CN109902344A - Short/Medium Span Bridge group structure performance prediction apparatus and system - Google Patents
Short/Medium Span Bridge group structure performance prediction apparatus and system Download PDFInfo
- Publication number
- CN109902344A CN109902344A CN201910061475.4A CN201910061475A CN109902344A CN 109902344 A CN109902344 A CN 109902344A CN 201910061475 A CN201910061475 A CN 201910061475A CN 109902344 A CN109902344 A CN 109902344A
- Authority
- CN
- China
- Prior art keywords
- bridge
- performance prediction
- short
- neural network
- bridge group
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction systems, the structural behaviour of Bridge Group for forming to the bridge by multiple Mid and minor spans is predicted, it is characterised by comprising: examining report storage unit, is stored with the examining report over the years of each bridge；Database sharing portion constructs the relational database of Bridge Group according to examining report over the years；Model construction portion constructs neural network model according to relational database and is trained and examines, which is used to predict the Deterioration of Structural Performance of Bridge Group；And performance prediction portion, the performance parameter of each bridge is obtained to predict the overall structure of Bridge Group and the performance change trend of partial component according to prediction neural network model.
Description
Technical field
Bridge security field of the present invention more particularly, to a kind of Short/Medium Span Bridge group structure performance prediction device and is
System.
Background technique
China is the country that highway bridge quantity is most in the world.According to Department of Transportation's statistical information, by the end of 2016
Year end, China have highway bridge 80.53 ten thousand, 4916.97 ten thousand linear meter(lin.m.) of cumulative length.And with the increasing of bridge Years Of Service
Add, large quantities of newly building bridges just progress into " aging " stage, and various forms of structure degradations inevitably occur.It can
See, it is very urgent for promoting the management and maintenance of all kinds of servicing bridges to work.However, for specified traffic network, although
Have accumulated the valuable material for largely containing valuable structural information in long-term structure inspection, but lack always corresponding means by its
It makes full use of, causes data disaster.On the other hand, existing Bridge Management & Maintenance method is only implemented in the level of monomer bridge, and
Do not planned as a whole in road network level, ignore many general character of the bridge structure in the same area, thus significantly reduces pipe
Support the efficiency of work.
The Structural Behavior Evaluation of Short/Medium Span Bridge group once endured the puzzlement of problems to the fullest extent in engineering practice.For example, building
The assessment models of vertical Short/Medium Span Bridge group need huge data volume as support.Therefore it needs to existing history bridge machinery
Data and each road section traffic volume Flow Observation record carry out data mining, extract interested, valuable information by data integration,
Data cleansing, data conversion obtain relational database.Meanwhile how based on the database simulation bridge performance degradation trend and its
Complex nonlinear and logical relation between every basic parameter are a large orders.Machine learning side neural network based
Method there is provided herein a practical and effective approach, and reasonable model can support for the pipe of all bridges in traffic network to be planned as a whole
Service.
Summary of the invention
To solve the above problems, the future developing trend of existing highway bridge can be effectively predicted based on neural network model,
And suggestion about maintenance program is provided, present invention employs following technical solutions:
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction devices, for by multiple Mid and minor spans
The structural behaviour of the Bridge Group of bridge composition is predicted characterized by comprising examining report storage unit is stored with each
The examining report over the years of bridge；Database sharing portion constructs the relational database of Bridge Group according to examining report over the years；Model is deposited
Storage portion is stored with and completes the neural network model that training is examined according to relational database, and the neural network model is for predicting bridge
The Deterioration of Structural Performance of Liang Qun；Scheme storage unit, there are many scheduled maintenance schemes for repairing bridge for storage；And performance is pre-
Survey portion is become using the performance change that neural network model carries out the overall structure for calculating to predict Bridge Group and partial component
Gesture.
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction devices, can also have the feature that,
Wherein, database sharing portion includes: data extracting unit, and the original assessment number of each bridge is extracted from examining report over the years
According to original assessment packet includes technology status scoring, bridge age, structure type, the volume of traffic and the maintenance behavior of each bridge；Clearly
Rule storage unit is washed, the data cleansing rule for being cleaned to original assessment data is stored with；Data cleansing unit, root
Cleaning is carried out to obtain forecast assessment data to original assessment data according to data cleansing rule；Data processing unit, to prediction
Assessment data are handled and are configured to a relational database based on forecast assessment data for property set.
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction devices, can also have the feature that,
Wherein, property set includes maintenance behavior property, structure type attribute, bridge age attribute, volume of traffic attribute and the technology shape of bridge
Condition scoring attribute, data processing unit include: maintenance behavior processing subelement, carry out binary transform to maintenance behavior property, if
On the contrary bridge is repaired, then the value of corresponding maintenance behavior property is set as 1, then be 0；Structure type processing is single
Member carries out vectorized process to structure type attribute；
Bridge age handles subelement, carries out normalizing transformation to bridge age attribute:
A' is normalizing transformed bridge age, and a is the bridge age before normalizing transformation, amaxFor the maximum value in bridge age；
The volume of traffic handles subelement, carries out normalizing transformation to volume of traffic attribute:
β ' is the transformed volume of traffic of normalizing, and β is the volume of traffic before normalizing transformation, βmaxFor the maximum value of the volume of traffic；
Technology status scoring processing subelement carries out normalizing transformation to technology status scoring attribute:
C' is the transformed technology status scoring of normalizing, and c is the technology status scoring before normalizing transformation.
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction devices, can also have the feature that,
Wherein, original neural network model is more hidden layer BP network models, including input layer and output layer, and input layer has 6
A input layer, input neuron in 3 it is corresponding with structure type, remaining 3 respectively with bridge age, the volume of traffic and
Maintenance behavior is corresponding, and output layer has 1 output layer neuron, and output neuron and the scoring of annual technology status are opposite
It answers, input layer and output layer neuron define connection in the interlayer of original neural network model, while in original mind
There is no connections in layer through network model.
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction devices, can also have the feature that,
Wherein, the e-learning rate of original neural network model is initialized as 0.1.
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction devices, can also have the feature that,
Wherein, original neural network model further includes hidden layer, and 20 hidden layer neurons are set in hidden layer.
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction devices, can also have the feature that,
Wherein, the loss function of original neural network model is defined as the mean square deviation between the predicted value and true value of model output.
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction devices, can also have the feature that,
Wherein, performance prediction portion includes: bridge record generation unit, according to the increase in bridge age for not from the last time to be predicted
Same bridge age generates corresponding bridge record；Maintenance behavior value unit obtains in bridge record according to scheduled maintenance scheme
The value of maintenance behavior；Bridge records fills unit, and structure type and the volume of traffic are filled respectively into each bridge record；Skill
Art situation scoring output unit, records the input as neural network model for bridge, so that output is under scheduled maintenance scheme
The Predicting Technique situation in each year scores.
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction devices, can also have the feature that,
Wherein, the volume of traffic is annual average daily traffic.
The present invention provides a kind of Short/Medium Span Bridge group structure performance prediction systems characterized by comprising performance is pre-
Device is surveyed, the structural behaviour of the Bridge Group for forming to the bridge by multiple Mid and minor spans is predicted；Storage server is deposited
Contain the examining report over the years of all bridges；And examining report acquisition device, for acquiring Bridge Group from storage server
In all bridges examining report over the years and be sent to performance prediction device, wherein performance prediction device is aforementioned present invention
Short/Medium Span Bridge group structure performance prediction device.
Invention action and effect
Short/Medium Span Bridge group structure performance prediction device according to the present invention, due to database sharing portion,
The region bridge machinery report accumulated throughout the year can be carried out the extraction of information, data it is integrated and regular, by structural parameters and
Its performance degradation trend is converted into relational database.Since training artificial neural network can be passed through with model storage unit
Network model, the relationship between simulation bridge bridge age, type, the volume of traffic, each parameter of maintenance behavior and configuration state.Due to performance
Prediction section, therefore the performance change of following region bridge can be further predicted based on mature neural network model
And degradation trend, it is verified by example of calculation, obtained Short/Medium Span Bridge group structure performance prediction result has very high reality
With property, effective decision support is provided for the management and maintenance of Short/Medium Span Bridge group.
To sum up, Short/Medium Span Bridge group structure performance prediction device of the invention combines data mining technology, to all the year round
Magnanimity detection data under bridge inspection work accumulation effectively, adequately utilize, and establishes neural network model, will extract
Data be converted into the valuable knowledge in Bridge Management & Maintenance field, realize the Bridge performance assessment prediction of Short/Medium Span Bridge group
Guidance is supported with pipe.
Detailed description of the invention
Fig. 1 is the structural block diagram of Short/Medium Span Bridge group structure performance prediction device of the invention；
Fig. 2 is the structural block diagram in data building portion of the invention；
Fig. 3 is the information exchange schematic diagram between relational database and neural network model of the invention；
Fig. 4 is the structural schematic diagram of neural network of the invention；
Fig. 5 is the prediction result contrast schematic diagram of bridge of the invention under different maintenance programs；
Fig. 6 is the work flow diagram of Short/Medium Span Bridge group structure performance prediction device of the invention；
Fig. 7 is the structural block diagram of Short/Medium Span Bridge group structure performance prediction system of the invention.
Specific embodiment
In order to be easy to understand the technical means, the creative features, the aims and the efficiencies achieved by the present invention, tie below
Attached drawing is closed to be specifically addressed beam bridge safety monitoring of the invention and assessment device.
<embodiment 1>
Fig. 1 is the structural block diagram of Short/Medium Span Bridge group structure performance prediction device of the invention.
As shown in Figure 1, Short/Medium Span Bridge group structure performance prediction device 100 of the invention, for the target network of communication lines
The structural behaviour for the Bridge Group being made of in network the bridge of multiple Mid and minor spans predicted, including examining report storage unit 10,
Database sharing portion 20, model storage unit 30, scheme storage unit 40, performance prediction portion 50, pipe, which are supported, suggests that acquisition unit 60, picture are deposited
Storage portion 70 and input display unit 80.
Examining report storage unit 10 is stored with the examining report over the years of each bridge.
Fig. 2 is the structural block diagram in data building portion of the invention.
As shown in Fig. 2, database sharing portion 20 is used to construct the relational database of Bridge Group, packet according to examining report over the years
Include data extracting unit 201, cleaning rule storage unit 202, data cleansing unit 203 and data processing unit 204.
Data extracting unit 201 is used to extract the original assessment data of each bridge from examining report over the years, original
Assessment packet includes technology status scoring, bridge age, structure type, the volume of traffic and the maintenance behavior of each bridge.It carries out special
Bridge age, the structure type, annual average daily traffic, maintenance behavior, annual technology status scoring field of bridge are screened in sign selection
As the property set of relational database, data extraction is carried out to every part of report accordingly and is integrated.
Cleaning rule storage unit 202 is stored with the data cleansing rule for being cleaned to original assessment data.
Data cleansing unit 203 is used to carry out cleaning to original assessment data according to data cleansing rule to be predicted
Assess data.Missing will be present, the data of mistake phenomenon are deleted.Specially according to data cleansing rule, for data
A certain item record in library is deleted, it is ensured that the validity of data if its any attribute value has missing, mistake phenomenon
And availability.For example, the bridge age attribute value of a record is -1 or blank, then whole record is left out.
It is category that data processing unit 204, which handles forecast assessment data and is configured to one based on forecast assessment data,
The relational database of property collection.
Property set includes maintenance behavior property, structure type attribute, bridge age attribute, volume of traffic attribute and the technology of bridge
Situation scoring attribute,
Data processing unit 204 includes: maintenance behavior processing subelement 204a, structure type processing subelement 204b, bridge
Age handles subelement 204c, volume of traffic processing subelement 204d, technology status scoring processing subelement 204e.
Maintenance behavior handles subelement 204a and is used to carry out binary transform to maintenance behavior property, if bridge is tieed up
It repairs, then the value of corresponding maintenance behavior property is set as 1, it is on the contrary then be 0.
Structure type handles subelement 204b and is used to carry out vectorized process to structure type attribute, will be in relational database
All bridges are divided into three classes, i.e., plate girder bridge, box girder bridge, other, be respectively converted into (1,0,0), (0,1,0), (0,0,1).
Bridge age handles subelement 204c and is used to carry out normalizing transformation to bridge age attribute, its calculation formula is:
Wherein, a' is normalizing transformed bridge age, and a is the bridge age before normalizing transformation, amaxFor the maximum value in bridge age.
The volume of traffic handles subelement 204d and is used to carry out normalizing transformation to volume of traffic attribute, its calculation formula is:
Wherein, β ' is the transformed volume of traffic of normalizing, and β is the volume of traffic before normalizing transformation, βmaxFor the maximum of the volume of traffic
Value.
Technology status scoring processing subelement 204e is used to carry out normalizing transformation to technology status scoring attribute, calculates public
Formula are as follows:
Wherein, c' is the transformed technology status scoring of normalizing, and c is the technology status scoring before normalizing transformation.
Fig. 3 is the information exchange schematic diagram between relational database and neural network model of the invention, and Fig. 4 is of the invention
The structural schematic diagram of neural network.
As shown in Fig. 3~Fig. 4, model storage unit 30, which is stored with, completes the neural network that training is examined according to relational database
Model.
Neural network model is used to predict the Deterioration of Structural Performance of Bridge Group, is more hidden layer BP network models, packet
Include input layer, output layer and hidden layer.
Input layer has 6 input layers, and 3 inputted in neuron are corresponding with structure type (to be respectively corresponded
Plate girder bridge, box girder bridge and other), remaining 3 respectively with bridge age, the volume of traffic and maintenance behavior it is corresponding.
Output layer has 1 output layer neuron, and output neuron is corresponding with annual technology status scoring.
Input layer and output layer neuron define connection in the interlayer of neural network model, and its is corresponding
Weight coefficient is initialized by 0~1 section stochastical sampling, while there is no connections in the layer of neural network model.
The quantity of hidden layer is 1 layer, is set with 20 hidden layer neurons.
In the present embodiment, the e-learning rate of neural network model is initialized as 0.1, the training side of neural network model
Method specifically:
The loss function of neural network model is defined as to the predicted value y of model outputi' with true value yiBetween it is square
Difference, model output are the technology status scoring after normalization, and data are used as in model if the n item record in database is imported
Collection, the then calculation formula lost areTraining is iterated to the network using BP algorithm again, until accidentally
Poor size is lower than preset value.
There are many scheduled maintenance schemes of bridge for the storage of scheme storage unit 40, including taking for each maintenance program
With.
Fig. 5 is the prediction result contrast schematic diagram of bridge of the invention under different maintenance programs.
As shown in figure 5, performance prediction portion 50 utilizes neural network model using attribute centrality energy parameter as input data
The skill score situation of corresponding bridge is exported, thus predict the overall structure of Bridge Group and the performance change trend of partial component,
It is commented including bridge record generation unit 501, maintenance behavior value unit 502, bridge record fills unit 503 and technology status
Divide output unit 504.
Bridge record generation unit 501 is used to according to the increase in bridge age be different bridges from the last time to be predicted
Age generates corresponding bridge record.
Maintenance behavior value unit 502 is used to obtain the value for repairing behavior in bridge record according to scheduled maintenance scheme.
Bridge record fills unit 503 is used to fill structure type and the volume of traffic respectively into each bridge record, i.e.,
It include bridge age, maintenance behavior, structure type and the volume of traffic of the bridge under scheduled maintenance scheme in bridge record.
Technology status scores output unit 504 for bridge to be recorded to the input data as neural network model, thus
The technology status scoring in each year under output scheduled maintenance scheme is calculated by model.
Pipe, which is supported, suggests that acquisition unit 60 is used to obtain each bridge in corresponding Bridge Group according to the scoring of the technology status in each year
The pipe of beam, which is supported, suggests.For example, the expense of every kind of scheduled maintenance scheme in association schemes storage unit 40, exists to every kind of maintenance program
The same-cost lower technology status scoring that can be improved is ranked up, and the technology status scoring that can be improved is higher, then corresponding
Maintenance program is then better, and as optimal pipe, which is supported, suggests.
Picture storage unit 70 is stored with the display picture, the display picture of forecast assessment data, performance of original assessment data
Prediction display picture, pipe support suggest display picture, the setting screen of neural network model, data cleansing rule setting picture
The setting screen of face and default maintenance program.
Input display unit 80 is the liquid crystal display with touch screen functionality, for showing the display picture of original assessment data
Face, the display picture of forecast assessment data, the display picture of performance prediction and pipe support the display picture suggested, input display unit
80 can be used for display allow user's setting model parameter neural network model setting screen, setting data cleansing rule
Setting screen and the setting screen for setting default maintenance program.
Fig. 6 is the work flow diagram of Short/Medium Span Bridge group structure performance prediction device of the invention.
Before the structural behaviour to Bridge Group is predicted, the target transportation network region of prediction needed for user first determines
And in selection region required prediction bridge as Bridge Group, regather, summarize all bridges in Bridge Group detection over the years report
It accuses, is then stored into examining report storage unit 10.
By input display unit 80 neural network model setting screen allow user set neural network model mould
Shape parameter, the setting screen for setting data cleansing rule allow user to set data cleansing rule, set default maintenance program
Setting screen allows user to set maintenance program and is stored into scheme storage unit 40.
Below in conjunction with the workflow of the Short/Medium Span Bridge group structure performance prediction device 100 of Detailed description of the invention the present embodiment
Journey, step specific as follows:
Step S1, bridge age, type, the annual day that database sharing portion 20 extracts each bridge in examining report over the years hand over
Flux, maintenance behavior and the scoring of the technology status in each year, construct relational database；
Step S2 establishes neural network model based on the data in relational database and is trained and examines, and will
It completes the neural network model that training is examined and is stored into model storage unit 30；
Step S3, performance prediction portion 50 using attribute centrality energy parameter as input data, bridge respectively to be predicted
Structure type, annual traffic, bridge age and maintenance behavior, the skill score shape of corresponding bridge is exported using neural network model
Condition, to predict the overall structure of Bridge Group and the performance change trend of partial component；
Step S4 proposes each bridge in corresponding Bridge Group according to the skill score situation of the bridge of model output
Pipe, which is supported, to be suggested and is shown on the feeding display picture suggested of pipe.
<embodiment 2>
Fig. 7 is the structural block diagram of Short/Medium Span Bridge group structure performance prediction system of the invention.
As shown in fig. 7, including embodiment 1 in Short/Medium Span Bridge group structure performance prediction system 500 in the present embodiment 2
In Short/Medium Span Bridge group structure performance prediction device 100, examining report acquisition device 300 and storage server 400.In
Small across footpath Bridge Group prediction of performance of structures device 100, examining report acquisition device 300 and storage server 400 pass through nothing
Line communication connection.
Storage server 400 is stored with the design drawing, examining report over the years and maintenance record of all bridges.For example, road
Political situation, investment in transportation group, the municipal examining report of Maintenance Company and the database of maintenance record, the original design figure of designing institute
The database etc. of paper.
Examining report acquisition device 300 is used for from acquiring going through for all bridges in Bridge Group to be measured in storage server 400
Year examining report is simultaneously sent to Short/Medium Span Bridge group structure performance prediction device 100.
Before the structural behaviour to Bridge Group is predicted, the target transportation network region of prediction needed for user first determines
And in selection region required prediction bridge as Bridge Group, examining report acquisition device 300 is received from storage server 400
The examining report over the years for collecting, summarizing all bridges in Bridge Group is then sent to Short/Medium Span Bridge group structure performance prediction dress
It sets 100 and is stored into examining report storage unit 10.
Embodiment action and effect
According to the Short/Medium Span Bridge group structure performance prediction device of the present embodiment, due to database sharing portion, because
This can report the region bridge machinery accumulated throughout the year the integrated and regular of the extraction, data for carrying out information, by structural parameters
And its performance degradation trend is converted into relational database.Since training artificial neuron can be passed through with model storage unit
Network model, the relationship between simulation bridge bridge age, type, the volume of traffic, each parameter of maintenance behavior and configuration state.Due to property
Energy prediction section, therefore based on mature neural network model, can further predict the performance change of following region bridge
Change and degradation trend, are verified by example of calculation, and obtained Short/Medium Span Bridge group structure performance prediction result has very high
Practicability provides effective decision support for the management and maintenance of Short/Medium Span Bridge group.
To sum up, Short/Medium Span Bridge group structure performance prediction device of the invention combines data mining technology, to all the year round
Magnanimity detection data under bridge inspection work accumulation effectively, adequately utilize, and establishes neural network model, will extract
Data be converted into the valuable knowledge in Bridge Management & Maintenance field, realize the Bridge performance assessment prediction of Short/Medium Span Bridge group
Guidance is supported with pipe.
Since database sharing portion includes data extracting unit, cleaning rule storage unit, data cleansing unit and number
According to processing unit, therefore it can get rid of in initial data and there is missing, the data of mistake phenomenon, so that neural network model
Calculated result is more accurate.
Since data processing unit includes maintenance behavior processing subelement, structure type processing subelement, bridge age processing
Unit, volume of traffic processing subelement, technology status scoring processing subelement, therefore each attribute in property set can be distinguished
Processing calculating is carried out, so that the data in relational database are more accurate, the computational efficiency of neural network model is higher.
Due to using more hidden layer BP network models, for mature effective computation model, including input layer, output layer
And hidden layer, therefore, the technology status of the bridge exported by the model score it is more accurate, to the structural of Bridge Group
It is also more accurate to predict.
Due to Short/Medium Span Bridge group structure performance prediction system of the invention further include examining report acquisition device and
Storage server, testing staff by wireless communication can be from acquiring going through for all bridges in Bridge Group to be measured in storage server
Year examining report is simultaneously sent to performance prediction device, therefore, so that acquisition of the testing staff for the examining report over the years of bridge
It is more accurate, comprehensive with data mining, further improve the accuracy and working efficiency of Bridge Group prediction result.
The preferred embodiment of the present invention has been described in detail above.It should be appreciated that the ordinary skill of this field is without wound
The property made labour, which according to the present invention can conceive, makes many modifications and variations.Therefore, all technician in the art
Pass through the available technology of logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Scheme, all should be within the scope of protection determined by the claims.
For example, being stored with scheduled maintenance program in scheme storage unit in embodiment, calculated by neural network model
The technology status scoring of each bridge under different maintenance programs is obtained, to obtain to develop skill under identical expense
Situation scores highest maintenance program as optimal pipe and supports suggestion, in Short/Medium Span Bridge group structure performance provided by the invention
In prediction meanss, scheme storage unit can also be stored with evaluation of maintenance cost corresponding with each maintenance program point, the maintenance at
This assessment point is that maintenance personal comments what the factors such as human and material resources spent by maintenance program and time were assessed
Point, it is higher to expend smaller scoring.Short/Medium Span Bridge group structure performance prediction device also has sequence portion, which has comprehensive
Close scoring computing unit and sequencing unit, comprehensive score computing unit for the human and material resources according to spent by maintenance program with
And maintenance program is weighted in time factor, to obtain the scoring of maintenance program, sequencing unit is used for maintenance side
Case is ranked up, and sort by is successively to be sorted according to the height of scoring.Further, setting weighting is stored in picture storage unit
The setting screen of value, input display unit are used to show that the setting screen of setting weighted value to allow user to set weighted value.
Claims (10)
1. a kind of Short/Medium Span Bridge group structure performance prediction device, the bridge for being formed to the bridge by multiple Mid and minor spans
The structural behaviour of group is predicted characterized by comprising
Examining report storage unit is stored with the examining report over the years of each bridge；
Database sharing portion constructs the relational database of the Bridge Group according to the examining report over the years；
Model storage unit is stored with and completes the neural network model that training is examined, the neural network according to the relational database
Model is used to predict the Deterioration of Structural Performance of the Bridge Group；
Scheme storage unit, there are many scheduled maintenance schemes for repairing the bridge for storage；And
Performance prediction portion carries out overall structure and the part for calculating to predict the Bridge Group using the neural network model
The performance change trend of component.
2. Short/Medium Span Bridge group structure performance prediction device according to claim 1, it is characterised in that:
Wherein, the database sharing portion includes:
Data extracting unit extracts the original assessment data of each bridge, the original from the examining report over the years
Beginning assessment packet includes technology status scoring, bridge age, structure type, the volume of traffic and the maintenance behavior of each bridge；
Cleaning rule storage unit is stored with the data cleansing rule for being cleaned to the original assessment data；
Data cleansing unit carries out cleaning to the original assessment data according to the data cleansing rule to obtain pre- assessment
Estimate data；
Data processing unit, handles the forecast assessment data and is configured to one based on the forecast assessment data and be
The relational database of property set.
3. Short/Medium Span Bridge group structure performance prediction device according to claim 2, it is characterised in that:
Wherein, the property set includes the maintenance behavior property, structure type attribute, bridge age attribute, volume of traffic category of the bridge
Property and technology status score attribute,
The data processing unit includes:
Maintenance behavior handles subelement, carries out binary transform to the maintenance behavior property, if the bridge is repaired,
The value of the corresponding maintenance behavior property is set as 1, it is on the contrary then be 0；
Structure type handles subelement, carries out vectorized process to the structure type attribute；
Bridge age handles subelement, carries out normalizing transformation, calculation formula to the bridge age attribute are as follows:
A' is normalizing transformed bridge age, and a is the bridge age before normalizing transformation, amaxFor the maximum value in bridge age；
The volume of traffic handles subelement, carries out normalizing transformation, calculation formula to the volume of traffic attribute are as follows:
β ' is the transformed volume of traffic of normalizing, and β is the volume of traffic before normalizing transformation, βmaxFor the maximum value of the volume of traffic；
Technology status scoring processing subelement carries out normalizing transformation, calculation formula to technology status scoring attribute are as follows:
C' is the transformed technology status scoring of normalizing, and c is the technology status scoring before normalizing transformation.
4. Short/Medium Span Bridge group structure performance prediction device according to claim 2, it is characterised in that:
Wherein, the neural network model is more hidden layer BP network models, including input layer and output layer,
The input layer has 6 input layers, and 3 in the input neuron are corresponding with the structure type,
Remaining 3 are corresponding with the bridge age, the volume of traffic and the maintenance behavior respectively,
The output layer has 1 output layer neuron, and the output neuron and the scoring of the annual technology status are opposite
It answers,
The input layer and the output layer neuron define connection in the interlayer of the neural network model, simultaneously
There is no connections in the layer of the neural network model.
5. Short/Medium Span Bridge group structure performance prediction device according to claim 4, it is characterised in that:
Wherein, the e-learning rate of the neural network model is initialized as 0.1.
6. Short/Medium Span Bridge group structure performance prediction device according to claim 4, it is characterised in that:
Wherein, the neural network model further includes hidden layer, and 20 hidden layer neurons are set in the hidden layer.
7. Short/Medium Span Bridge group structure performance prediction device according to claim 4, it is characterised in that:
Wherein, the loss function of the neural network model is defined as square between the predicted value and true value of model output
Difference.
8. Short/Medium Span Bridge group structure performance prediction device according to claim 2, it is characterised in that:
Wherein, the performance prediction portion includes:
Bridge record generation unit is generated according to the increase in the bridge age from the last time to be predicted for different bridge ages
Corresponding bridge record；
Maintenance behavior value unit obtains taking for maintenance behavior described in the bridge record according to the scheduled maintenance scheme
Value；
Bridge records fills unit, and the structure type and the volume of traffic are filled respectively into each bridge record；
Technology status scoring output unit, records the input as the neural network model for the bridge, so that output exists
The Predicting Technique situation scoring in each year under the scheduled maintenance scheme.
9. Short/Medium Span Bridge group structure performance prediction device according to claim 8, it is characterised in that:
Wherein, the volume of traffic is annual average daily traffic.
10. a kind of Short/Medium Span Bridge group structure performance prediction system characterized by comprising
The structural behaviour of performance prediction device, the Bridge Group for forming to the bridge by multiple Mid and minor spans is predicted；
Storage server is stored with the examining report over the years of all bridges；And
Examining report acquisition device, for from acquiring the over the years of all bridges in the Bridge Group in the storage server
Examining report is simultaneously sent to the performance prediction device,
Wherein, the performance prediction device is Short/Medium Span Bridge group structure described in any one of claim 1~9
It can prediction meanss.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910061475.4A CN109902344B (en) | 2019-01-23 | 2019-01-23 | Device and system for predicting structural performance of middle-small span bridge group |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910061475.4A CN109902344B (en) | 2019-01-23 | 2019-01-23 | Device and system for predicting structural performance of middle-small span bridge group |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109902344A true CN109902344A (en) | 2019-06-18 |
| CN109902344B CN109902344B (en) | 2023-06-27 |
Family
ID=66944011
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910061475.4A Active CN109902344B (en) | 2019-01-23 | 2019-01-23 | Device and system for predicting structural performance of middle-small span bridge group |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109902344B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111242347A (en) * | 2019-12-28 | 2020-06-05 | 浙江大学 | Bridge management and maintenance aid decision-making system based on historical weight updating |
| CN111738996A (en) * | 2020-06-09 | 2020-10-02 | 交通运输部公路科学研究所 | Bridge health monitoring and early warning system based on machine learning |
| CN113469374A (en) * | 2021-09-02 | 2021-10-01 | 北京易真学思教育科技有限公司 | Data prediction method, device, equipment and medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106017556A (en) * | 2016-05-20 | 2016-10-12 | 清华大学合肥公共安全研究院 | Bridge safety operation monitoring and management system |
| CN108629144A (en) * | 2018-06-11 | 2018-10-09 | 湖北交投智能检测股份有限公司 | A kind of bridge health appraisal procedure |
-
2019
- 2019-01-23 CN CN201910061475.4A patent/CN109902344B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106017556A (en) * | 2016-05-20 | 2016-10-12 | 清华大学合肥公共安全研究院 | Bridge safety operation monitoring and management system |
| CN108629144A (en) * | 2018-06-11 | 2018-10-09 | 湖北交投智能检测股份有限公司 | A kind of bridge health appraisal procedure |
Non-Patent Citations (4)
| Title |
|---|
| 向木生等: "基于BP神经网络的桥梁技术状态评估", 《交通科技》 * |
| 时笑鹏: "中小跨径混凝土梁桥退化预测和维护策略研究", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
| 李晓辉: "寒区桥梁养护管理***研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 赵进: "公路桥梁养护决策与管理***研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111242347A (en) * | 2019-12-28 | 2020-06-05 | 浙江大学 | Bridge management and maintenance aid decision-making system based on historical weight updating |
| CN111738996A (en) * | 2020-06-09 | 2020-10-02 | 交通运输部公路科学研究所 | Bridge health monitoring and early warning system based on machine learning |
| CN111738996B (en) * | 2020-06-09 | 2023-04-07 | 交通运输部公路科学研究所 | Bridge health monitoring and early warning system based on machine learning |
| CN113469374A (en) * | 2021-09-02 | 2021-10-01 | 北京易真学思教育科技有限公司 | Data prediction method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109902344B (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Biancofiore et al. | Recursive neural network model for analysis and forecast of PM10 and PM2. 5 | |
| Zhang et al. | A feature selection and multi-model fusion-based approach of predicting air quality | |
| Gao et al. | Digital twin and its application in transportation infrastructure | |
| CN106022518B (en) | A kind of piping failure probability forecasting method based on BP neural network | |
| CN104700321B (en) | A kind of power transmission and transformation equipment state operation trend analysis method | |
| CN108172301A (en) | A kind of mosquito matchmaker's epidemic Forecasting Methodology and system based on gradient boosted tree | |
| CN106779457A (en) | A kind of rating business credit method and system | |
| CN105825342A (en) | Pipeline failure possibility evaluation method and system | |
| CN109902344A (en) | Short/Medium Span Bridge group structure performance prediction apparatus and system | |
| CN109614669A (en) | Net grade Bridge performance assessment prediction method | |
| CN110111576A (en) | A kind of urban transportation elastic index and its implementation based on space-time congestion group | |
| CN104798043A (en) | Data processing method and computer system | |
| Van et al. | Research trends on machine learning in construction management: a scientometric analysis | |
| CN110162445A (en) | The host health assessment method and device of Intrusion Detection based on host log and performance indicator | |
| CN105631612A (en) | System and method of evaluating individual performance and capability of public servant based on big data | |
| CN116739368A (en) | Industrial park carbon emission level monitoring and evaluating method based on energy big data | |
| CN111199298A (en) | Flood forecasting method and system based on neural network | |
| Ye et al. | A new environmental governance cost prediction method based on indicator synthesis and different risk coefficients | |
| KR102359902B1 (en) | Crossroads LOS Prediction Method Based on Big Data and AI, and Storage Medium Having the Same | |
| Whipple | Predictive storm damage modeling and optimizing crew response to improve storm response operations | |
| TW201810133A (en) | Neural networks based water level predicting system and method for reservoirs | |
| CN116662860A (en) | User portrait and classification method based on energy big data | |
| CN115660645A (en) | Method, device, equipment and storage medium for evaluating skills of maintenance personnel | |
| CN108429771A (en) | Software defined network secure state evaluating method based on catastrophe theory and device | |
| CN114897262A (en) | Rail transit equipment fault prediction method based on deep learning |
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 |