | Waterlogging disaster risk assessment system based on causal sum or graph |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110968701A (en) | Relationship map establishing method, device and equipment for graph neural network | |
| US11741373B2 (en) | Turbulence field update method and apparatus, and related device thereof | |
| CN111091247A (en) | Power load prediction method and device based on deep neural network model fusion | |
| CN113077090A (en) | Passenger flow prediction method, system and computer readable storage medium | |
| CN107886160B (en) | BP neural network interval water demand prediction method | |
| CN108090677B (en) | Reliability evaluation method for key infrastructure | |
| CN111160959B (en) | User click conversion prediction method and device | |
| CN117078048B (en) | Digital twinning-based intelligent city resource management method and system | |
| CN114898121A (en) | Concrete dam defect image description automatic generation method based on graph attention network | |
| CN115775116B (en) | BIM-based road and bridge engineering management method and system | |
| WO2023207013A1 (en) | Graph embedding-based relational graph key personnel analysis method and system | |
| CN116307215A (en) | Load prediction method, device, equipment and storage medium of power system | |
| CN111198897A (en) | Scientific research hotspot topic analysis method and device and electronic equipment | |
| CN112749791A (en) | Link prediction method based on graph neural network and capsule network | |
| CN116489038A (en) | Network traffic prediction method, device, equipment and medium | |
| CN114595764A (en) | Method and system for acquiring influence degree of urban factors on inland inundation disaster loss | |
| CN106816871B (en) | State similarity analysis method for power system | |
| CN106550387B (en) | A kind of wireless sensor network routing layer QoS evaluating method | |
| CN114357160A (en) | Early rumor detection method and device based on generation propagation structure characteristics | |
| CN114566048A (en) | Traffic control method based on multi-view self-adaptive space-time diagram network | |
| CN110362831B (en) | Target user identification method, device, electronic equipment and storage medium | |
| Wang et al. | Air quality data analysis and forecasting platform based on big data | |
| CN116501924B (en) | Graph link prediction method based on robust enhancement loss function | |
| CN116933896B (en) | Super-parameter determination and semantic conversion method, device, equipment and medium | |
| CN116188834B (en) | Full-slice image classification method and device based on self-adaptive training model |
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 |