CN113298426B - Knowledge graph driven dam safety evaluation weight dynamic drafting method and system - Google Patents
Knowledge graph driven dam safety evaluation weight dynamic drafting method and system Download PDFInfo
- Publication number
- CN113298426B CN113298426B CN202110672706.2A CN202110672706A CN113298426B CN 113298426 B CN113298426 B CN 113298426B CN 202110672706 A CN202110672706 A CN 202110672706A CN 113298426 B CN113298426 B CN 113298426B
- Authority
- CN
- China
- Prior art keywords
- dam
- knowledge
- graph
- path
- entity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000011156 evaluation Methods 0.000 title claims description 20
- 239000013598 vector Substances 0.000 claims abstract description 32
- 230000003993 interaction Effects 0.000 claims abstract description 22
- 230000008447 perception Effects 0.000 claims abstract description 17
- 238000011176 pooling Methods 0.000 claims abstract description 13
- 238000013507 mapping Methods 0.000 claims abstract description 9
- 230000006870 function Effects 0.000 claims description 9
- 230000002452 interceptive effect Effects 0.000 claims description 6
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 238000011144 upstream manufacturing Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 description 11
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 230000001276 controlling effect Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000035882 stress Effects 0.000 description 1
- 230000002195 synergetic effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Biomedical Technology (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 knowledge graph driven dynamic setting method and a system for dam safety judgment weight.A dam knowledge perception path is generated by combining semantics of entities and relations in a dam safe operation knowledge graph; then mapping each entity, entity corresponding type and relationship in the knowledge graph to a low-dimensional vector for representation by using a graph volume network; sequentially encoding the elements by adopting an LSTM (localized surface texture TM), and capturing the combined semantics of the entities with the relation as a condition; and finally, combining a plurality of paths by using the pooling layers, and outputting a final score of interaction between a given user and the project, namely a dam measuring point weight value. The method is used for evaluating the influence degree of each measuring point of the dam on the safety of the dam, assists comprehensive judgment of the safety of the dam, and has real-time performance and reusability.
Description
Technical Field
The invention relates to a dynamic setting method and a dynamic setting system for dam safety comprehensive evaluation weights, in particular to a dynamic setting method and a dynamic setting system for dam safety comprehensive evaluation weights under a special working condition driven by a knowledge map, and belongs to the technical field of dam safety comprehensive evaluation.
Background
The dam is an important engineering measure for regulating and controlling the spatial and temporal distribution of water resources and optimizing the allocation of the water resources, but the potential safety problem of the dam is widely concerned by the problems of natural disasters or self aging along with the lapse of time. The dam grading comprehensive evaluation is one of important means for evaluating dam safety, and the dam safety is evaluated from local to overall and from small to large, namely, the grading comprehensive evaluation is carried out from a single measuring point, an instrument type, a monitoring project, a basic part and finally to a building. Because the importance degree of each measuring point is different, the weight value of each measuring point is evaluated through expert experience and site conditions in the past, the dam can often meet natural disasters such as earthquake, rainstorm, debris flow and the like, and the expert can not be called in time every time to obtain the weight value of each measuring point evaluated by the dam.
Disclosure of Invention
The invention aims to: aiming at the problems and the defects in the prior art, the invention provides a dynamic method and a system for comprehensive dam safety evaluation weight, in particular to a dynamic method and a system for comprehensive dam safety evaluation weight under a special working condition driven by a knowledge map.
The technical scheme is as follows: a knowledge graph driven dam safety judgment weight dynamic setting method comprises the following steps:
(1) combining the semantics of the entities and the relations in the dam safe operation knowledge graph to generate a dam safe operation knowledge perception path;
(2) mapping each entity, entity corresponding type and relation in the knowledge graph to a low-dimensional vector representation by using a graph convolution network;
(3) sequentially encoding the elements by adopting an LSTM (least squares TM), and capturing the combined semantics of the entities with the relation as a condition;
(4) and combining multiple paths by using the pooling layer, and outputting a final score of the given user interacting with the project, namely a dam measuring point weight value.
A knowledge graph driven dynamic setting method for dam safety evaluation weights is suitable for a dynamic setting method for comprehensive dam safety evaluation weights under special working conditions, wherein the special working conditions refer to the following steps:
compared with the daily working condition, the working state of the dam under natural disasters has the advantages that the generation of special working conditions is sudden, and the influence on the integrity and stability of the dam is larger. The natural disasters include earthquakes, rainstorms and debris flows.
Further, the concrete steps of combining the semantics of the entities and the relations in the dam safe operation knowledge graph in the step (1) to generate the dam knowledge perception path are as follows:
(1.1) defining the safe operation knowledge graph of the dam as
Wherein each triplet (h, r, t) represents a fact that there is a relationship r from the head entity h to the tail entity t; user item interaction data is typically represented as a bipartite graph, employing
And
respectively representing a user set and an item set, and the interaction between the user and the item is represented by a triple tau (u, interactive, i), wherein u istThe user refers to dam working conditions (including special working conditions such as earthquake, heavy rain and debris flow), itThe project comprises all measuring points of the dam, M and N respectively indicate the number of users and the number of the project, and the interact represents the interaction between u and i (namely, a relation exists between the u and the i);
(1.2) triplets in the knowledge-graph clearly describe the direct or indirect relational attributes of an item, which should constitute one or more paths between a given user and an item pair, namely the dam knowledge-aware path, defined as
Where e is an entity, r is a relationship, e1＝u，eL＝i，(el,rl,el+1) Is the ith triplet in p, l represents three in the pathThe number of tuples.
Further, the specific step of mapping each entity, entity corresponding type and relationship in the knowledge graph to a low-dimensional vector representation by using the graph convolution network in the step (2) is as follows:
given a path pkProjecting the type (such as special working condition, hydraulic inspection position, dam safety monitoring instrument) and specific value (such as dam measuring point (such as tension line EP 2-12) of each entity to two independent low-dimensional embedded vectors by using a graph convolution network
And
in the method, the semantics of the relationship are integrated into path representation learning, wherein d is the size of an embedded vector, a bracing wire instrument is used for measuring the horizontal displacement of the dam in the upstream and downstream directions and the left and right bank directions, and EP2-12 belongs to a measuring point monitored by using the bracing wire;
further, the step (3) of sequentially encoding the elements by using LSTM, and the specific step of capturing the combined semantics of the relationship-conditioned entities is as follows:
(3.1) concatenating the current entity e when the path step is l-1l-1、e'l-1And relation rl-1Generates a vector xl-1(i.e. the
Wherein
Is operated in series, e'l-1Is a low-dimensional embedded vector of a particular value, rl-1Is a relationship vector) and then x is addedl-1As input vector of LSTM, output hidden state vector hl-1And using the final state hlRepresenting the entire path pk；
(3.2) establishing a Path pkAfter vector representation, the final state is input into two fully-connected layers to obtain a path pkPrediction of (2)Fraction, the formula is shown below, where w1And w2Coefficient weights for the first and second layers, respectively, ReLU is an activation function, and τ is an interactive representation between the user and the item.
Further, the step (4) of merging multiple paths by using a pooling layer and outputting a final score of a given user interacting with a project, that is, a dam measuring point weight value, specifically comprises the following steps:
different paths in the dam knowledge perception path have different contributions to the preference of the model user, so that the pooling layer needs to perform weighted pool operation on the total scores of all paths and output the final score of interaction between a given user and a target item, the formula is as follows, firstly, the scores of all paths are aggregated through one weighted pool operation, and then the scores are converted into [0,1 ] by using a sigmoid function]Within the interval, where skIs the predicted score of the kth path, gamma is the hyperparameter controlling each exponential weight, sigma is the sigmoid function,
is the final station weight.
According to the dynamic setting method for the dam safety comprehensive evaluation weight, effective reasoning is carried out by utilizing the sequential dependence in the dam safety perception path, the weight of each measuring point of the dam is dynamically evaluated, and the real-time performance and the reusability of setting the dam weight are guaranteed while the labor cost is saved.
A knowledge-graph-driven dam safety judgment weight dynamic drafting system comprises:
a module for generating a knowledge perception path for safe operation of the dam: combining the semantics of the entities and the relations in the dam safe operation knowledge graph to generate a dam safe operation knowledge perception path;
a vector representation module: mapping each entity, entity corresponding type and relation in the knowledge graph to a vector representation by using a graph convolution network;
LSTM module: sequentially encoding the elements by adopting an LSTM (least squares TM), and capturing the combined semantics of the entities with the relation as a condition;
a weight acquisition module: and combining multiple paths by using the pooling layer, and outputting a final score of the given user interacting with the project, namely a dam measuring point weight value.
Has the advantages that: compared with the prior art, the knowledge graph driven dynamic dam safety evaluation weight drafting method and the knowledge graph driven dynamic dam safety evaluation weight drafting system have the following advantages: the dam knowledge perception path is generated by combining the semantics of entities and relations in the dam safe operation knowledge graph, and the weight of each measuring point of the dam is dynamically evaluated by using a weight drafting model formed by connecting a graph convolution network, an LSTM and a pooling layer in series, so that the labor cost is saved, and the real-time performance and reusability of dam weight drafting are ensured.
Drawings
FIG. 1 is an example of a dam operational safety knowledge graph seismic module;
FIG. 2 is a visualization case of three predicted scoring paths under dam earthquake conditions;
FIG. 3 is a flow chart of an embodiment of the present invention.
Detailed Description
The present invention is further illustrated by the following examples, which are intended to be purely exemplary and are not intended to limit the scope of the invention, as various equivalent modifications of the invention will occur to those skilled in the art upon reading the present disclosure and fall within the scope of the appended claims.
Fig. 1 shows an example of a safety knowledge graph seismic module for dam operation, which has two seismic events, namely yuxi Tonghai and Yunnan Yongde, and makes the production area of a bay slightly shocky, so that the bay power plant carries out comprehensive special inspection. The key inspection objects are key monitoring instruments in a dam safety automatic system, such as a tension line, a vertical line, a stress meter and the like, wherein each monitoring instrument is provided with a plurality of monitoring measuring points; there are important parts in the bay hydraulic structure such as various elevation galleries, first-stage factory buildings, traffic holes, etc. The more times of investigation of monitoring instruments or key parts where the measuring points are located, the higher the importance degree of the measuring points is, and meanwhile, the influence of abnormal conditions of the measuring points on the safety of the dam can be searched through backtracking of the knowledge graph.
FIG. 2 is a visual example of three predicted score paths under dam seismic conditions, showing S1And S2The paths all provide relatively high scores, 0.372 and 0.376, respectively. When earthquake occurs, firstly, safety monitoring data is provided, whether the measuring point is abnormal or not is judged through the safety value range of the measuring point, historical data and the current day measuring value, and the method has real-time performance and reliability, so that S1The path score is highest for the actual situation. The key parts of the dam are inspected afterwards, the abnormal conditions can be visually judged, the accuracy is high only under the condition that the abnormal conditions are obvious due to the judgment of naked eyes, and otherwise, certain errors exist, so that S2Path score lower than S1The path is in accordance with the actual situation. The collaborative filtering plays an important role in judging the importance degree of the measuring point EP2-12, and the measuring point still belongs to an important investigation measuring point under other working conditions, which indicates that the measuring point has higher weight than a common measuring point.
As shown in fig. 3, the method for dynamically formulating the safety evaluation weight of the dam driven by the knowledge-map comprises the following steps:
(1) combining the semantics of entities and relations in the dam safe operation knowledge graph to generate a dam knowledge perception path; the method specifically includes the following steps.
(1.1) defining the safe operation knowledge graph of the dam as
Wherein each triplet (h, r, t) represents the fact that there is a relationship r from the head entity h to the tail entity t, as in fig. 2 (hydraulic building, detection site, 961 corridor (tension line)) is a triplet, "hydraulic building" is the head entity h, "detection site" is the relationship r, "961 corridor (tension line))" is the tail entity t; user item interaction data is typically represented as a bipartite graph, employing
And
respectively representing a user set and an item set, and the interaction between the user and the item is represented by a triple tau (u, interactive, i), wherein u istThe user refers to dam working conditions (including special working conditions such as earthquake, heavy rain and debris flow), itThe project comprises all measuring points of the dam, M and N respectively indicate the number of users and the number of projects, and the interact represents the interaction between u and i (namely the relationship exists between the two).
(1.2) triplets in the knowledge-graph clearly describe the direct or indirect relational attributes of an item, which should constitute one or more paths between a given user and an item pair, namely the dam knowledge-aware path, defined as
Wherein e1＝u，eL＝i，(el,rl,el+1) Is the ith triplet in p, l represents the number of triplets in the path; as FIG. 2 shows, these paths from the same user "earthquake" to the same project "EP 2-12" clearly express their different multi-step relationships and imply different combination semantics and possible interpretations of the "earthquake" regime. The two paths of the 'hydraulic structure' and the 'dam safety monitoring system' show that the data of the monitoring measuring point EP2-12 is one of important bases for safety judgment during earthquake; while the "rainstorm" path reflects a synergistic filtering effect, similar users tend to have similar preferences. Wherein the 961 corridor is a part of a dam, and the arranged monitoring instrument comprises a tension line EP2-12 measuring point; the tension instrument is used for measuring the horizontal displacement of the dam in the upstream and downstream directions and the left and right bank directions; EP2-12 belongs to a station which is monitored by means of a tension wire.
(2) Mapping each entity, entity corresponding type and relation in the knowledge graph to a low-dimensional vector representation by using a graph convolution network, and specifically comprising the following steps:
given a path pkProjecting the type (such as special working conditions, hydraulic inspection positions, dam safety monitoring instruments and the like) and specific value (namely a dam measuring point, such as a tension line EP2-12 measuring point in the figure 2) of each entity to two independent low-dimensional embedded vectors by using a graph convolution network
And
in the method, the semantics of the relationship are integrated into the path representation learning, wherein d is the size of an embedded vector, a bracing wire instrument is used for measuring the horizontal displacement of the dam in the upstream and downstream directions and the left and right bank directions, and EP2-12 belongs to a measuring point monitored by using the bracing wire.
(3) Sequentially encoding the elements by adopting an LSTM (least squares TM), and capturing the combined semantics of the entities with the relation as a condition; the method comprises the following steps:
(3.1) concatenating the current entity e when the path step is l-1l-1、e'l-1And relation rl-1Generates a vector xl-1(i.e. the
Wherein
Is a series operation) and then x is addedl-1As input vector of LSTM, output hidden state vector hl-1And using the final state hlRepresenting the entire path pk；
(3.2) establishing a Path pkAfter vector representation, the final state is input into two fully-connected layers to obtain a path pkThe formula is shown below, wherein w1And w2Coefficient weights for the first and second layers, respectively, ReLU is an activation function, and τ is an interactive representation between the user and the item.
(4) Combining a plurality of paths by using the pooling layer, and outputting a final score of interaction between a given user and a project, namely a dam measuring point weight value, wherein the specific steps are as follows:
different paths in the dam knowledge perception path have different contributions to the preference of the model user, so that the pooling layer needs to perform weighted pool operation on the total scores of all paths and output the final score of interaction between a given user and a target item, the formula is as follows, firstly, the scores of all paths are aggregated through one weighted pool operation, and then the scores are converted into [0,1 ] by using a sigmoid function]Within a range of skIs the predicted score of the kth path, gamma is the hyperparameter controlling each exponential weight, sigma is the sigmoid function,
is the final station weight.
(5) The method is characterized in that the weight of each measuring point of the dam is applied to dam safety judgment, and the method comprises the following specific steps:
because the importance degree of each measuring point is different, the weight of each measuring point of the dam can be evaluated by utilizing the steps (1) to (4), then the analytic hierarchy process is carried out, a weight table required by judgment is generated, and then the comprehensive analysis is carried out on the building by combining the fuzzy theory and the membership matrix, so that the safety state of each part of the building and the building can be judged.
A knowledge-graph-driven dam safety judgment weight dynamic drafting system comprises:
a module for generating a knowledge perception path for safe operation of the dam: combining the semantics of the entities and the relations in the dam safe operation knowledge graph to generate a dam safe operation knowledge perception path;
a vector representation module: mapping each entity, entity corresponding type and relation in the knowledge graph to a vector representation by using a graph convolution network;
an LSTM module: sequentially encoding the elements by adopting an LSTM (least squares TM), and capturing the combined semantics of the entities with the relation as a condition;
a weight acquisition module: and combining multiple paths by using the pooling layer, and outputting a final score of the given user interacting with the project, namely a dam measuring point weight value.
According to the embodiment, the weight of each measuring point of the dam is evaluated dynamically, and reliable basis is provided for the comprehensive evaluation of the dam safety in real time; meanwhile, the method realizes high multiplexing and can be applied to other dams.
Claims (6)
1. A knowledge graph driven dam safety evaluation weight dynamic drafting method is characterized by comprising the following steps:
(1) combining semantics of entities and relations in the dam safe operation knowledge graph to generate a dam safe operation knowledge perception path;
(2) mapping each entity, entity corresponding type and relation in the knowledge graph to a vector representation by using a graph convolution network;
(3) sequentially encoding the elements by adopting an LSTM (least squares TM), and capturing the combined semantics of the entities with the relation as a condition;
(4) combining a plurality of paths by using the pooling layer, and outputting a final score of interaction between a given user and a project, namely a dam measuring point weight value;
the concrete steps of combining the semantics of the entities and the relations in the dam safe operation knowledge graph to generate the dam knowledge perception path in the step (1) are as follows:
(1.1) defining the safe operation knowledge graph of the dam as
Wherein each triplet (h, r, t) represents a factNamely, a relation r exists from the head entity h to the tail entity t; user item interaction data is represented as a bipartite graph, employing
And
respectively representing a user set and an item set, and the interaction between the user and the item is represented by a triple tau (u, interactive, i), wherein u istThe user refers to the dam condition, itThe project comprises all measuring points of the dam, M and N respectively indicate the number of users and the number of the project, and the interact represents the interaction between u and i, namely the relationship exists between the u and the i;
(1.2) triplets in the knowledge-graph clearly describe the direct or indirect relational attributes of an item, which should constitute one or more paths between a given user and an item pair, namely the dam knowledge-aware path, defined as
Wherein e1＝u，eL＝i，(el,rl,el+1) Is the ith triplet in p, l represents the number of triplets in the path.
2. The knowledge-graph-driven dynamic determination method for dam safety evaluation weights according to claim 1, wherein the method is a dynamic determination method for dam safety comprehensive evaluation weights under special conditions, and the special conditions refer to: the dam works in natural disasters.
3. The method for dynamically formulating the safety evaluation weights of the dams driven by the knowledge-graph according to the claim 1, wherein the specific steps of mapping each entity, the corresponding type and the relation of the entity in the knowledge-graph to a vector representation by using the graph convolution network in the step (2) are as follows:
given a path pkEach is formed by a graph convolution networkProjecting the type and specific value of an entity onto two independent embedded vectors
And
and (3) integrating the semantics of the relationship into path representation learning, wherein d is the size of an embedded vector, a bracing wire instrument is used for measuring the horizontal displacement of the dam in the upstream and downstream directions and the left and right bank directions, and EP2-12 belongs to a measuring point monitored by using a bracing wire.
4. The method for dynamically formulating knowledge-graph-driven dam security assessment weights according to claim 1, wherein the step (3) of sequentially encoding elements by using LSTM and capturing the combined semantics of the relation-conditioned entities comprises the following steps:
(3.1) concatenating the current entity e when the path step is l-1l-1、e'l-1And relation rl-1Generates a vector xl-1Then x is addedl-1As input vector of LSTM, output hidden state vector hl-1And using the final state hlRepresenting the entire path pk；
(3.2) establishing a Path pkAfter vector representation, the final state is input into two fully-connected layers to obtain a path pkThe formula is shown below, wherein w1And w2Coefficient weights for the first and second layers, respectively, ReLU being an activation function, τ being a representation of the interaction between the user and the item
s(τ|pk)＝W2 TReLU(W1 Tpk)。
5. The method for dynamically formulating knowledge-graph-driven dam safety assessment weights according to claim 1, wherein in the step (4), a plurality of paths are merged by using a pooling layer, and a final score of interaction between a given user and a project is output, namely a dam measuring point weight value is obtained by the following specific steps:
different paths in the dam knowledge perception path have different contributions to the preference of the model user, so that the pooling layer needs to perform weighted pool operation on the total scores of all paths and output the final score of interaction between a given user and a target item, the formula is as follows, firstly, the scores of all paths are aggregated through one weighted pool operation, and then the scores are converted into [0,1 ] by using a sigmoid function]Within the interval, where skIs the predicted score of the kth path, gamma is the hyperparameter controlling each exponential weight, sigma is the sigmoid function,
is the weight of the final measured point,
6. a knowledge-graph-driven dam safety judgment weight dynamic drafting system is characterized by comprising:
a module for generating a knowledge perception path for safe operation of the dam: combining the semantics of the entities and the relations in the dam safe operation knowledge graph to generate a dam safe operation knowledge perception path;
the vector representation module: mapping each entity, entity corresponding type and relation in the knowledge graph to a vector representation by using a graph convolution network;
LSTM module: sequentially encoding the elements by adopting an LSTM (least squares TM), and capturing the combined semantics of the entities with the relation as a condition;
a weight acquisition module: combining a plurality of paths by using the pooling layer, and outputting a final score of interaction between a given user and a project, namely a dam measuring point weight value;
the specific steps of generating the dam safe operation knowledge sensing path by combining the semantics of the entity and the relation in the dam safe operation knowledge graph in the module for generating the dam safe operation knowledge sensing path are as follows:
(1.1) defining the safe operation knowledge graph of the dam as
Wherein each triplet (h, r, t) represents a fact that there is a relationship r from the head entity h to the tail entity t; user item interaction data is represented as a bipartite graph, employing
And
respectively representing a user set and an item set, and the interaction between the user and the item is represented by a triple tau (u, interactive, i), wherein u istThe user refers to the dam condition, itThe project comprises all measuring points of the dam, M and N respectively indicate the number of users and the number of the project, and the interwork represents the interaction between u and i, namely the interaction between the u and the i exists;
(1.2) triplets in the knowledge-graph clearly describe the direct or indirect relational attributes of an item, which should constitute one or more paths between a given user and an item pair, namely the dam knowledge-aware path, defined as
Wherein e1＝u，eL＝i，(el,rl,el+1) Is the ith triplet in p, and l represents the number of triplets in the path.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110672706.2A CN113298426B (en) | 2021-06-17 | 2021-06-17 | Knowledge graph driven dam safety evaluation weight dynamic drafting method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110672706.2A CN113298426B (en) | 2021-06-17 | 2021-06-17 | Knowledge graph driven dam safety evaluation weight dynamic drafting method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113298426A CN113298426A (en) | 2021-08-24 |
| CN113298426B true CN113298426B (en) | 2022-06-24 |
Family
ID=77328602
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110672706.2A Active CN113298426B (en) | 2021-06-17 | 2021-06-17 | Knowledge graph driven dam safety evaluation weight dynamic drafting method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113298426B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115982374B (en) * | 2022-12-02 | 2023-07-04 | 河海大学 | Multi-view learning entity alignment method and system for dam emergency response knowledge base linkage |
| CN116861785B (en) * | 2023-07-06 | 2024-08-27 | 河海大学 | Knowledge-graph-based concrete dam key part partitioning method and device |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205329A (en) * | 2015-09-29 | 2015-12-30 | 南京南瑞集团公司 | Comprehensive evaluation method for dam safety |
| CN105976104A (en) * | 2016-05-03 | 2016-09-28 | 中国电建集团昆明勘测设计研究院有限公司 | Dam safety evaluation method based on monitoring data |
| CN106909622A (en) * | 2017-01-20 | 2017-06-30 | 中国科学院计算技术研究所 | Knowledge mapping vector representation method, knowledge mapping relation inference method and system |
| CN107480341B (en) * | 2017-07-21 | 2018-10-23 | 河海大学 | A kind of dam safety comprehensive method based on deep learning |
| CN109657074B (en) * | 2018-09-28 | 2023-11-10 | 北京信息科技大学 | News knowledge graph construction method based on address tree |
| CN110188399B (en) * | 2019-05-07 | 2020-10-16 | 河海大学 | Dam safety monitoring single-measuring-point evaluation method based on multiple correlation sequences |
| CN110334221B (en) * | 2019-07-18 | 2022-12-09 | 桂林电子科技大学 | Interpretability recommendation method based on knowledge graph path |
| CN110516160B (en) * | 2019-08-30 | 2022-04-01 | 中国科学院自动化研究所 | Knowledge graph-based user modeling method and sequence recommendation method |
| CN111241295B (en) * | 2020-01-03 | 2022-05-03 | 浙江大学 | Knowledge graph relation data extraction method based on semantic syntax interactive network |
| CN111832922B (en) * | 2020-06-30 | 2021-10-01 | 北方工业大学 | Food safety event risk studying and judging method and device based on knowledge graph reasoning |
| CN112115358B (en) * | 2020-09-14 | 2024-04-16 | 中国船舶重工集团公司第七0九研究所 | Personalized recommendation method utilizing multi-hop path characteristics in knowledge graph |
| CN112287608B (en) * | 2020-11-19 | 2023-11-24 | 中国水利水电科学研究院 | Dam safety prediction system and method based on fuzzy comprehensive evaluation |
| CN112540832B (en) * | 2020-12-24 | 2022-01-28 | 中山大学 | Cloud native system fault analysis method based on knowledge graph |
| CN112765322B (en) * | 2021-01-25 | 2023-05-23 | 河海大学 | Remote sensing image searching recommendation method based on water conservancy domain knowledge graph |
| CN112836120B (en) * | 2021-01-27 | 2024-03-22 | 深圳大学 | Movie recommendation method, system and terminal based on multi-mode knowledge graph |
-
2021
- 2021-06-17 CN CN202110672706.2A patent/CN113298426B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113298426A (en) | 2021-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102866321B (en) | Self-adaptive stealing-leakage prevention diagnosis method | |
| CN113298426B (en) | Knowledge graph driven dam safety evaluation weight dynamic drafting method and system | |
| Wang et al. | Testing and modelling autoregressive conditional heteroskedasticity of streamflow processes | |
| Pyayt et al. | Machine learning methods for environmental monitoring and flood protection | |
| CN103292762B (en) | Displacement monitoring method for judging dam stability | |
| CN108764601A (en) | A kind of monitoring structural health conditions abnormal data diagnostic method based on computer vision and depth learning technology | |
| CN105573997A (en) | Method and device for determining electric larceny suspect user | |
| CN111861211B (en) | System with double-layer anti-electricity-stealing model | |
| CN102128022A (en) | Drilling engineering early warning method and system thereof | |
| CN110348150A (en) | A kind of fault detection method based on dependent probability model | |
| CN104535346B (en) | Anti-slip stability detection method for concrete gravity dangerous dam | |
| CN112101811A (en) | Water supply network pipe explosion risk prediction method and system | |
| CN112541666B (en) | Shield tunnel risk assessment method considering uncertainty of earthquake vulnerability model | |
| Turrisi et al. | A cointegration-based approach for automatic anomalies detection in large-scale structures | |
| CN115187033A (en) | Intelligent construction site construction safety risk early warning system and method thereof | |
| CN115906663A (en) | Building safety evaluation model establishing method, evaluation method, server and system | |
| CN117392815A (en) | Alarm method and system for secondary water supply based on Internet of things | |
| CN115977736A (en) | Coal and gas outburst early warning method based on field real-time data drive | |
| CN117113157B (en) | Platform district power consumption fault detection system based on artificial intelligence | |
| CN109902265A (en) | A kind of underground method for early warning based on hidden Markov model | |
| CN113221440A (en) | Optimal arrangement and real-time global inversion method for monitoring points of drainage system | |
| CN101382474A (en) | Multi-purpose evaluating method for bridge structure safety | |
| CN108108839A (en) | Grid information system equipment state method for early warning based on reverse Fuzzy Level Analytic Approach | |
| CN107220921B (en) | Verification method for data collected by energy consumption online monitoring system | |
| Ahmadi-Nedushan | Multivariate statistical analysis of monitoring data for concrete dams |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Zhou Hua Inventor after: Liao Guineng Inventor after: Tan Bin Inventor after: Xiong Xiaozhong Inventor after: Zhang Peng Inventor after: Peng Xinxin Inventor after: Yu Yi Inventor after: Wu Guangyao Inventor after: Wang Shunbo Inventor after: Zhai Li Inventor after: Nie Bingbing Inventor after: Mao Yingchi Inventor after: Zhao Huan Inventor after: Chen Hao Inventor after: Wang Qiang Inventor after: Guo Youan Inventor after: Wang Longbao Inventor after: Word Chen Bo Inventor after: Chen Weidong Inventor after: Li Hongbo Inventor before: Zhou Hua Inventor before: Liao Guineng Inventor before: Tan Bin Inventor before: Xiong Xiaozhong Inventor before: Zhang Peng Inventor before: Peng Xinxin Inventor before: Yu Yi Inventor before: Wu Guangyao Inventor before: Wang Shunbo Inventor before: Zhai Li Inventor before: Nie Bingbing Inventor before: Mao Yingchi Inventor before: Zhao Huan Inventor before: Chen Hao Inventor before: Wang Qiang Inventor before: Guo Youan Inventor before: Wang Longbao Inventor before: Word wave Inventor before: Chen Weidong Inventor before: Li Hongbo |
|
| CB03 | Change of inventor or designer information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |