CN115345411B - Matrix fusion algorithm-based dam break emergency plan field ontology evolution method - Google Patents
Matrix fusion algorithm-based dam break emergency plan field ontology evolution method Download PDFInfo
- Publication number
- CN115345411B CN115345411B CN202210478625.3A CN202210478625A CN115345411B CN 115345411 B CN115345411 B CN 115345411B CN 202210478625 A CN202210478625 A CN 202210478625A CN 115345411 B CN115345411 B CN 115345411B
- Authority
- CN
- China
- Prior art keywords
- ontology
- new
- emergency plan
- evolution
- dam
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
- G06Q10/06—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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/374—Thesaurus
-
- 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)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Animal Behavior & Ethology (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a matrix fusion algorithm-based dam-break emergency plan field ontology evolution method, which is characterized in that through the evolution of a dam-break emergency plan field ontology model, the logic relation among elements in the field ontology is modified, so that the dam-break emergency plan field ontology knowledge base is changed, the uncertainty of emergency elements is eliminated, the dynamic adjustment of the emergency plan is realized, the uncertainty in the dam-break emergency is effectively overcome, and a scientific basis is provided for the dynamic adjustment of the dam-break emergency plan under complex live-action.
Description
Technical Field
The invention belongs to the field of dam risk management of hydraulic engineering technology, and particularly relates to a matrix fusion algorithm-based dam break emergency plan field ontology evolution method.
Background
China is the country with the largest number of reservoir dams in the world, and has built various reservoir dams 98112, and the total reservoir capacity is 8983 hundred million m 3 . The water reservoir dams are national important infrastructure, are basic guarantees of economic and social development and implementation of national important strategy, are important carriers for ecological civilization construction, and play various functional benefits of flood control, power generation, irrigation, water supply and the like. However, along with superposition of the operation time of the reservoir dam, the complexity of engineering operation, the uncertainty of operation environment change and the influence of some external acting forces, such as congenital deficiency, weak management, engineering aging and other multiple factors, the possibility of abnormal or even breaking of the reservoir dam exists for a long time, and the life and property safety of people in downstream areas is seriously threatened.
The dam risk management system in China is still in a development stage, is relatively behind in the construction of non-engineering measures for preventing and controlling dam break risks, most of water reservoirs, particularly small-sized reservoirs, are weak in emergency plans, corresponding emergency management response mechanisms are imperfect, and the dam break events can occur at intervals when the corresponding emergency management response mechanisms face emergency events. It can be seen that the emergency plan has a significant effect on reducing dam break losses.
With the rapid development of economy and society and the increasing safety awareness of China society and public, casualties and serious economic losses possibly caused by dam break are not tolerated. The emergency plan is one of the most effective and important tools for emergency management work, and needs to have stronger specialty and pertinence, and more emphasis is placed on the possibility of dam-break emergencies and the scientificity and accuracy of analysis of the results thereof. However, the dam break emergency has the characteristics of disaster propagation, induced complexity, multi-category treatment and the like, and the dam break emergency faces a large amount of uncertainty in the actual treatment situation. Emergency plan uncertainty comprises subjective uncertainty of people about things awareness, uncertainty and the like, and objective uncertainty of flood evolution caused by influence of various factors such as hydrology, weather, geography and human interference. The uncertainty of dam break emergency is eliminated, and the feasibility and the accuracy of the dam break emergency plan are improved, so that the dam break emergency plan has important significance for reservoir dam risk management and control and emergency disposal.
The body theory realizes the acquisition of unstructured knowledge, and has certain application in the field of emergency plans, but the study on the reservoir dam-break emergency plan is less, and most of the body theory is to construct a body model aiming at the knowledge of personnel structures, emergency treatment, emergency response and the like in the emergency plan. In the application of the prior ontology theory in emergency plans, the method only aims at the construction of an ontology model, and is not involved in the treatment of uncertainty in the dam-break emergency plans and a corresponding ontology evolution method.
Disclosure of Invention
The invention aims to provide a method for body evolution in the field of dam-break emergency plans, aiming at the defects in the prior art, so as to realize uncertainty analysis of body theory in application of the dam-break emergency plans and dynamic adjustment of the emergency plans.
In order to achieve the aim of the invention, the invention provides a matrix fusion algorithm-based dam break emergency plan field ontology evolution method, which is characterized by comprising the following steps:
step one, constructing a dam break emergency plan frame
Forming a text of the dam-break emergency plan based on body theory, constructing a basic vocabulary term table related to the dam-break emergency plan and a body knowledge base among the basic vocabulary terms, defining rules inside the body knowledge base by using body language, and establishing a body model in the field of the dam-break emergency plan;
step two, uncertainty factor change capture
Collecting uncertainty factors which cause the emergency plan to change;
step three, body evolution operation
According to the evolution source of the collected uncertainty factors, fusing a new ontology with the source ontology, continuously acquiring new attribute values, concepts and relations in the field, extracting basic vocabulary terms and relations among the basic vocabulary terms, and fusing the basic vocabulary terms into a dam break emergency plan implicit field ontology model through a matrix fusion algorithm to convert the basic vocabulary terms and relations into a dam break emergency plan explicit field ontology model;
step four, ontology evolution management
And (3) propagating and diffusing the ontology changes, integrating the newly generated ontology model into an ontology knowledge base, and storing the ontology knowledge base in a data form.
The invention has the beneficial effects that:
the invention provides a matrix fusion algorithm-based dam-break emergency plan field ontology evolution method, which modifies the logic relationship among elements in the field ontology through the evolution of the dam-break emergency plan field ontology model, thereby changing the dam-break emergency plan field ontology knowledge base, eliminating the uncertainty of emergency elements, realizing the dynamic adjustment of the emergency plan, effectively aiming at the uncertainty in the dam-break emergency, and providing scientific basis for the dynamic adjustment of the dam-break emergency plan under complex live-action.
Drawings
FIG. 1 is a schematic diagram of an analysis of ontology evolution consistency;
FIG. 2 is a schematic diagram of an ontology matrix fusion model;
FIG. 3 is a bulk O 1 experiment A structural diagram;
FIG. 4 is a diagram of a body O 2 experiment A structural diagram;
fig. 5 is a diagram of a body structure of s=0;
fig. 6 is a diagram of a body structure of s=1;
fig. 7 is a diagram of the body structure of s=3;
FIG. 8 is an initial body structure diagram;
FIG. 9 is a diagram of a body to be fused O To be treated A structural diagram;
FIG. 10 is a diagram of the body structure after fusion;
fig. 11 is a core concept of implicit domain ontology model extraction.
Detailed Description
Example 1
The embodiment provides a matrix fusion algorithm-based dam break emergency plan field ontology evolution method, which comprises the following steps:
step one, constructing a dam break emergency plan frame
Based on ontology theory, the text of the dam-break emergency plan is formalized, a basic vocabulary term table related to the dam-break emergency plan and an ontology knowledge base among the vocabularies are constructed, rules inside the basic vocabulary term table are defined by using ontology language, and an ontology model in the field of the dam-break emergency plan is built.
The body model in the dam-break emergency plan field comprises two parts: the system is a dominant field ontology model, the part of the field ontology is mainly content which can be visually presented by an emergency plan, and the main basis of extraction is related guidance rules, laws and regulations and technical standards in the field. In this embodiment, the core concept of dominant domain ontology model extraction is shown in table 1. Secondly, the model of the body in the implicit field is not visually displayed, and is an implicit knowledge expression of the element information of the plan, and the extraction is mainly based on subjective uncertainty and objective uncertainty derived from dam break emergency. Knowledge affected by uncertainty factors such as dam break results, emergency evacuation time and the like is classified into the implicit field. The subjective uncertainty is mainly in the aspects of emergency evacuation and emergency resource management, such as emergency evacuation routes, emergency evacuation time, emergency resource requirements, uncertainty of emergency resource allocation and the like; the objective uncertainty is mainly reflected in aspects of space-time difference of rainfall, data observation, warehousing flood grade, dam break mode, breach flow, uncertainty of flood process and hydraulic parameters and the like. In this embodiment, the core concept of implicit domain ontology model extraction is shown in fig. 11.
Table 1 core concept of dominant domain ontology model and its subordinate concept
Computers cannot handle natural language statements, so ontologies act as an intermediate architecture. The ontology formalizes the text of the dam-break emergency plan, constructs a basic vocabulary glossary about the dam-break emergency plan and an ontology knowledge base among the vocabularies, uses the Web ontology language (OWL) of the ontology to define rules inside the ontology to carry out ontology modeling, and provides a semantic basis for concept understanding in an intelligent analysis system of the emergency plan. The ontology knowledge base thus corresponds to a dictionary, and the knowledge is words in the ontology model.
Step two, uncertainty factor change capture
The change of the body is mainly divided into two cases:
(1) user-driven changes, which are determined by specific requirements or other uncertainties of the user during actual use, are required by the application system or other software according to external changes, so that the source body is updated, and the expert discovers and requests the change operation;
(2) when the attribute value of one index in the initial body of the dam-break emergency plan changes to further change the knowledge source, a change request of body evolution is needed.
Therefore, the ontology evolution needs to capture uncertainty factors, namely, collect concepts, relations, instance data and the like which cause the emergency plan to change, such as the number of evacuees, a dam break mode, a flood level and the like.
Step three, body evolution operation
After the capture of the uncertainty factor change is completed, the new ontology and the source ontology are fused according to the evolution source of the uncertainty factor change, new attribute values, concepts and relations in the field are continuously acquired, basic vocabulary terms and relations in the new ontology are extracted and fused into the dam break emergency plan recessive field ontology through a matrix fusion algorithm, and the new ontology is converted into a dominant field ontology model.
3.1 converting the implicit Domain ontology of the initial Emergency plan into OWL form using the ontology editing tool Prot e, and recording as ontology O 1 The implicit domain ontology for the change caused by the uncertainty of the emergency plan is denoted as O 2 。
3.2 bulk fusion is to fuse O 1 With O 2 Merging according to a certain rule, converting two bodies to be fused into corresponding structural matrixes according to a body fusion algorithm, and obtaining the structural matrix F of the new fused body through operation calculation of the matrix fusion algorithm n Finally, the matrix is converted into a body structure form, so that the implicit field body evolution is realized, and the field body which is more in line with the current situation is generated.
The ontology fusion algorithm is as follows:
according to the fusion algorithm model of the matrix and the ontology, the algorithms corresponding to different concept numbers s are different, and specific algorithms are divided into two types.
Algorithm one: when s=0, the fusion algorithm is called Add algorithm, and the specific steps are as follows:
step 1: c (C) new ＝C 1 ∪C 2 ∪{c 30 }＝{c 3i |c 3i ∈C 1 ∪C 2 I=1, 2,..m+n, and c 30 E is a new set of concepts, where C 1 ＝{c 11 ，c 12 ，...，c 1m Sum C 2 ＝{c 21 ，c 22 ，...，c 2n Two sets of concepts corresponding to source ontology on domain E, c 30 For C in field E 1 And C 2 A parent concept of all concepts in (a);
step 2: constructing a new concept vector g from the new concept set new ＝(c 30 ，c 31 ，...，c 3，m+n ) T ；
Step 3: converting two bodies to be fused into two structural matrixes F in a natural traversal mode 1 And F 2 ；
Step 4: from F 1 *F 2 The operation can be obtained
e 1 ＝(1，0，...，0) 1×m ，e 2 ＝(1，0，...，0) 1×n ；
Step 5: from F new g new And
obtaining c 3i I=0, 1,2,..m+n.
Step 6: thus, a new body O after fusion can be constructed new 。
Algorithm II: when s > 0, the fusion algorithm is called Merge algorithm, and the specific steps are as follows:
step 1: c (C) 1 ＝{c 11 ，c 12 ，...，c 1m Sum C 2 ＝{c 21 ，c 22 ，...，c 2n Two concept sets corresponding to source ontology on domain E, C is based on one of the concept sets 1 And C 2 All concepts forming a new concept set are denoted as C new Only one is taken for the same concept;
step 2: constructing a new concept vector g from the new concept set new ＝(c 31 ，c 32 ，...，c 3，m+n-s ) T ；
Step 3: converting two bodies to be fused into two structural matrixes F in a natural traversal mode 1 And F 2 ；
Step 4: from F 1 *F 2 The operation can obtain F new ；
Step 5: from F new g new And
all concepts of ci were derived, i=0, 1,2,..m+n-s.
Step 6: thus, a new body O after fusion can be constructed new 。
Step four, ontology evolution management
The ontology evolution management needs to propagate and diffuse the confirmed ontology evolution operation, integrate the newly generated ontology model into an ontology knowledge base, and store the ontology knowledge base as data, wherein the file format is mainly XML and OWL.
The work is mainly to complete some operations after the evolution of the ontology, such as new creation, addition, deletion, modification and the like of the changed ontology. The management of ontology evolution is the alternation between different versions, and some versions may need to be traced or cancelled in the later stage, so that the modification process of the ontology needs to be managed. The change is transmitted through data call to confirm the change, and the operation of each step is recorded in a log form and stored in a database, so that the later check is convenient. And meanwhile, ontology evolution management also provides knowledge inquiry and storage, provides knowledge support for other application modules, assists a user to access the domain ontology in the ontology knowledge base and establishes a bridge for interaction between other application modules and the ontology knowledge base.
Specific introduction to related content:
1. ontology evolution definition
First, introducing definition of Ontology, ontology (ontologiy) is commonly understood as "object", which may be an object actually existing, or exist in virtual logic such as relationship, rule, etc., and is a method introduced for researching philosophy, which is used to describe an object existing in objective world, and is now used in computer research such as artificial intelligence, machine learning, etc. With the rise of Artificial Intelligence (AI), ontology theory is widely applied to the computer field, and although the ontology theory is mature, the definition of the ontology is controversial, and as the ontology research is further advanced, the concept is also perfected, wherein the main stages are shown in table 2:
TABLE 2 ontology definition
The ontology evolution is that the current ontology model cannot completely express all knowledge in a certain field, so that according to the evolution process performed by an external new knowledge source, the compatibility and logic consistency of the changed ontology and the adaptive application are considered in the whole evolution process, and any small change in the ontology model affects the whole system, and even affects related intelligent agents, services and applications. The ontology evolution is a series of continuous and perfect consistency propagation processes of concepts, relationships among concepts, attributes and the like in the ontology evolution mainly according to corresponding theories and methods, and the ontology evolution mainly relates to two aspects:
(1) The ontology concept is rich. The ontology construction is a step-by-step work, new concepts need to be continuously added to the initial ontology according to the external knowledge change, and the ontology concepts are enriched to meet the application requirements.
(2) Updating the ontology concept. According to the actual application requirements, not only the concept, the relationship and the like of the ontology are required to be added, but also the outdated concept, relationship and the like are sometimes required to be deleted. Modification of the ontology local changes the overall chain reaction, so that the modification of the ontology concept also needs to maintain the consistency of the data of each part of the ontology.
2. Ontology evolution reason and consistency analysis
There are various reasons for the evolution of ontologies, mainly including the following:
(1) Changes in the field. The change in domain is a very common occurrence. For example, when new dam-break emergency plan field ontology knowledge is added to the emergency plan field ontology, changes must be made to reflect such changes.
(2) Changes to the shared conceptual model. When the domain of the description changes, the meaning of some classes in the ontology will also represent different meanings due to semantic changes.
(3) Changes in representation. When the description language of the ontology is changed, the grammar structure, the expression mode and the like among different languages are different, so that the ontology semantics are difficult to be ensured to be consistent in the conversion process.
In addition, after the ontology evolves, the ontology non-uniformity detection is also needed to prevent system conflict caused by evolution, so that after additional transformation operation is implemented, the ontology structure, logic and user-defined uniformity are still maintained, and the analysis flow is shown in fig. 1.
3. Main body structure matrix in dam break emergency plan field
The ontology is defined by natural language description and mathematical language description, the former is mainly defined in the philosophy sense, and the latter is defined in the information science sense. The definition of the ontology is from objective description to deep processing, and the intelligent development requirement of a certain field is met. Here, according to the constituent elements of the established dam-break emergency plan field ontology model, the dam-break emergency plan field ontology model O is defined as a five-tuple:
O＝(C，R，F，A，I) (1)
wherein: c represents a concept set in the whole system of the dam-break emergency plan; r represents a finite set of relationships between dam-break emergency plan concepts; f represents a function set in a dam break emergency plan; a represents a finite set of axiom in a dam break emergency plan; i represents a specific entity set of the dam break emergency plan. In addition, in R, isA epsilon R, isA represents the classification relation among concepts, and forms a hierarchical structure among concepts.
Other related concepts are introduced to describe the relationship of the ontology and the matrix. In the ontology o= (C, R, F, a, I), if the topology of (C, R) is tree, the ontology o= (C, R, F, a, I) is referred to as tree ontology. If the topology of (C, R) is graph, it is referred to as graph ontology. In the tree body, a traversing mode of 'top to bottom and left to right' is adopted, and is called a natural traversing mode. The general ontology only discusses the relationship between the concept set C and the relationship set R, and only one tree-shaped ontology of the relationship IsA exists in R, namely R= { IsA }, so the expression form of the ontology in the dam-break emergency plan field can be recorded as O= (C, isA).
Let o= (C, isA) be one ontology in the field of dam break emergency plans, where c= { C 1 ，c 2 ，...，c n }. Assume that the traversal pattern in pair C is
Wherein i is 1 ，i 2 ，...，i n For an n-level arrangement, defined as:
then the matrix f= (F) kl ) n×n Traversing for ontology O
The underlying structural matrix, also called the body structural matrix, vector +.>
Is an ontology concept vector. As can be seen from the definition of the above body structure matrix, the structure matrix of the same body in different traversal modes is the same, and the body O and the body structure matrix F are in one-to-one correspondence in the same traversal mode.
4. Matrix fusion model for ontology in dam break emergency plan field
The body fusion is to fuse the source body with the existing new body to establish a new body, and simultaneously keep consistency. The current common fusion methods are the System Chimaera, PROMPT method developed by the university of Steady, the FCA-Merge method, which is a set of fusion and diagnostics. The mapping technology of the current ontology is developed rapidly, and an ontology matrix model is built by fusing the existing ontology with a new ontology. A specific algorithm model is shown in figure 2.
Assume that two bodies O are arranged in the dam break emergency plan field E 1 And O 2 The traversing mode is natural traversing, so that a body structure matrix F corresponding to two bodies can be obtained 1 And F 2 Obtaining a new matrix F through matrix fusion calculation new Then constructing a new body O after the two bodies are fused according to the new matrix new 。
5. Ontology matrix correlation operation
To implement the ontology matrix algorithm, it is necessary to introduce first a matrix straight sum and a definition of hadamard Ma Chengji. Let matrix a= (a) ij ) m×m Matrix b= (B ij ) n×n The straight sum C of the matrix is equal to taking A and B as diagonal lines, the off-diagonal line is 0, and is recorded as
Namely:
there are three main ways of multiplying the matrix, mainly used herein are Hadamard products, also called Hadamard Ma Chengji. Let matrix a= (a) ij ) m×n Matrix b= (B ij ) m×n The Hadamard product of the two matrices is c= (C) ij ) m×n Is marked as
Namely:
for two bodies O in dam break emergency plan field E 1 ＝(C1，IsA)、O 2 ＝(C 2 IsA), wherein C 1 ＝{c 11 ，c 12 ，...，c 1m Sum C 2 ＝{c 21 ，c 22 ，...，c 2n Two sets of concepts on domain E corresponding to source ontologies. c 11 And c 21 Respectively is O 1 With O 2 F, F 1 And F 2 Respectively is O 1 With O 2 Is a matrix of body structures. By O 1 For reference to the body, at C 1 And C 2 In which if there are s.gtoreq.0 concepts identical, then the ontology O is considered 1 The expression of knowledge is more reasonable and complete.
When s=0, it is represented that there is no identical concept in both ontologies. Definition F 1 And F 2 Mapping relationship to new structure matrix:
wherein:
e 1 ＝(1，0，...，0) 1×m ，e 2 ＝(1，0，...，0) 1×n 。
so when s=0,
when s > 0, define F 1 And F 2 Mapping relationship to new structure matrix:
F new ＝F 1 ·F 2 (6)
wherein F is new The construction of (a) can be accomplished in an iterative manner, so that
Obtaining a new matrix through iteration>
Finally, it can be seen that F new Is a uniquely determined matrix, the remaining diagonal elements are all 0, and only the first column is the only one zero vector, and the remaining columns are standard single bit vectors.
According to the method, when the same concept number s in the two bodies is more than or equal to 0, the following body matrix fusion operation is adopted and is marked as F 1 ＊F 2 The method comprises the following steps:
example 2 dam break emergency plan field ontology fusion
Get two bodies O on dam break emergent plan field E 1 experiment With O 2 experiment Wherein C 1，i 、C 2，j (i=1, 2, 10; j=1, 2, 8) represent concepts in the field body of the dam burst emergency plan, respectively, the patterning of the structure of which is shown in fig. 3 and 4, and only the relationship of isas is maintained.
Under natural traversal, the respective structural matrix is obtained as F 1 And F is equal to 2 。
In order to realize the fusion of two bodies, experiments are mainly carried out in three cases, namely, the cases when two identical concepts do not exist in the two cases, and only one identical concept and more than two identical concepts exist in the two cases.
Case one: when there is no identical concept in the two bodies to be fused, i.e. s=0, then it is necessary to find a concept C for it in the dam break emergency plan field E 0 As their top level concept.
At this time, according to the Add algorithm, it is known that:
as shown in FIG. 5, the two bodies are sub-classes of the new body, the knowledge, the concept and the like of the two bodies are perfectly reserved, and as the two bodies have no same concept, no incompatibility exists, the fusion of the bodies is the simplest form, and the meaning of the new body is more complete.
And a second case: when there is only one identical concept in the two ontologies to be fused, i.e. s=1, assume C 21 ＝C 11 。
At this time, according to the Merge algorithm, it is known that:
as shown in FIG. 6, in the new ontology, the root concepts of the two ontologies are combined into one concept, other concept relationships are inherited, the generated new ontology knowledge is more complete in expression, and the new ontology knowledge is matched with the structure of the source ontology, and has the same structural form.
Case three: when there are two or more of the bodies to be fused, s=3 is taken here, assuming C 21 ＝C 11 ，C 13 ＝C 23 ，C 25 ＝C 12 。
At this time, according to the Merge algorithm, it is known that:
as shown in fig. 7, in the new ontology, first, the root concepts of the two ontologies are combined into one concept, the number of concepts is one less, and the rest of the concepts continue to be kept as they are. Next, due to C 13 ＝C 23 And they all have a common upper concept, so they can be combined into one concept C 13 C is trimmed off 23 ，C 13 The concept relationships contained below are preserved and inherited. And due to C 25 ＝C 12 Comparing the layer depths of the two to obtain C 25 Greater than C 12 Thus selecting C 25 Upper concept C of (2) 22 As the upper concept after the two are combined, C is cut off 12 And C 25 Merging into one concept, C 12 The conceptual relationships contained below are inherited. The knowledge relationship presented by the newly generated ontology is' C 12 IsAC 22 IsAC 11 "C in the new ontology can be obtained according to the transmissibility of IsA 12 IsA C 11 ", original body O 1 experiment The relation of (C) 12 IsA C 11 By the method, the knowledge of the fused ontology expression accords with the structure in the original ontology, and the concept relationship is inherited.
Through the fusion experiment, the algorithm can well realize ontology fusion, the generated new ontology semantics are more complete, other conceptual relations are reserved in the new ontology, the respective attributes of the new ontology are well fused, and the new attribute covers the old attribute value, so that the ontology model which is more in line with the actual situation is obtained.
Example 3 evolution case analysis of an ontology model in the field of dam-break emergency plans
The reservoir dam burst can form oversized flood flooding, life and property safety of a downstream society is threatened, a virtual specific flood burst scene is set for a flood-roof dam burst emergency caused by upstream continuous storm of a certain earth-rock dam, evolution of a dam burst emergency plan field ontology model is further analyzed how to complete under complex real scenes, a system dynamic matching plan flow is realized, proper emergency adjustment is made, and instantaneity of an emergency plan is improved.
The scenario settings are as follows: the upstream water level of the reservoir continuously rises, a large amount of water flows impact the dam and overflow the dam until the dam body breaks and breaks the dam, flood water invades a flood discharging channel and a coastal building after the dam break, various levels of power plants and hydraulic facilities at the downstream side, villages at the coastal part of a river basin at the downstream side and the like are damaged and lost by personnel at different degrees, and preliminary statistical information of the damage after the disaster is shown in a table 3 under the conditions of fuzzy information and uncertainty in the evolution process of flood discharging.
TABLE 3 description of dam break scenario for certain earth-rock dam
| Grading index | Description of the situation |
| Damaged road | Good road traffic condition |
| Area of influenceAnd range of | Partial areas such as XX City, YY county and ZZ county |
| Infrastructure impairment | Urban road, power transmission (water) line, oil gas pipeline and enterprise |
| Casualties of people | 7 people |
| Direct economic loss | 0.37 hundred million yuan |
| Disaster-stricken population | 35 thousands of resident people on the coast behind and downstream of the dam |
| Natural and cultural landscape | Provincial and municipal level nature and cultural landscape |
| Animal and plant habitat | National second and third level protection animals and plants and living environment thereof |
| River channel shape | Serious damage to medium and small river |
| Town | Villages and towns |
| Predicted recovery deadlines | 90 days |
According to the simulation scene description table, the hazard condition is analyzed and judged, the risk degree is predicted, the maximum hazard result is selected, according to intelligent matching of the body knowledge in the field of emergency plans in the system, the system makes a three-level response instruction to be started, and decision-making staff performs orderly emergency treatment according to the system prediction result.
Along with the development of the event, some uncertainty factors can force the current emergency scheme to be not well matched with the change of the dynamic environment, so that the emergency guiding function cannot be realized, and the effect is greatly reduced. As can be seen from the core concept of the hidden field body of the dam-break emergency plan shown in fig. 11, the uncertainty that causes the deviation of the emergency plan is mainly divided into evacuation time, dam-break result, dam-break flood and emergency resources. Assuming that in the dam break emergency event, the uncertainty factor causing the deviation of the emergency scheme is the evacuation time, and the other items are not found to change, so that in order to be able to make corresponding adjustment of the dam break emergency scheme in response to the uncertainty of the evacuation time, a part of concepts are extracted from the hidden ontology model OWL format, only classes containing the evacuation time and relations thereof are extracted, and the IsA relations are reserved, so as to form an ontology structure under the initial condition, and the ontology structure is marked as O Initial initiation The structure is shown in fig. 8.
The analysis and the feedback of the data know that the movement capacity and presupposes great access in walking evacuation of disaster-stricken people are influenced, and the influence of uncertainty of crowd composition is also influenced, and the entity to be fused is marked as O at the moment To be treated The structure is shown in fig. 9.
According to the definition of the body structure matrix in the upper section, O under natural traversal can be obtained Initial initiation With O To be treated The matrix of the body structure is F 1 、F 2 ：
According to the ontology evolution operation rule, the number of concepts in the initial ontology structure which is the same as that of the ontology to be fused is more than two, so that the matrix of the fused ontology structure is F according to Merge algorithm new The following is shown:
the fused body structure O can be obtained by a Merge fusion algorithm new The structure is shown in fig. 10, wherein: "crowd composition" is an emerging concept; "movement capability" is a knowledge concept that has not changed, but whose class attribute values have changed.
In summary, in the new ontology, ontology O Initial initiation The time used for the root concept evacuation activity is taken as the root concept of a new ontology, the concepts of age, population density, sex, personnel psychology and the like which are positioned on the same layer and have the same upper concepts are combined, the new concept knowledge is formed by the extended crowd, the concept relationship is inherited, the ontology structure of more complete knowledge is finally obtained, the structural form is consistent with the previous one, and the evolution of the ontology model in the dam-break emergency plan field is realized.
Therefore, the body model in the dam-break emergency plan field is not unchanged, the body model can be modified by utilizing the body evolution method according to specific conditions, the influence of uncertainty factors in emergency is transmitted and analyzed, the body structure of the dam-break emergency plan is enabled to have directivity, directionality and accuracy, the body logic structure is modified through the body evolution, the change or degree change of constituent elements of the category is realized, different changes are finally embodied by the change of a plan instruction, the dynamic update of emergency decision under complex actual conditions can be realized by utilizing the method, and the capability of coping with the dam-break emergency uncertainty is improved.
Claims (7)
1. A matrix fusion algorithm-based dam break emergency plan field ontology evolution method is characterized by comprising the following steps:
step one, constructing a dam break emergency plan frame
Forming text of a dam break emergency plan based on ontology theory, constructing a basic vocabulary term table related to the dam break emergency plan and an ontology knowledge base among the basic vocabulary terms, defining rules in the ontology knowledge base by using ontology language, and establishing dam break stressThe body model of the emergency plan field uses a body editing tool Prot g to convert the hidden field body of the initial emergency plan into an OWL form, and records the OWL form as a body O 1 ；
Step two, uncertainty factor change capture
Collecting uncertainty factors causing the change of the emergency plan, and recording an implicit domain ontology caused by the uncertainty of the emergency plan to be changed as an ontology O 2 ；
Step three, body evolution operation
Two bodies O to be fused are fused by utilizing a body fusion algorithm 1 Body O 2 Respectively converting the basic vocabulary terms into corresponding structural matrixes, fusing a new body with a source body according to an evolution source of the collected uncertainty factors, continuously acquiring new attribute values, concepts and relations in the field, extracting the basic vocabulary terms and the relations among the basic vocabulary terms, calculating the structural matrixes of the fused new body through an Add matrix fusion algorithm or a Merge matrix fusion algorithm, converting the structural matrixes into body structural forms, enabling the structural matrixes to be fused into a dam break emergency plan recessive field ontology model, converting the structural matrixes into a dam break emergency plan dominant field ontology model, and realizing recessive field ontology evolution;
step four, ontology evolution management
And (3) propagating and diffusing the ontology changes, integrating the newly generated ontology model into an ontology knowledge base, and storing the ontology knowledge base in a data form.
2. The matrix fusion algorithm-based dam break emergency plan field ontology evolution method according to claim 1, wherein the uncertainty factors include concepts, relations and instance data.
3. The matrix fusion algorithm-based dam break emergency plan field ontology evolution method according to claim 1, wherein the ontology evolution management completes the alternation between the ontologies of different versions, the propagation and the diffusion of the ontology changes realize the confirmation of the changes through the data call, and each operation is recorded in a log form and stored in a database.
4. The matrix fusion algorithm-based dam break emergency plan field ontology evolution method according to claim 1, wherein the ontology evolution management provides the query and storage functions of an ontology knowledge base, provides support of the ontology knowledge base for other application modules, and assists a user in accessing a field ontology in the ontology knowledge base.
5. The method for the evolution of the ontology in the dam break emergency plan field based on the matrix fusion algorithm as claimed in claim 1, wherein when the concept number s=0, the ontology fusion algorithm adopts an Add algorithm, and specifically comprises the following steps:
step 1: c (C) new ＝C 1 ∪C 2 ∪{c 30 }＝{c 3i |c 3i ∈C 1 ∪C 2 I=1, 2,..m+n, and c 30 E is a new set of concepts, where C 1 ＝{c 11 ，c 12 ，...，c 1m Sum C 2 ＝{c 21 ，c 22 ，...，c 2n Two sets of concepts corresponding to source ontology on domain E, c 30 For C in field E 1 And C 2 A parent concept of all concepts in (a);
step 2: constructing a new concept vector g from the new concept set new ＝(c 30 ，c 31 ，...，c 3，m+n ) T ；
Step 3: converting two bodies to be fused into two structural matrixes F in a natural traversal mode 1 And F 2 ；
Step 4: from F 1 ·F 2 The operation is obtained
e 1 ＝(1，0，...，0) 1×m ，e 2 ＝(1，0，...，0) 1×n ；/>
Step 5: from F new g new And
obtaining c 3i I=0, 1,2,., m+n;
step 6: constructing a new body O after fusion new 。
6. The matrix fusion algorithm-based dam break emergency plan field ontology evolution method according to claim 1, wherein when the concept number s is more than 0, the ontology fusion algorithm adopts a Merge algorithm, and specifically comprises the following steps:
step 1: c (C) 1 ＝{c 11 ，c 12 ，...，c 1m Sum C 2 ＝{c 21 ，c 22 ，...，c 2n Two concept sets corresponding to source ontology on domain E, C is based on one of the concept sets 1 And C 2 All concepts forming a new concept set are denoted as C new Only one is taken for the same concept;
step 2: constructing a new concept vector g from the new concept set new ＝(c 31 ，c 32 ，...，c 3,m+n-s ) T ；
Step 3: converting two bodies to be fused into two structural matrixes F in a natural traversal mode 1 And F 2 ；
Step 4: from F 1 ＊F 2 Calculation to obtain F new ；
Step 5: from F new g new And
deriving all concepts of ci, i=0, 1,2,..m+n-s;
step 6: constructing a new body O after fusion new 。
7. The matrix fusion algorithm-based dam break emergency plan field ontology evolution method according to claim 1, wherein the ontology evolution operation further comprises the step of ontology non-uniformity detection after completion, wherein the ontology non-uniformity detection is used for ensuring the self-defined consistency of ontology structures, logics and users.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210478625.3A CN115345411B (en) | 2022-04-29 | 2022-04-29 | Matrix fusion algorithm-based dam break emergency plan field ontology evolution method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210478625.3A CN115345411B (en) | 2022-04-29 | 2022-04-29 | Matrix fusion algorithm-based dam break emergency plan field ontology evolution method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115345411A CN115345411A (en) | 2022-11-15 |
| CN115345411B true CN115345411B (en) | 2023-05-05 |
Family
ID=83948796
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210478625.3A Active CN115345411B (en) | 2022-04-29 | 2022-04-29 | Matrix fusion algorithm-based dam break emergency plan field ontology evolution method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115345411B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112836018A (en) * | 2021-02-07 | 2021-05-25 | 北京联创众升科技有限公司 | Method and device for processing emergency plan |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100505970B1 (en) * | 2003-11-28 | 2005-08-30 | 한국전자통신연구원 | System and method for knowledge modeling using ontology |
| FR2865055A1 (en) * | 2004-01-12 | 2005-07-15 | Thomson Licensing Sa | Knowledge base for tennis, has interrogative knowledge base with lexical unit database having key words associated to classes of interrogation syntax structures of request in natural language and to interrogation object concepts and roles |
| CN111914100A (en) * | 2020-08-11 | 2020-11-10 | 中科院合肥技术创新工程院 | Emergency decision knowledge representation method based on ontology |
| CN111967800A (en) * | 2020-09-08 | 2020-11-20 | 四川华能宝兴河水电有限责任公司 | Hydropower station dam break disaster event scene construction and emergency evaluation analysis system |
| CN112328810B (en) * | 2020-11-11 | 2022-10-14 | 河海大学 | Knowledge graph fusion method based on self-adaptive mixed ontology mapping |
| CN113344457A (en) * | 2021-07-05 | 2021-09-03 | 四川华能宝兴河水电有限责任公司 | Hydropower station dam break disaster event simulation evaluation analysis system |
-
2022
- 2022-04-29 CN CN202210478625.3A patent/CN115345411B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112836018A (en) * | 2021-02-07 | 2021-05-25 | 北京联创众升科技有限公司 | Method and device for processing emergency plan |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115345411A (en) | 2022-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109992672A (en) | Knowledge mapping construction method based on disaster scene | |
| Xie et al. | Rule acquisition and optimal scale selection in multi-scale formal decision contexts and their applications to smart city | |
| Haimes | Risk modeling of interdependent complex systems of systems: Theory and practice | |
| CN111798355B (en) | Urban public security event chain formalized organization method and event chain database | |
| Xu et al. | Large group emergency decision-making method with linguistic risk appetites based on criteria mining | |
| CN107331132A (en) | A kind of method and system of Urban Fires hidden danger dynamic prediction monitoring | |
| CN113434634A (en) | Knowledge graph construction method and device | |
| CN112559766A (en) | Legal knowledge map construction system | |
| Ma et al. | City profile: using smart data to create digital urban spaces | |
| CN113468320B (en) | Method and system for quickly visualizing earthquake emergency information | |
| CN114117065A (en) | Knowledge graph construction method and system based on power production statistical service | |
| Zhong et al. | Pipeline risk big data intelligent decision-making system based on machine learning and situation awareness | |
| CN115345411B (en) | Matrix fusion algorithm-based dam break emergency plan field ontology evolution method | |
| CN117235219A (en) | Reservoir knowledge intelligent question-answering system based on flood prevention demands | |
| CN107274324A (en) | A kind of method that accident risk assessment is carried out based on cloud service | |
| Linghu et al. | A conceptual model for flood disaster risk assessment based on agent-based modeling | |
| CN116304070A (en) | Automatic generation method of power grid emergency treatment scheme based on knowledge graph construction | |
| CN108829786A (en) | A kind of geographical big data operation method of the society of time-space-logical relation | |
| Liu et al. | Modelling a large scale system for risk assessment | |
| Ma et al. | Strategic analysis of the James Bay hydro-electric dispute in Canada | |
| Goodall et al. | Exploring disaster ontologies from Chinese and Western perspectives: commonalities and nuances | |
| Fan et al. | Effect evaluation of eco-environmental big data resource integration and data sharing construction | |
| Chen et al. | Research on the construction of intelligent community emergency service platform based on convolutional neural network | |
| Memiş et al. | An Application of Soft Decision-Making Methods to Energy Planning of Turkey | |
| De la Vega Hernández et al. | Smart cities and urban dimensions to ensure sustainable development: a bibliometric analysis |
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 |