CN114565252B - Decision-making method for determining bridge maintenance sequence - Google Patents
Decision-making method for determining bridge maintenance sequence Download PDFInfo
- Publication number
- CN114565252B CN114565252B CN202210157389.5A CN202210157389A CN114565252B CN 114565252 B CN114565252 B CN 114565252B CN 202210157389 A CN202210157389 A CN 202210157389A CN 114565252 B CN114565252 B CN 114565252B
- Authority
- CN
- China
- Prior art keywords
- bridge
- weight
- evaluation
- type
- score
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention provides a decision-making method for determining a bridge maintenance sequence, and belongs to the technical field of bridge maintenance. According to the method, a comprehensive evaluation type representative to the bridge maintenance score is extracted through principal component analysis, and the evaluation type and the initial weight of an influence factor are obtained through migration learning and a Delphi method; obtaining the difference between the maintenance score of the bridge and the actual score; optimizing the weight of the evaluation type; optimizing influence factor weight according to the evaluation type errors; obtaining an updated bridge maintenance score; sorting the updated bridge maintenance scores by adopting a merging sorting method; and maintaining the bridge according to the bridge maintenance sequence. The method adopts the transfer learning and the Delphi method to obtain the evaluation type and the initial weight of the influence factor, optimizes the evaluation type weight, and optimizes the influence factor weight according to the evaluation type error; and after the bridge maintenance scores are recalculated, sequencing the bridge maintenance scores, so that the accuracy of maintaining the bridges is improved.
Description
Technical Field
The invention relates to the technical field of bridge maintenance, in particular to a decision-making method for determining a bridge maintenance sequence.
Background
At present, the bridge maintenance work in China is to simply maintain the bridge with problems, namely, the bridge is in the passive maintenance stage. Because the infrastructure construction of China occupies most of the financial budget and is limited by the national financial resources, the capital invested in the bridge maintenance is far from meeting the requirement, and therefore, the limited capital needs to be reasonably distributed.
The sequencing and optimization of the bridge maintenance projects need to consider the degree of urgency of each project and the balance among regions. The current bridge maintenance sequencing methods are common in the following two types, firstly, sequencing among a plurality of selectable maintenance schemes; and secondly, sequencing maintenance schemes among bridges in the road network. The common goal of both methods is to bring the greatest economic and social benefits into play with limited maintenance capital.
The content of the road network level bridge maintenance decision optimization ordering is to determine the maintenance sequence of the bridge facilities in the road network, namely, to make a reasonable maintenance sequence under the condition of limited funds, and to preferentially arrange bridge projects with large influence on traffic transportation, poorer technical conditions and high maintenance level so as to reasonably distribute and use the limited funds, ensure that the bridge network is at the specified service level and enable the maintenance funds to exert the best economic benefit and social benefit.
The technical condition of the bridge is constantly changing along with time in the service life cycle of the bridge, the degradation of the bridge structure is an inevitable natural process, and the bridge cannot always maintain a high service level. Factors influencing the technical condition or structural deterioration of the bridge mainly comprise the engineering structure of the bridge, the load of the bridge, the traffic flow of the bridge, the natural climate, the material of the bridge, the level of maintenance management of the bridge and the like.
The bridge evaluation is divided into general evaluation and adaptability evaluation. The general evaluation is to determine the technical condition grade of the bridge and provide various maintenance measures of the bridge by comprehensively evaluating the technical conditions of all parts of the bridge according to the regular inspection data of the bridge. The bridge adaptability evaluation comprises the following contents: according to the regular inspection and special inspection data of the bridge, the actual bearing capacity, the traffic capacity and the disaster resistance capacity of the bridge are evaluated by combining the test and the structural stress analysis, and a bridge maintenance and transformation scheme is provided.
Chinese patent application CN109816250A discloses a method for determining maintenance priority of concrete bridges in a highway network, which is an attribute reduction-based method, and obtains the information content of each attribute set of a bridge, and further obtains the importance of the attributes, and then obtains the weight of the attributes through importance normalization. According to the scheme, the attribute values of bridge technical state, bridge safety, bridge durability, bridge adaptability and economics are used, and in the face of all bridges in the existing road network, data behind the 5 attributes cannot be acquired for all concrete bridges in the existing highway network. In real life, the service state of a bridge changes according to the change of the external environment, and an evaluation system which can better reflect the change is needed to make a decision on the maintenance priority.
The prior art has the following defects:
1. in the decision of bridge maintenance, all bridges in the road network cannot be judged, and the method is not suitable for being widely applied in actual engineering practice. The method of attribute reduction is used, which easily causes the explosive increase of the calculation amount.
Disclosure of Invention
In order to solve the technical problems in the prior art, the invention provides a decision method for determining a bridge maintenance sequence, which comprises the steps of converting indexes influencing the bridge maintenance score into a few comprehensive evaluation types with better representativeness through principal component analysis, and obtaining the evaluation types and the initial weight of an influence factor through migration learning and a Delphi method; obtaining the difference between the maintenance score of the bridge and the actual score; optimizing the weight of the evaluation type; optimizing influence factor weight according to the evaluation type errors; obtaining an updated bridge maintenance score; sequencing the updated bridge maintenance scores by adopting a merging sequencing method, and determining the sequence as a bridge maintenance sequence; and maintaining the bridge according to the bridge maintenance sequence. The method adopts the transfer learning and the Delphi method to obtain the evaluation type and the initial weight of the influence factor, optimizes the evaluation type weight, and optimizes the influence factor weight according to the evaluation type error; and after the bridge maintenance scores are recalculated, the bridge maintenance scores are sequenced, so that the accuracy of maintaining the bridge is improved, and the scientific decision of the in-service bridge maintenance of all levels of roads and materials in the whole road network can be realized.
The invention provides a decision method for determining a bridge maintenance sequence, which comprises the following steps:
determining the evaluation type and the influence factor of the bridge;
respectively determining the evaluation type and the initialization weight of the influence factor, specifically comprising the following steps: obtaining a first initialization weight through transfer learning, obtaining a second initialization weight through a Delphi method, and determining the initialization weights of the evaluation types and the influence factors according to the first initialization weight and the second initialization weight;
obtaining the difference between the maintenance score of the bridge and the actual score;
optimizing the evaluation type weight according to the difference between the bridge maintenance score and the actual score;
optimizing influence factor weight according to the evaluation type errors;
obtaining an updated bridge maintenance score;
and sequencing the updated bridge maintenance scores by adopting a merging sequencing method, and determining the sequence as a bridge maintenance sequence.
Preferably, the initialization weights of the evaluation type and the influence factor are respectively determined by the following formulas:
initialization weight W for each evaluation type x (i) x(i) Comprises the following steps:
W x(i) =λ A x(i) W A x(i) +λ B x(i) W B x(i) (1)
initialization weight W for each influence factor y (j) y(j) Comprises the following steps:
W y(j) =λ A y(j) W A y(j) +λ B y(j) W B y(j) (4)
wherein the content of the first and second substances,
x (i) is the ith evaluation type;
y (j) is the influence factor corresponding to the evaluation type x (i);
W A x(i) first initialization weight, W, for evaluation type of transfer learning method B x(i) Second initialization weight, W, for the type of evaluation of the Delphi method x(i) Initialization weights for evaluation type x (i);
a first initialization weight for an impact factor of the transfer learning method,second initialization weight, W, of influence factor for Delphi y(j) An initialization weight for the impact factor y (i);
the weight of the evaluation factor weight of the transfer learning method in the evaluation type weight is taken up;
the weight of the evaluation factor of the Delphi method accounts for the weight of the evaluation type;
the influence factor weight of the transfer learning method accounts for the proportion of the influence factor weight;
Preferably, the evaluation type of the bridge is determined by a principal component analysis method according to the index influencing the maintenance score of the bridge.
Preferably, the evaluation types include a component technical status, a bridge structure form and scale, an adaptability, a trafficability, and a disaster resistance.
Preferably, the influence factors of the technical condition of the part component comprise disease type quantity scale, component evaluation result and component evaluation result; the influence factors of the bridge structure form and scale comprise a bridge structure type and a bridge span type; influence factors of the adaptability comprise design load and bridge age; influence factors of traffic capacity comprise wide roads, narrow bridges, traffic volume and heavy traffic; the influence factors of the disaster resistance include flood resistance, frost resistance and earthquake resistance.
Preferably, the bridge maintenance score is obtained according to the scale grade of the influence factors.
Preferably, the shock resistance of the kth bridge and the wide and narrow bridges are divided into 2 scale grades; the anti-freezing capacity and the flood resistance are divided into 3 scale grades; dividing the disease condition, the component evaluation result, the bridge structure type and the bridge span type into 4 scale grades; the design load, bridge age, traffic volume and heavy traffic were divided into 5 scales.
Preferably, the scale levels are classified into 2, with scores of 50 and 100, respectively; for the scale grade division into 3 grades, the scores are 40, 60 and 100, respectively; for a scale level of 4, scores of 40, 60, 80 and 100, respectively; for the scale grade division into 5 grades, the scores are 20, 40, 60, 80 and 100 respectively.
Preferably, the obtaining of the bridge maintenance score by constructing a decision tree for bridge maintenance specifically includes: the bridge maintenance Score is taken as a root node, the bridge maintenance Score is divided into five sub-nodes including a part of component technical condition x (1), a bridge structure form and scale x (2), an adaptive capacity x (3), a traffic capacity x (4) and an anti-disaster capacity x (5), and the bridge maintenance Score is obtained through the following formula:
the sub-nodes of the part component technical condition x (1) are divided into three leaf nodes of a disease type quantity scale y (1), a component evaluation result y (2) and a part evaluation result y (3);
the bridge structure form and the scale x (2) are divided into two leaf nodes of a bridge structure type y (4) and a bridge span type y (5);
dividing the adaptive capacity x (3) into two leaf nodes of a design load y (6) and a bridge age y (7);
the traffic capacity x (4) is divided into two leaf nodes of a wide road narrow bridge y (8) and a traffic volume and heavy traffic y (9);
the disaster resistance capability x (5) is divided into three leaf nodes of flood resistance capability y (10), frost damage resistance capability y (11) and earthquake resistance capability y (12);
the bridge maintenance Score is obtained according to the following formula:
wherein the content of the first and second substances,
x (i) is the ith evaluation type;
y (j) is an influence factor corresponding to the evaluation type x (i), the value of j is related to x (i), and if i is 1, the value of j is 1, 2 and 3; if i takes 2, j takes 4 and 5; if i takes 3, j takes 6 and 7; if i takes 4, j takes 8 and 9; if i takes 5, j takes values of 10, 11 and 12.
Preferably, the process of obtaining the updated bridge maintenance score according to the optimized evaluation type and the influence factor weight is as follows:
writing equation (7) into vector form as follows:
Score=W x T x (10)
wherein the content of the first and second substances,
x T where x is x (1) x (2) L x (5) ═ x T The transposed matrix of (2);
in the process of training the model weight, obtaining a difference value of the bridge maintenance Score according to the trained bridge maintenance Score Score and the actual bridge maintenance Score Score', and obtaining the difference value according to the following formula:
ΔS=|Score-Score'| (11)
wherein, the delta S is the difference value of the maintenance scores of the bridge;
calculating the difference value delta W of each evaluation type weight by adopting the following formula according to the difference value of the bridge maintenance scores x(i) :
The difference value Δ W of each evaluation type weight is learned by using the following formula x(i) To update the evaluation type weights:
in the formula, λ x(i) To evaluate the learning rate, W, of type x (i) x(i) ' weight of updated evaluation type x (i);
the error value Δ x (i) of the evaluation type x (i) is obtained using the following equation:
the weight difference value formula of the influence factors y (i) is expressed as:
the weight of the influence factor y (i) is optimized as follows:
W y(j) '=W y(j) +λ y(j) ΔW y(j) (16)
and recalculating the updated bridge maintenance Score by using the following formula:
in the formula, λ y(j) W is the learning rate of the influencing factor y (i) y(j) ' is the weight of the updated impact factor y (i).
Compared with the prior art, the invention has the following beneficial effects:
1. according to the method, the migration learning method and the Delphi method are combined to obtain the evaluation types and the initial weights of the influence factors, so that the inheritance relationships of bridge technologies and the like are considered, and the bridge weights are learned on a time level; the method also performs weight learning on the bridge from top to bottom, from inside to outside and other spaces from the technical condition or structural deterioration of the current bridge and from the property and characteristic of the current bridge needing maintenance, and finally fuses the weights on the time and space level;
2. according to the method, after the bridge maintenance score is obtained, the evaluation type weight is optimized, the influence factor weight is optimized according to the optimized evaluation type, the bridge maintenance score is recalculated and sequenced, and bridge maintenance is performed according to the sequencing, so that the finally obtained bridge maintenance sequencing is more accurate, and a certain meaning reference is provided for the sequence of the bridge maintenance engineering.
3. The method converts indexes influencing the maintenance score of the bridge into a few representative comprehensive evaluation types through principal component analysis, and starts from the properties and characteristics of the bridge needing maintenance at present, so that the finally obtained bridge maintenance sequence decision is more practical.
4. The invention is not only suitable for the maintenance of all road networks and bridges made of various materials such as steel structures, masonry structures, concrete and the like, but also can be used for expressways, first-level roads, second-level roads, third-level roads and the like, and has wider application range.
Drawings
FIG. 1 is a flow chart of a decision method for determining a bridge maintenance sequence according to an embodiment of the present invention;
FIG. 2 is a flow chart of a decision method for determining a bridge maintenance sequence according to another embodiment of the present invention;
FIG. 3 is a schematic diagram of an evaluation type and impact factor weight optimization process according to an embodiment of the invention.
Detailed Description
The following detailed description of the embodiments of the present invention is provided in conjunction with the accompanying drawings of fig. 1-3.
The invention provides a decision method for determining a bridge maintenance sequence, which comprises the following steps:
determining the evaluation type and the influence factor of the bridge;
respectively determining the evaluation type and the initialization weight of the influence factor, specifically comprising the following steps: acquiring a first initialization weight through transfer learning, acquiring a second initialization weight through a Delphi method, and determining the initialization weights of the evaluation types and the influence factors according to the first initialization weight and the second initialization weight;
obtaining the difference between the maintenance score of the bridge and the actual score;
optimizing the evaluation type weight according to the difference between the bridge maintenance score and the actual score;
optimizing the weight of the influence factors according to the evaluation type errors;
obtaining an updated bridge maintenance score;
sequencing the updated bridge maintenance scores by adopting a merging sequencing method, and determining the sequence as a bridge maintenance sequence;
and maintaining the bridge according to the bridge maintenance sequence.
According to a specific embodiment of the present invention, the initialization weights of the evaluation types and the influence factors are respectively determined by the following formulas:
initialization weight W for each evaluation type x (i) x(i) Comprises the following steps:
W x(i) =λ A x(i) W A x(i) +λ B x(i) W B x(i) (1)
initialization weight W for each influence factor y (j) y(j) Comprises the following steps:
W y(j) =λ A y(j) W A y(j) +λ B y(j) W B y(j) (4)
wherein the content of the first and second substances,
x (i) is the ith evaluation type;
y (j) is the influence factor corresponding to the evaluation type x (i);
W A x(i) first initialization weight, W, for the type of evaluation of the transfer learning method B x(i) Second initialization weight, W, for the type of evaluation of the Delphi method x(i) Initialization weights for evaluation type x (i);
a first initialization weight for an impact factor of the transfer learning method,second initialization weight, W, of influence factor for Delphi y(j) An initialization weight for the impact factor y (i);
the evaluation factor weight of the transfer learning method accounts for the proportion of the evaluation type weight;
the weight of the evaluation factor weight of the Delphi method accounts for the weight of the evaluation type;
the influence factor weight of the transfer learning method accounts for the proportion of the influence factor weight;
is the influence of the Delphi methodThe factor weight is a proportion of the influence factor weight.
The formula is set by adopting the sizes of the migration learning method and the Delphi method evaluation type weight, and the purpose is to combine the advantages of the two methods when W is A x(i) When the value of (b) is larger, the migration learning method gives more weight to the ith evaluation type, and when W is larger, W gives more weight B x(i) When the value of (a) is larger, the Delphi method gives a higher weight to the ith evaluation type, and only when the values of the two methods are smaller, the two methods do not give a weight to the evaluation type. Only if two methods are not regarded as the evaluation types, the method gives a lower initial weight, and only if one method is regarded as the evaluation type, the method gives a high score, and correction is carried out in the training process, so that the training pressure is reduced.
The invention determines the weight under the current evaluation system by taking the weights of the previous bridge evaluation types and the influence factors as reference by a transfer learning method. The conventional weights and the weights under the current model are all directed to the bridge technical condition or the structural deterioration, so the weights of the conventional bridge evaluation types and the influence factors obtained by the transfer learning method can be used as a part of the weights of the current model.
The delphi method is essentially a feedback anonymity function query method. The general process is as follows: after the opinions of the experts are characterized for the problems to be predicted, the problems are sorted, induced and counted, and then fed back to each expert anonymously, the opinions are solicited again, concentrated and fed back again until the consistent opinions are obtained.
The process of obtaining the weight by the delphi method is as follows:
firstly, normalizing each column of a judgment matrix A:
wherein the content of the first and second substances,
judgment matrixThe method is obtained according to the Delphi method, namely, the expert can obtain the intelligence collectively, so that the method is more reasonable; m rows and m columns.
a ij Judging the element of the ith row and the jth column in the matrix A;
secondly, adding the normalized matrixes of each column according to rows to obtain a vector M, wherein the ith element in the vector M is obtained as follows:
then, the vector M is set to (M) 1 ,M 2 ,Λ,M m ) T And normalizing to obtain a vector W, wherein the ith element in the vector W is obtained as follows:
the obtained characteristic vector W meets the condition of consistency check B =(W 1 ,W 2 ,Λ,W m ) T I.e. the weight.
According to a specific embodiment of the invention, the evaluation type of the bridge is determined by a principal component analysis method according to the index affecting the maintenance score of the bridge.
According to an embodiment of the present invention, the evaluation types include a construction technical status, a bridge structural form and scale, an adaptability, a trafficability, and a disaster resistance. In order to determine the maintenance sequence of bridge facilities in a road network and achieve the purpose of reasonably distributing and using limited funds, the invention converts indexes influencing the maintenance score of the bridge into a few representative comprehensive evaluation types through principal component analysis to obtain the five evaluation types.
According to a specific embodiment of the invention, the influence factors of the technical condition of the component include disease type number scale, component evaluation result and component evaluation result; the influence factors of the bridge structure form and scale comprise a bridge structure type and a bridge span type; influence factors of the adaptability comprise design load and bridge age; influence factors of traffic capacity comprise wide roads, narrow bridges, traffic volume and heavy traffic; the influence factors of the disaster resistance include flood resistance, frost resistance and earthquake resistance.
According to an embodiment of the present invention, the bridge maintenance score is obtained according to the scale grade of the influencing factor.
According to a specific embodiment of the invention, the shock resistance of the kth bridge and the wide and narrow bridges are divided into 2 scale grades; the anti-freezing capacity and the flood resistance are divided into 3 scale grades; dividing the disease condition, the component evaluation result, the bridge structure type and the bridge span type into 4 scale grades; the design load, bridge age, traffic volume and heavy traffic were divided into 5 scales.
According to a particular embodiment of the invention, the scale grade is divided into 2 grades with scores of 50 and 100, respectively; for the scale grade division into 3 grades, the scores are 40, 60 and 100, respectively; for a scale level of 4, scores of 40, 60, 80 and 100, respectively; for the scale grade division into 5 grades, the scores are 20, 40, 60, 80 and 100 respectively. The different influence factor scores can be determined according to the scale grades, and the setting of the scale grades and the scores refers to general Specification for design of highway bridges and culverts (JTG D60-2015), < Specification for earthquake resistance of highway engineering (JTG B02-2013), < Specification for earthquake resistance of highway bridges and culverts > (JTG/T B02-01-2008).
According to a specific embodiment of the present invention, obtaining a bridge maintenance score by constructing a decision tree for bridge maintenance specifically includes: the bridge maintenance Score is taken as a root node, the bridge maintenance Score is divided into five sub-nodes including a part of component technical condition x (1), a bridge structure form and scale x (2), an adaptive capacity x (3), a traffic capacity x (4) and an anti-disaster capacity x (5), and the bridge maintenance Score is obtained through the following formula:
the sub-nodes of the part component technical condition x (1) are divided into three leaf nodes of a disease type quantity scale y (1), a component evaluation result y (2) and a part evaluation result y (3);
the bridge structure form and the scale x (2) are divided into two leaf nodes of a bridge structure type y (4) and a bridge span type y (5);
dividing the adaptability x (3) into two leaf nodes of a design load y (6) and a bridge age y (7);
the traffic capacity x (4) is divided into two leaf nodes of a wide road narrow bridge y (8) and a traffic volume and heavy traffic y (9);
the disaster resistance capability x (5) is divided into three leaf nodes of flood resistance capability y (10), frost damage resistance capability y (11) and earthquake resistance capability y (12);
the bridge maintenance Score is obtained according to the following formula:
wherein the content of the first and second substances,
x (i) is the ith evaluation type;
y (j) is an influence factor corresponding to the evaluation type x (i), the value of j is related to x (i), and if i is 1, the value of j is 1, 2 and 3; if i takes 2, j takes 4 and 5; if i takes 3, j takes 6 and 7; if i takes 4, j takes 8 and 9; if i takes 5, j takes values of 10, 11 and 12.
According to a specific embodiment of the present invention, the process of obtaining the updated bridge maintenance score according to the optimized evaluation type and the influence factor weight is as follows:
writing equation (7) into vector form as follows:
Score=W x T x (10)
wherein the content of the first and second substances,
x T where x is x (1) x (2) L x (5) ═ x T The transposed matrix of (2);
in the process of training the model weight, obtaining a difference value of the bridge maintenance Score according to the trained bridge maintenance Score Score and the actual bridge maintenance Score Score', and obtaining the difference value according to the following formula:
ΔS=|Score-Score'| (11)
wherein, the delta S is the difference value of the maintenance scores of the bridge;
calculating the difference value delta W of each evaluation type weight by adopting the following formula according to the difference value between the actual scores of the bridge maintenance score x(i) :
The following formula is adopted to learn the difference value delta W of each evaluation type weight x(i) To update the evaluation type weights:
in the formula, λ x(i) To evaluate the learning rate of type x (i), W x(i) ' weight of updated evaluation type x (i);
the error value Δ x (i) of the evaluation type x (i) is obtained using the following equation:
the weight difference value formula of the influence factors y (i) is expressed as:
the weight of the influence factor y (i) is optimized as follows:
W y(j) '=W y(j) +λ y(j) ΔW y(j) (16)
and recalculating the bridge maintenance Score by adopting the following formula:
in the formula, λ y(j) W is the learning rate of the influencing factor y (i) y(j) ' is the weight of the updated impact factor y (i).
Example 1
The decision method for determining the maintenance sequence of the bridge according to the present invention is described in detail below according to an embodiment of the present invention.
The invention provides a decision method for determining a bridge maintenance sequence, which comprises the following steps:
determining the evaluation type and the influence factor of the bridge;
respectively determining the evaluation type and the initialization weight of the influence factor, specifically comprising the following steps: obtaining a first initialization weight through transfer learning, obtaining a second initialization weight through a Delphi method, and determining the initialization weights of the evaluation types and the influence factors according to the first initialization weight and the second initialization weight;
obtaining the difference between the maintenance score of the bridge and the actual score;
optimizing the evaluation type weight according to the difference between the bridge maintenance score and the actual score;
optimizing influence factor weight according to the evaluation type errors;
obtaining an updated bridge maintenance score;
and sequencing the updated bridge maintenance scores by adopting a merging sequencing method, and determining the sequence as a bridge maintenance sequence.
Example 2
The decision method for determining the maintenance sequence of the bridge according to the present invention is described in detail below according to an embodiment of the present invention.
The invention provides a decision method for determining a bridge maintenance sequence, which comprises the following steps:
determining the evaluation type and the influence factor of the bridge;
respectively determining the evaluation type and the initialization weight of the influence factor, specifically comprising the following steps: obtaining a first initialization weight through transfer learning, obtaining a second initialization weight through a Delphi method, and determining the initialization weights of the evaluation types and the influence factors according to the first initialization weight and the second initialization weight; the evaluation type and the initialization weight of the influence factor are respectively determined by the following formulas:
initialization weight W for each evaluation type x (i) x(i) Comprises the following steps:
W x(i) =λ A x(i) W A x(i) +λ B x(i) W B x(i) (1)
initialization weight W for each influence factor y (j) y(j) Comprises the following steps:
W y(j) =λ A y(j) W A y(j) +λ B y(j) W B y(j) (4)
wherein, the first and the second end of the pipe are connected with each other,
x (i) is the ith evaluation type;
y (j) is the influence factor corresponding to the evaluation type x (i);
W A x(i) first initialization weight, W, for the type of evaluation of the transfer learning method B x(i) Second initialization weight, W, for the type of evaluation of the Delphi method x(i) Initialization weights for evaluation type x (i);
a first initialization weight for an impact factor of the transfer learning method,second initialization weight, W, of influence factor for Delphi y(j) An initialization weight for the impact factor y (i);
the evaluation factor weight of the transfer learning method accounts for the proportion of the evaluation type weight;
the weight of the evaluation factor of the Delphi method accounts for the weight of the evaluation type;
the influence factor weight of the transfer learning method accounts for the proportion of the influence factor weight;
the influence factor weight of the Delphi method accounts for the proportion of the influence factor weight;
obtaining the difference between the maintenance score of the bridge and the actual score;
optimizing the evaluation type weight according to the difference between the bridge maintenance score and the actual score;
optimizing the weight of the influence factors according to the evaluation type errors;
obtaining an updated bridge maintenance score;
sequencing the updated bridge maintenance scores by adopting a merging sequencing method, and determining the sequence as a bridge maintenance sequence;
and maintaining the bridge according to the bridge maintenance sequence.
Example 3
The decision method for determining the maintenance sequence of the bridge according to the present invention is described in detail below according to an embodiment of the present invention.
The invention provides a decision method for determining a bridge maintenance sequence, which comprises the following steps:
determining the evaluation type and the influence factor of the bridge; determining the evaluation type of the bridge through a principal component analysis method according to the indexes influencing the maintenance score of the bridge; the evaluation types comprise the technical condition of a component, the structural form and scale of the bridge, the adaptability, the traffic capacity and the disaster resistance; the influence factors of the technical condition of the part component comprise disease type quantity scale, component evaluation results and component evaluation results; the influence factors of the form and scale of the bridge structure comprise the type of the bridge structure and the type of the bridge span; influence factors of the adaptability comprise design load and bridge age; influence factors of traffic capacity comprise wide roads, narrow bridges, traffic volume and heavy traffic; the influence factors of the disaster resistance include flood resistance, frost damage resistance and earthquake resistance;
in order to more accurately determine the evaluation types and the weight of the influence factors, the indexes influencing the maintenance score of the bridge are converted into a few comprehensive evaluation types with better representativeness through principal component analysis, and then the initialization weight is determined from two aspects of transfer learning and the Delphi method.
The technical condition of the bridge part component reflects the structural defect, the functional condition and the working condition of the bridge, and the technical condition of the bridge part component is used as an important factor of maintenance sequencing. The method is preferably classified according to the technical condition grades of the bridge part components, wherein the first group is four-type and five-type bridges, the second group is three-type bridges, the third group is two-type bridges, and the fourth group is one-type bridges.
Respectively determining the evaluation type and the initialization weight of the influence factor, specifically comprising the following steps: obtaining a first initialization weight through transfer learning, obtaining a second initialization weight through a Delphi method, determining the initialization weights of the evaluation type and the influence factor by adopting a linear combination mode according to the first initialization weight and the second initialization weight, and respectively determining the initialization weights of the evaluation type and the influence factor by adopting the following formulas:
initialization weight W for each evaluation type x (i) x(i) Comprises the following steps:
W x(i) =λ A x(i) W A x(i) +λ B x(i) W B x(i) (1)
initialization weight W for each influence factor y (j) y(j) Comprises the following steps:
W y(j) =λ A y(j) W A y(j) +λ B y(j) W B y(j) (4)
wherein the content of the first and second substances,
x (i) is the ith evaluation type;
y (j) is the influence factor corresponding to the evaluation type x (i);
W A x(i) first initialization weight, W, for the type of evaluation of the transfer learning method B x(i) Second initialization weight, W, for the type of evaluation of the Delphi method x(i) Initialization weights for evaluation type x (i);
a first initialization weight for an impact factor of the transfer learning method,second initialization weight, W, of influence factor for Delphi y(j) An initialization weight for the impact factor y (i);
the evaluation factor weight of the transfer learning method accounts for the proportion of the evaluation type weight;
the weight of the evaluation factor of the Delphi method accounts for the weight of the evaluation type;
the influence factor weight of the transfer learning method accounts for the proportion of the influence factor weight;
the influence factor weight of the Delphi method accounts for the proportion of the influence factor weight;
obtaining a bridge maintenance score by constructing a decision tree according to the scale grade of the influence factor, and obtaining a difference value between the bridge maintenance score and the actual score;
optimizing the evaluation type weight according to the difference between the bridge maintenance score and the actual score;
optimizing the weight of the influence factors according to the evaluation type errors;
obtaining an updated bridge maintenance score;
sequencing the updated bridge maintenance scores by adopting a merging sequencing method, and determining the sequence as a bridge maintenance sequence;
and maintaining the bridge according to the bridge maintenance sequence.
Example 4
The decision method for determining the maintenance sequence of the bridge according to the present invention is described in detail below according to an embodiment of the present invention.
The invention provides a decision method for determining a bridge maintenance sequence, which comprises the following steps:
determining the evaluation type and the influence factor of the bridge; determining the evaluation type of the bridge through a principal component analysis method according to the indexes influencing the maintenance score of the bridge; the evaluation types comprise the technical condition of a component, the structural form and scale of the bridge, the adaptability, the traffic capacity and the disaster resistance; the influence factors of the technical condition of the part component comprise disease type quantity scale, component evaluation results and component evaluation results; the influence factors of the bridge structure form and scale comprise a bridge structure type and a bridge span type; influence factors of the adaptability comprise design load and bridge age; influence factors of traffic capacity comprise wide roads, narrow bridges, traffic volume and heavy traffic; the influence factors of the disaster resistance include flood resistance, frost damage resistance and earthquake resistance;
respectively determining the evaluation type and the initialization weight of the influence factor, specifically comprising the following steps: acquiring a first initialization weight through transfer learning, acquiring a second initialization weight through a Delphi method, and determining the initialization weights of the evaluation types and the influence factors according to the first initialization weight and the second initialization weight; the evaluation type and the initialization weight of the influence factor are respectively determined by the following formulas:
initialization weight W for each evaluation type x (i) x(i) Comprises the following steps:
W x(i) =λ A x(i) W A x(i) +λ B x(i) W B x(i) (1)
initialization weight W for each influence factor y (j) y(j) Comprises the following steps:
W y(j) =λ A y(j) W A y(j) +λ B y(j) W B y(j) (4)
wherein the content of the first and second substances,
x (i) is the ith evaluation type;
y (j) is the influence factor corresponding to the evaluation type x (i);
W A x(i) first initialization weight, W, for the type of evaluation of the transfer learning method B x(i) Second initialization weight, W, for the type of evaluation of the Delphi method x(i) Initialization weights for evaluation type x (i);
a first initialization weight for an impact factor of the transfer learning method,second initialization weight, W, of influence factor for Delphi y(j) For initialisation of the influencing factors y (i)A weight;
the evaluation factor weight of the transfer learning method accounts for the proportion of the evaluation type weight;
the weight of the evaluation factor of the Delphi method accounts for the weight of the evaluation type;
the influence factor weight of the transfer learning method accounts for the proportion of the influence factor weight;
the influence factor weight of the Delphi method accounts for the proportion of the influence factor weight;
obtaining a bridge maintenance score by constructing a decision tree according to the scale grade of the influence factor, and obtaining a difference value between the bridge maintenance score and the actual score; dividing the shock resistance of the kth bridge and the wide and narrow bridges into 2 scale grades; the freezing damage resistance and flood resistance are divided into 3 scale grades; dividing the disease condition, the component evaluation result, the bridge structure type and the bridge span type into 4 scale grades; dividing the design load, the bridge age, the traffic volume and the heavy traffic into 5 scales; the scale levels are classified into 2, with scores of 50 and 100, respectively; for the scale grade division into 3 grades, the scores are 40, 60 and 100, respectively; for a scale level of 4, scores of 40, 60, 80 and 100, respectively; for the scale grade division into 5 grades, the scores are 20, 40, 60, 80 and 100 respectively; the classification of the technical condition grades of the parts of the bridge is classified according to the evaluation standard of the technical condition of the highway bridge. And ranking the bridge maintenance scores in each group by using a bridge maintenance decision priority algorithm model.
Obtaining a bridge maintenance score by constructing a decision tree for bridge maintenance, which specifically comprises the following steps: according to the regular inspection and special inspection data of the bridge, combined with the test and structural stress analysis, the bridge maintenance Score is taken as a root node, the bridge maintenance Score is divided into five sub-nodes of a part of component technical condition x (1), a bridge structural form and scale x (2), an adaptive capacity x (3), a traffic capacity x (4) and a disaster resistance capacity x (5), and the bridge maintenance Score is obtained through the following formula:
the sub-nodes of the part component technical condition x (1) are divided into three leaf nodes of a disease type quantity scale y (1), a component evaluation result y (2) and a part evaluation result y (3);
the bridge structure form and the scale x (2) are divided into two leaf nodes of a bridge structure type y (4) and a bridge span type y (5);
dividing the adaptive capacity x (3) into two leaf nodes of a design load y (6) and a bridge age y (7);
the traffic capacity x (4) is divided into two leaf nodes of a wide road narrow bridge y (8) and a traffic volume and heavy traffic y (9);
the disaster resistance capability x (5) is divided into three leaf nodes of flood resistance capability y (10), frost damage resistance capability y (11) and earthquake resistance capability y (12);
the bridge maintenance Score is obtained according to the following formula:
wherein the content of the first and second substances,
x (i) is the ith evaluation type;
y (j) is an influence factor corresponding to the evaluation type x (i), the value of j is related to x (i), and if i is 1, the value of j is 1, 2 and 3; if i takes 2, j takes 4 and 5; if i takes 3, j takes 6 and 7; if i takes 4, j takes 8 and 9; if i takes 5, j takes values of 10, 11 and 12.
Obtaining a bridge maintenance score, obtaining a difference value between the bridge maintenance score and the actual score, optimizing evaluation type weight according to the difference value between the bridge maintenance score and the actual score, and optimizing influence factor weight according to evaluation type errors to obtain an updated bridge maintenance score, wherein the specific process comprises the following steps:
writing equation (7) into a vector form as follows:
Score=W x T x (10)
wherein the content of the first and second substances,
x T where x is x (1) x (2) L x (5) ═ x T The transposed matrix of (2);
in the process of training the model weight, obtaining a difference value of the bridge maintenance Score according to the trained bridge maintenance Score Score and the actual bridge maintenance Score Score', and obtaining the difference value according to the following formula:
ΔS=|Score-Score'| (11)
wherein, the delta S is the difference value of the maintenance scores of the bridge;
calculating the difference value delta W of each evaluation type weight by adopting the following formula according to the difference value of the bridge maintenance scores x(i) :
Considering that there may be data with errors in the data, and the weight of the influence factor also influences the difference of the bridge score, the difference Δ W of the weight of a part of the evaluation types is learned x(i) To update the evaluation type weights by learning the difference Δ W of each evaluation type weight using the following formula x(i) To update the evaluation type weights:
in the formula, λ x(i) To evaluate the learning rate of type x (i), W x(i) ' weight of updated evaluation type x (i);
the error value Δ x (i) of the evaluation type x (i) is obtained using the following equation:
since the influence factor is linear with the evaluation type, the weight difference value of the influence factor y (i) is expressed as:
the weight of the influence factor y (i) is optimized as follows:
W y(j) '=W y(j) +λ y(j) ΔW y(j) (16)
and recalculating the updated bridge maintenance Score by using the following formula:
in the formula, λ y(j) W is the learning rate of the influencing factor y (i) y(j) ' is the weight of the updated impact factor y (i).
Sorting the updated bridge maintenance scores Score from small to large by adopting a merging sorting method, and determining the sequence as a bridge maintenance sequence;
and maintaining the bridge according to the bridge maintenance sequence.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (9)
1. A decision method for determining a bridge maintenance sequence is characterized by comprising the following steps:
determining the evaluation type and the influence factor of the bridge;
respectively determining the evaluation type and the initialization weight of the influence factor, specifically comprising the following steps: obtaining a first initialization weight through transfer learning, obtaining a second initialization weight through a Delphi method, and determining the initialization weights of the evaluation types and the influence factors according to the first initialization weight and the second initialization weight;
obtaining the difference between the maintenance score of the bridge and the actual score;
optimizing the evaluation type weight according to the difference between the bridge maintenance score and the actual score;
optimizing influence factor weight according to the evaluation type errors;
obtaining an updated bridge maintenance score;
sequencing the updated bridge maintenance scores by adopting a merging sequencing method, and determining the sequence as a bridge maintenance sequence;
the evaluation type and the initialization weight of the influence factor are determined by the following formulas respectively:
initialization weight W for each evaluation type x (i) x(i) Comprises the following steps:
W x(i) =λ A x(i) W A x(i) +λ B x(i) W B x(i) (1)
initialization weight W for each influence factor y (j) y(j) Comprises the following steps:
W y(j) =λ A y(j) W A y(j) +λ B y(j) W B y(j) (4)
wherein, the first and the second end of the pipe are connected with each other,
x (i) is the ith evaluation type;
y (j) is the influence factor corresponding to the evaluation type x (i);
W A x(i) first initialization weight, W, for the type of evaluation of the transfer learning method B x(i) Second initialization weight, W, for the evaluation type of Delphi x(i) Initialization weights for evaluation type x (i);
a first initialization weight for an impact factor of the transfer learning method,second initialization weight, W, of influence factor for Delphi y(j) An initialization weight for the impact factor y (i);
the evaluation factor weight of the transfer learning method accounts for the proportion of the evaluation type weight;
the weight of the evaluation factor of the Delphi method accounts for the weight of the evaluation type;
the influence factor weight of the transfer learning method accounts for the proportion of the influence factor weight;
2. The decision-making method for determining a bridge maintenance sequence according to claim 1, characterized in that the evaluation type of the bridge is determined by a principal component analysis method according to an index affecting the maintenance score of the bridge.
3. The decision method for determining a bridge maintenance sequence according to claim 1, wherein the evaluation type comprises a component technical status, a bridge structure form and scale, an adaptability, a trafficability and a disaster resistance.
4. The decision method for determining a bridge maintenance sequence according to claim 1, wherein the influence factors of the technical conditions of the component include a disease type number scale, a component evaluation result and a component evaluation result; the influence factors of the form and scale of the bridge structure comprise the type of the bridge structure and the type of the bridge span; influence factors of the adaptability comprise design load and bridge age; influence factors of traffic capacity comprise wide roads, narrow bridges, traffic volume and heavy traffic; the influence factors of the disaster resistance include flood resistance, frost resistance and earthquake resistance.
5. The decision method for determining a bridge maintenance sequence according to claim 1, wherein the bridge maintenance score is obtained according to the scale level of the influence factors.
6. The decision method for determining the maintenance sequence of the bridge according to claim 5, wherein the shock resistance and the wide and narrow bridges of the kth bridge are divided into 2 scale levels; the anti-freezing capacity and the flood resistance are divided into 3 scale grades; dividing the disease condition, the component evaluation result, the bridge structure type and the bridge span type into 4 scale grades; the design load, bridge age, traffic volume and heavy traffic were divided into 5 scales.
7. The decision-making method for determining the bridge maintenance sequence according to claim 6, wherein the scale grade is divided into 2 grades, and the scores are respectively 50 and 100; for the scale grade division into 3 grades, the scores are 40, 60 and 100, respectively; for a scale rating of 4, scores of 40, 60, 80 and 100, respectively; for the scale grade division into 5 grades, the scores are 20, 40, 60, 80 and 100 respectively.
8. The decision method for determining a bridge maintenance sequence according to claim 7, wherein obtaining a bridge maintenance score by constructing a decision tree for bridge maintenance specifically comprises: the bridge maintenance Score is divided into five sub-nodes, namely a partial component technical condition x (1), a bridge structure form and scale x (2), an adaptive capacity x (3), a traffic capacity x (4) and a disaster-resistant capacity x (5), and is obtained through the following formula:
the sub-nodes of the part component technical condition x (1) are divided into three leaf nodes of a disease type quantity scale y (1), a component evaluation result y (2) and a part evaluation result y (3);
the bridge structure form and the scale x (2) are divided into two leaf nodes of a bridge structure type y (4) and a bridge span type y (5);
dividing the adaptive capacity x (3) into two leaf nodes of a design load y (6) and a bridge age y (7);
the traffic capacity x (4) is divided into two leaf nodes of a wide road narrow bridge y (8) and a traffic volume and heavy traffic y (9);
the disaster resistance capability x (5) is divided into three leaf nodes of flood resistance capability y (10), frost damage resistance capability y (11) and earthquake resistance capability y (12);
the bridge maintenance Score is obtained according to the following formula:
wherein, the first and the second end of the pipe are connected with each other,
x (i) is the ith evaluation type;
y (j) is an influence factor corresponding to the evaluation type x (i), the value of j is related to x (i), and if i is 1, the value of j is 1, 2 and 3; if i takes 2, j takes 4 and 5; if i takes 3, j takes 6 and 7; if i takes 4, j takes 8 and 9; if i takes 5, j takes values of 10, 11 and 12.
9. The decision-making method for determining a bridge maintenance sequence according to claim 8, wherein the process of obtaining the updated bridge maintenance score according to the optimized evaluation type and the influence factor weight is as follows:
writing equation (7) into vector form as follows:
Score=W x T x (10)
wherein the content of the first and second substances,
x T where x is x (1) x (2) L x (5) ═ x T The transposed matrix of (2);
in the process of training the model weight, obtaining a difference value of the bridge maintenance Score according to the trained bridge maintenance Score Score and the actual bridge maintenance Score Score', and obtaining the difference value according to the following formula:
ΔS=|Score-Score'| (11)
wherein, the delta S is the difference value of the maintenance scores of the bridge;
calculating the difference value delta W of each evaluation type weight by adopting the following formula according to the difference value of the bridge maintenance scores x(i) :
The following formula is adopted to learn the difference value delta W of each evaluation type weight x(i) To update the evaluation type weights:
in the formula, λ x(i) To evaluate the learning rate of type x (i), W x(i) ' weight of updated evaluation type x (i);
the error value Δ x (i) of the evaluation type x (i) is obtained using the following equation:
the weight difference value of the influence factors y (i) is formulated as:
the weight of the influence factor y (i) is optimized as follows:
W y(j) '=W y(j) +λ y(j) ΔW y(j) (16)
and recalculating the updated bridge maintenance Score by using the following formula:
in the formula, λ y(j) W is the learning rate of the influencing factor y (i) y(j) ' is the weight of the updated impact factor y (i).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210157389.5A CN114565252B (en) | 2022-02-21 | 2022-02-21 | Decision-making method for determining bridge maintenance sequence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210157389.5A CN114565252B (en) | 2022-02-21 | 2022-02-21 | Decision-making method for determining bridge maintenance sequence |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114565252A CN114565252A (en) | 2022-05-31 |
CN114565252B true CN114565252B (en) | 2022-08-30 |
Family
ID=81713870
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210157389.5A Active CN114565252B (en) | 2022-02-21 | 2022-02-21 | Decision-making method for determining bridge maintenance sequence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114565252B (en) |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101739416A (en) * | 2008-11-04 | 2010-06-16 | 未序网络科技(上海)有限公司 | Method for sequencing multi-index comprehensive weight video |
CN106548210B (en) * | 2016-10-31 | 2021-02-05 | 腾讯科技(深圳)有限公司 | Credit user classification method and device based on machine learning model training |
CN109816250B (en) * | 2019-01-28 | 2021-07-30 | 东北大学 | Method for determining maintenance priority of concrete bridges in highway network |
CN109948926A (en) * | 2019-03-14 | 2019-06-28 | 东北大学 | A kind of highway concrete-bridge maintenance technology Selection Method based on project period |
CN111160728B (en) * | 2019-12-09 | 2023-05-23 | 华南农业大学 | Road and bridge maintenance decision optimization method and device |
CN113222260A (en) * | 2021-05-19 | 2021-08-06 | 南方电网数字电网研究院有限公司 | Lightning trip-out rate prediction model of power transmission line |
CN113393094A (en) * | 2021-05-31 | 2021-09-14 | 欣旺达电子股份有限公司 | Energy system evaluation method, device, equipment and storage medium |
-
2022
- 2022-02-21 CN CN202210157389.5A patent/CN114565252B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN114565252A (en) | 2022-05-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111737916B (en) | Road and bridge disease analysis and maintenance decision method based on big data | |
CN104835103A (en) | Mobile network health evaluation method based on neural network and fuzzy comprehensive evaluation | |
CN107818237A (en) | The evaluation method of Damages of Asphalt Road Surface situation | |
Xia et al. | A data‐driven approach for regional bridge condition assessment using inspection reports | |
CN105913196A (en) | Navigation channel rectifying social stability risk automatically analyzing method and system | |
CN113554323A (en) | Emergency guarantee optimization design method for key nodes of road network | |
CN113052441A (en) | Emergency food supply risk analysis method | |
Attoh‐Okine | Grouping Pavement Condition Variables for Performance Modeling Using Self‐Organizing Maps | |
CN116739376A (en) | Highway pavement preventive maintenance decision method based on data mining | |
CN114529038A (en) | Intelligent matching business recruitment strategy system and method based on enterprise demands | |
CN113743831A (en) | Bridge network comprehensive performance evaluation method and device and storage medium | |
CN114565252B (en) | Decision-making method for determining bridge maintenance sequence | |
Zhao | Steel columns under fire—a neural network based strength model | |
CN112686396B (en) | Pavement maintenance property selection method, medium and system based on disease quantity | |
CN115545387A (en) | Method for evaluating fragility of immovable cultural relics | |
CN113536415A (en) | Comprehensive index system bridge comparison and selection method based on typical environment difference | |
CN114118688A (en) | Power grid engineering cost risk early warning method based on sequence relation analysis | |
CN113642162A (en) | Simulation comprehensive analysis method for urban road traffic emergency plan | |
CN110826779A (en) | Road network level bridge tunnel maintenance method, device and storage medium | |
CN114936786B (en) | Comprehensive efficiency evaluation method of road traffic energy source consistent system | |
Hanin et al. | Implementation Of Ahp-Topsis As A Support For Making Decisions On Micro Business Funding In Sambas Regency | |
CN115034648B (en) | Bridge engineering risk assessment method based on BP neural network under condition of few samples | |
CN117688657B (en) | Asphalt pavement durable service risk assessment method under influence of temperature in climate change | |
CN117951596A (en) | Intelligent decision method and system for pavement maintenance, controller and control method | |
JP4247160B2 (en) | Support priority decision support system and method for existing bridges |
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 |