CN110889155B - Steel bridge deck corrosion prediction model and construction method - Google Patents
Steel bridge deck corrosion prediction model and construction method Download PDFInfo
- Publication number
- CN110889155B CN110889155B CN201911081350.4A CN201911081350A CN110889155B CN 110889155 B CN110889155 B CN 110889155B CN 201911081350 A CN201911081350 A CN 201911081350A CN 110889155 B CN110889155 B CN 110889155B
- Authority
- CN
- China
- Prior art keywords
- corrosion
- bridge deck
- steel bridge
- steel
- neural network
- 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
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 119
- 239000010959 steel Substances 0.000 title claims abstract description 119
- 238000005260 corrosion Methods 0.000 title claims abstract description 107
- 230000007797 corrosion Effects 0.000 title claims abstract description 106
- 238000010276 construction Methods 0.000 title claims abstract description 18
- 238000013528 artificial neural network Methods 0.000 claims abstract description 43
- 230000002068 genetic effect Effects 0.000 claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 15
- 238000012360 testing method Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 14
- 239000000203 mixture Substances 0.000 claims description 13
- 239000010426 asphalt Substances 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000002844 melting Methods 0.000 abstract description 9
- 230000008018 melting Effects 0.000 abstract description 7
- 239000011384 asphalt concrete Substances 0.000 abstract description 6
- 239000000463 material Substances 0.000 abstract description 4
- 238000005266 casting Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 239000003795 chemical substances by application Substances 0.000 description 2
- 230000005684 electric field Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000004020 conductor Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000003487 electrochemical reaction Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000003892 spreading Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Genetics & Genomics (AREA)
- Physiology (AREA)
- Testing Resistance To Weather, Investigating Materials By Mechanical Methods (AREA)
- Bridges Or Land Bridges (AREA)
Abstract
The invention provides a steel bridge deck corrosion prediction model and a construction method thereof, which are characterized in that the method takes temperature, humidity, power-on times and power-on time as steel bridge deck corrosion influencing factors, adopts a genetic algorithm to optimize an extreme learning machine neural network to obtain an optimized neural network, and trains the optimized neural network to obtain the steel bridge deck corrosion prediction model. The pre-estimated model has the characteristics of high learning speed, wide application range and high prediction precision. The invention applies the extreme learning machine neural network prediction model optimized by the genetic algorithm to the field of steel bridge deck pavement for the first time, predicts the corrosion degree of the steel bridge deck based on the relevant data such as the temperature, the humidity, the power-on time, the power-on times and the like of the steel bridge deck pouring type conductive asphalt concrete, and provides a new thought for selecting materials and making corrosion-proof measures of the pouring type conductive asphalt concrete steel plates for melting snow and deicing.
Description
Technical Field
The invention belongs to the field of engineering materials, relates to a casting type conductive asphalt mixture, and in particular relates to a steel bridge deck corrosion prediction model and a construction method.
Background
Although the construction mileage of the Chinese highway is the first place in the world, the capability of the highway traffic facilities to cope with severe weather is insufficient, and the technology is limited. Especially, the snowfall and the icing in cold seasons seriously reduce the skid resistance of road (bridge) surfaces, so that traffic accidents frequently occur, and the existing service capacity and the social and economic development of a highway system are restricted. The traditional passive methods such as snow-melting agent spreading or mechanical snow removal have great influence on traffic, the snow-melting agent pollutes water resources and soil, the mechanical snow removal efficiency is low, road (bridge) surfaces are damaged, and the like. In recent years, the casting asphalt mixture paving layer becomes the first choice of bridge deck paving materials by virtue of excellent performance, and the casting conductive asphalt mixture is a research hot spot of the current bridge deck snow melting and deicing technology. Namely, the pouring type conductive asphalt mixture is prepared by mixing proper type and mixing amount of conductive materials into the pouring type asphalt mixture. The device can not only realize timely and efficient snow and ice melting of the bridge deck and powerfully ensure smooth road and safe driving, but also avoid potential safety hazards such as electrode damage and short circuit caused by water and electricity contact in construction rolling.
The casting type conductive asphalt mixture can realize timely and efficient snow and ice melting of the bridge deck. However, the current formed after the steel bridge surface is electrified generates a micro electric field/magnetic field, the potential around the steel plate is changed, a potential difference is formed between the current and the surrounding environment, the corrosion current is enhanced, the corrosion speed of the steel bar is accelerated, meanwhile, the strength of the magnetic field of the electric field gradually weakens outwards from the surface of the steel plate, and electrons migrate towards the direction of electromagnetic intensity, so that more electrons are accumulated on the surface of the steel plate, and the electrochemical reaction is accelerated. The different types of casting conductive asphalt mixtures have corrosion to the steel plate to a certain extent, damage the original bridge deck structure and influence the traffic running environment and safety.
The corrosion degree of the steel bridge deck is commonly acted by a plurality of factors such as the material, the stress state, the environment and the like of the steel plate. The corrosion of the steel plate is affected by a plurality of factors, the working environments of the bridge are different, such as a humid environment, a dry environment, a polluted environment and a non-polluted environment, and the corrosion degree of the steel plate is different. Corrosion can cause uneven stress of the bridge deck structure, stress concentration occurs, so that bridge panels are damaged to different degrees, the service life of paving the bridge deck is further shortened, and the driving safety of the bridge deck is seriously influenced.
Disclosure of Invention
Aiming at the defects and the shortcomings of the prior art, the invention aims to provide a steel bridge deck corrosion prediction model and a construction method thereof, which solve the technical problem of low steel bridge deck corrosion prediction accuracy in the prior art.
In order to solve the technical problems, the invention adopts the following technical scheme:
the construction method is characterized in that temperature, humidity, power-on times and power-on time are used as steel bridge deck corrosion influence factors, a genetic algorithm is adopted to optimize an extreme learning machine neural network, an optimized neural network is obtained, and the optimized neural network is trained to obtain the steel bridge deck corrosion prediction model.
Specifically, the construction method comprises the following steps:
step one, collecting corrosion data of a steel bridge deck plate:
selecting steel bars in different types of pouring type conductive asphalt mixture test pieces as data acquisition points, placing the combined structure into an environment control box after selecting corrosion data acquisition points, setting temperature, humidity, power-on times and power-on time, and testing and calculating to obtain corrosion current density of the steel bars after the steel bars are kept in the environment control box for 96 hours;
step two, establishing a steel bridge deck corrosion database:
the steel bar corrosion current density data obtained in the first step is stored in a steel bridge deck corrosion database, and in the steel bridge deck corrosion database, the collected steel bar corrosion current density data is steel bridge deck corrosion sample data;
step three, determining corrosion influence factors of the steel bridge deck plate:
the corrosion influence factors of the steel bridge deck are set as follows: temperature, humidity, number of times of energization, and energization time;
step four, preprocessing steel bridge deck corrosion sample data:
carrying out smooth denoising treatment on the fluctuation and burrs of the sample data caused by noise signals on the steel bridge deck corrosion sample data obtained in the step two;
step five, standardized processing of steel bridge deck corrosion sample data:
in order to reduce prediction misdetection, the mapmin max function in matlab is used for carrying out standardized processing on steel bridge deck corrosion sample data, so that the processed data are distributed in [0,1 ]]In the range of (2), the standardized formula is:wherein X is i Is normalized data, X is sample data, X max For maximum sample data, X min Minimum value of sample data;
step six, optimizing the extreme learning machine neural network by adopting a genetic algorithm to obtain an optimized neural network;
step seven, training a steel bridge deck corrosion estimation model:
taking the corrosion influence factors of the steel bridge deck in the step three as input parameters of the optimized neural network obtained in the step six, taking the number of the input parameters as the number of neurons of an input layer, taking the corrosion current density of the steel bars as output parameters of the optimized neural network obtained in the step six, and training the optimized neural network to obtain a corrosion estimation model of the steel bridge deck;
the invention also has the following technical characteristics:
in the first step, the temperature setting range in the environment control box is-15 ℃ to-5 ℃, the humidity setting range is 40-80%, the power-on frequency setting range is 1-4, and the power-on time setting range is 0.5 h-1.5 h.
In step seven, 400 sets of data are used for training and 100 sets of data are used for testing in every 500 sets of input parameter test data.
The invention also provides a steel bridge deck corrosion prediction model, which is obtained by the construction method of the steel bridge deck corrosion prediction model.
Compared with the prior art, the invention has the following technical effects:
the predictive model training method provided by the invention determines more reasonable steel bridge deck corrosion influence factors, trains the extreme learning machine neural network optimized by adopting the genetic algorithm to obtain the predictive model, and has the characteristics of high learning speed, wide application range and high prediction precision.
The invention applies the extreme learning machine neural network prediction model optimized by the genetic algorithm to the field of steel bridge deck pavement for the first time, predicts the corrosion degree of the steel bridge deck based on the temperature, humidity, energizing time, energizing times and other related data of the steel bridge deck pouring type conductive asphalt concrete, and provides a new idea for the selection of the pouring type conductive asphalt concrete steel plate for melting snow and deicing and the formulation of anti-corrosion measures.
And (III) by combining the characteristic of global optimization of the genetic algorithm, the invention acquires the optimal input weight and hidden node bias of the extreme learning machine, and compared with the traditional neural network, the invention omits the step of optimizing the output weight, improves the operation efficiency of the model, and simultaneously solves the problem of lower prediction precision caused by randomly generating the input weight and hidden node bias of the single extreme learning machine.
The invention can be used for predicting the corrosion of the steel bridge surface of different types of cast conductive asphalt concrete, can monitor the environment and working data of the steel bridge surface for a long time, and can be used as a sample for training a GA-ELM model and predicting the corrosion of a steel bridge deck.
Drawings
FIG. 1 is a schematic diagram of a construction flow of a steel deck corrosion prediction model.
Fig. 2 is a schematic diagram of an algorithm flow of a genetic algorithm optimization extreme learning machine neural network.
FIG. 3 is a schematic diagram of a network structure of a steel deck corrosion prediction model.
FIG. 4 is a schematic diagram of a training process for a steel deck corrosion prediction model.
FIG. 5 is a schematic flow chart of a steel deck corrosion estimation process.
The following examples illustrate the invention in further detail.
Detailed Description
Aiming at the prior art, a model for estimating corrosion of a steel deck plate of a cast conductive asphalt mixture is needed, and the model can estimate corrosion of a steel deck plate from the number of times of electrifying, temperature, humidity and electrifying time of the cast conductive asphalt mixture based on a neural network of an extreme learning machine optimized by a genetic algorithm, so that damage to the steel deck plate is reduced on the basis of ensuring snow melting and ice melting of the bridge deck.
The following specific embodiments of the present invention are provided, and it should be noted that the present invention is not limited to the following specific embodiments, and all equivalent changes made on the basis of the technical solutions of the present application fall within the protection scope of the present invention.
Example 1:
the embodiment provides a construction method of a steel bridge deck corrosion prediction model, as shown in fig. 1, the construction method takes temperature, humidity, power-on times and power-on time as steel bridge deck corrosion influence factors, adopts a genetic algorithm to optimize an extreme learning machine neural network to obtain an optimized neural network, and trains the optimized neural network to obtain the steel bridge deck corrosion prediction model.
Specifically, the construction method comprises the following steps:
step one, collecting corrosion data of a steel bridge deck plate:
selecting steel bars in different types of pouring type conductive asphalt mixture test pieces as data acquisition points, placing the combined structure into an environment control box after selecting corrosion data acquisition points, setting temperature, humidity, power-on times and power-on time, and testing and calculating to obtain corrosion current density of the steel bars after the steel bars are kept in the environment control box for 96 hours;
in the first step, the temperature setting range in the environment control box is-15 ℃ to-5 ℃, the humidity setting range is 40-80%, the power-on frequency setting range is 1-4, and the power-on time setting range is 0.5 h-1.5 h.
Specifically, the relation between the current density and the corrosion of the steel bar is shown in table 1;
TABLE 1 correspondence between reinforcement corrosion current density and corrosion rate
Step two, establishing a steel bridge deck corrosion database:
the steel bar corrosion current density data obtained in the first step is stored in a steel bridge deck corrosion database, and in the steel bridge deck corrosion database, the collected steel bar corrosion current density data is steel bridge deck corrosion sample data;
step three, determining corrosion influence factors of the steel bridge deck plate:
the corrosion influence factors of the steel bridge deck are set as follows: temperature, humidity, number of times of energization, and energization time;
step four, preprocessing steel bridge deck corrosion sample data:
carrying out smooth denoising treatment on the fluctuation and burrs of the sample data caused by noise signals on the steel bridge deck corrosion sample data obtained in the step two;
step five, standardized processing of steel bridge deck corrosion sample data:
in order to reduce prediction misdetection, the mapmin max function in matlab is used for carrying out standardized processing on steel bridge deck corrosion sample data, so that the processed data are distributed in [0,1 ]]In the range of (2), the standardized formula is:wherein X is i Is normalized data, X is sample data, X max For maximum sample data, X min Minimum value of sample data;
step six, optimizing the extreme learning machine neural network by adopting a genetic algorithm to obtain an optimized neural network;
genetic Algorithms (GA) and Extreme Learning Machine (ELM) neural networks are well known mature technologies.
The extreme learning machine is a novel feedforward neural network, compared with the traditional single hidden layer neural network, the hidden layer of the extreme learning machine does not need iteration, has very fast learning speed, and the input weight and the hidden node bias areAnd then determining, and optimizing an Extreme Learning Machine (ELM) neural network by adopting a Genetic Algorithm (GA) in order to eliminate the influence of the model input weight and the hidden layer threshold random value on the prediction model precision. Compared with the conventional neural network which needs to optimize the input weight, the hidden node bias and the output weight, the method only needs to optimize the input weight omega of the ELM i Hidden node bias b i The optimization selection is carried out, so that the calculated amount is reduced;
as shown in fig. 2, the specific process of optimizing the neural network of the extreme learning machine by adopting the genetic algorithm is as follows:
step S61, determining a basic topological structure of the ELM neural network, and adopting binary coding for input weights and hidden node biases of the ELM model to obtain an initial population. Each individual is a binary string, θ= [ ω ] 11 ,ω 12 …,ω 1m ,ω 21 ,…,ω 2m ,…,ω n1 ,ω nm ,b 1 ,b 2 ,…,b m ]Wherein θ is an individual in the population; omega ij 、b j To be initialized to [ -0.5,0.5]Random values for the intervals. The dimension of the individual depends on the number of parameters to be optimized of the ELM neural network model, namely input weight and hidden node bias;
and step S62, decoding to obtain input weights and hidden node biases, and assigning the weights and biases to the ELM neural network. The network is trained with training samples while testing is performed using test samples. In order to reduce the residual error between the predicted value and the actual value to the maximum extent, setting a network objective function as follows:wherein n is the number of test samples, y i As predicted value, y i ' is the actual value;
step S63, determining fitness function and evolution algebra G. The fitness function employs a ranking fitness distribution function ranking, i.e. V Fit =ranking(V),Wherein s isDifferential pressure, ps is the position of the individual in the ordered population, d is the number of individuals in the population;
step S64, locally solving the optimal fitness function V Fit . Solving the fitness function of each individual one by one according to V Fit Determining individuals with better fitness;
step S65, solving the global optimal fitness function V Fit . And setting the initial value of the evolution algebra as 0, after each generation of local optimal fitness function solution is carried out, intersecting and mutating individuals with better fitness to generate a child population, calculating the fitness function of the child population again, selecting and inserting the individuals of the child population into the father population according to the fitness function value to replace the individuals with the minimum fitness in the father population to obtain a new population, and carrying out self-addition operation on the evolution algebra. When the evolution algebra is greater than G, the operation is ended. Calculating V at this time Fit And decoding the parameters corresponding to the optimal fitness function to obtain the optimal input weight and the hidden node bias, thereby establishing an optimal GA-ELM neural network model.
Step seven, training a steel bridge deck corrosion estimation model:
as shown in fig. 3, the corrosion influence factor of the steel bridge deck in the third step is taken as the input parameter of the optimized neural network obtained in the sixth step, the number of the input parameters is taken as the number of neurons of the input layer, the corrosion current density of the steel bars is taken as the output parameter of the optimized neural network obtained in the sixth step, and the optimized neural network is trained to obtain the corrosion estimation model of the steel bridge deck.
In step seven, 400 sets of data are used for training and 100 sets of data are used for testing in every 500 sets of input parameter test data.
As shown in fig. 4, the training process of the genetic algorithm optimization extreme learning machine neural network steel bridge deck corrosion estimation model comprises the following steps:
step S71, calculating a hidden layer output matrix: the expression is h=g (ωx T +b), where H is the hidden layer output matrix, i.eOmega is input weight, b is hidden layer node bias, omega and b are determined by genetic algorithm optimization, g is hidden layer activation function;
step S72, calculating the weight from the hidden layer to the output layer: the training output sample Y is adopted to replace the output value of the neural network, and according to the expression Y= (H) T Beta) solving least squares solution of hidden layer to output layer weight beta, i.eWherein (H) T ) + Moore-Penrose generalized inverse of the transposed matrix of the hidden layer output matrix.
Example 2:
the embodiment provides a steel bridge deck corrosion prediction model, which is obtained by adopting the construction method of the steel bridge deck corrosion prediction model in the embodiment 1.
The steel bridge deck corrosion degree is predicted by adopting a steel bridge deck corrosion prediction model, and as shown in fig. 5, the prediction process comprises the following steps:
step A, inputting environment humidity and temperature of cast conductive asphalt concrete, and power-on times and power-on time parameters under 54V power-on voltage to a GA-ELM steel bridge deck corrosion prediction model, and obtaining a steel bridge deck corrosion degree prediction value y by calculating and performing inverse normalization processing by using a mapmin max function i ;
Step B, in order to ensure the accuracy and the reliability of the GA-ELM steel bridge deck corrosion prediction model, calculating a steel bridge deck corrosion degree prediction value y i And the actual value y i The calculation formulas of the average absolute error, the average absolute percentage error and the root mean square error are as follows:
average absolute error:
average absolute percentage error:
root mean square error:
Claims (3)
1. the construction method of the steel bridge deck corrosion prediction model is characterized in that the construction method takes temperature, humidity, power-on times and power-on time as corrosion influence factors of the steel bridge deck, adopts a genetic algorithm to optimize an extreme learning machine neural network to obtain an optimized neural network, and trains the optimized neural network to obtain the steel bridge deck corrosion prediction model;
the construction method comprises the following steps:
step one, collecting corrosion data of a steel bridge deck plate:
selecting steel bars in different types of pouring type conductive asphalt mixture test pieces as data acquisition points, placing the combined structure into an environment control box after selecting corrosion data acquisition points, setting temperature, humidity, power-on times and power-on time, and testing and calculating to obtain corrosion current density of the steel bars after the steel bars are kept in the environment control box for 96 hours;
step two, establishing a steel bridge deck corrosion database:
the steel bar corrosion current density data obtained in the first step is stored in a steel bridge deck corrosion database, and in the steel bridge deck corrosion database, the collected steel bar corrosion current density data is steel bridge deck corrosion sample data;
step three, determining corrosion influence factors of the steel bridge deck plate:
the corrosion influence factors of the steel bridge deck are set as follows: temperature, humidity, number of times of energization, and energization time;
step four, preprocessing steel bridge deck corrosion sample data:
carrying out smooth denoising treatment on the fluctuation and burrs of the sample data caused by noise signals on the steel bridge deck corrosion sample data obtained in the step two;
step five, standardized processing of steel bridge deck corrosion sample data:
to reduce predictive false positives, the mapmin max function in matlab was used to sample steel bridge deck corrosionThe data is standardized to distribute the processed data in [0,1 ]]In the range of (2), the standardized formula is:wherein X is i Is normalized data, X is sample data, X max For maximum sample data, X min Minimum value of sample data;
step six, optimizing the extreme learning machine neural network by adopting a genetic algorithm to obtain an optimized neural network;
step seven, training a steel bridge deck corrosion estimation model:
and D, taking the corrosion influence factors of the steel bridge deck in the step three as input parameters of the optimized neural network obtained in the step six, taking the number of the input parameters as the number of neurons of an input layer, taking the corrosion current density of the steel bars as output parameters of the optimized neural network obtained in the step six, and training the optimized neural network to obtain a corrosion estimation model of the steel bridge deck.
2. The method for constructing a steel bridge deck corrosion prediction model according to claim 1, wherein in the first step, the setting range of the temperature in the environment control box is-15 ℃ to-5 ℃, the setting range of the humidity is 40% -80%, the setting range of the power-on times is 1-4, and the setting range of the power-on time is 0.5 h-1.5 h.
3. The method of constructing a predictive model of corrosion of a steel deck as recited in claim 1 wherein in step seven, 400 sets of data are used for training and 100 sets of data are used for testing for every 500 sets of input parametric test data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911081350.4A CN110889155B (en) | 2019-11-07 | 2019-11-07 | Steel bridge deck corrosion prediction model and construction method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911081350.4A CN110889155B (en) | 2019-11-07 | 2019-11-07 | Steel bridge deck corrosion prediction model and construction method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110889155A CN110889155A (en) | 2020-03-17 |
CN110889155B true CN110889155B (en) | 2024-02-09 |
Family
ID=69747021
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911081350.4A Active CN110889155B (en) | 2019-11-07 | 2019-11-07 | Steel bridge deck corrosion prediction model and construction method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110889155B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111814401B (en) * | 2020-07-08 | 2023-10-27 | 重庆大学 | LED life prediction method of BP neural network based on genetic algorithm |
CN112768056A (en) * | 2021-01-14 | 2021-05-07 | 新智数字科技有限公司 | Disease prediction model establishing method and device based on joint learning framework |
CN113111549B (en) * | 2021-03-31 | 2024-02-27 | 北京首钢股份有限公司 | Erosion model modeling method and modeling system for casting repaired blast furnace hearth |
CN113344219A (en) * | 2021-06-11 | 2021-09-03 | 广东电网有限责任公司 | Concrete reinforcement corrosion state evaluation method, system, terminal and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105069486A (en) * | 2015-09-09 | 2015-11-18 | 西安石油大学 | Extreme learning machine-based petrochemical equipment corrosion prediction method |
WO2017197626A1 (en) * | 2016-05-19 | 2017-11-23 | 江南大学 | Extreme learning machine method for improving artificial bee colony optimization |
CN109766990A (en) * | 2019-01-09 | 2019-05-17 | 西安建筑科技大学 | A kind of submerged pipeline corrosion rate prediction technique based on AGA-ELM hybrid algorithm |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10531825B2 (en) * | 2016-10-14 | 2020-01-14 | Stoecker & Associates, LLC | Thresholding methods for lesion segmentation in dermoscopy images |
-
2019
- 2019-11-07 CN CN201911081350.4A patent/CN110889155B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105069486A (en) * | 2015-09-09 | 2015-11-18 | 西安石油大学 | Extreme learning machine-based petrochemical equipment corrosion prediction method |
WO2017197626A1 (en) * | 2016-05-19 | 2017-11-23 | 江南大学 | Extreme learning machine method for improving artificial bee colony optimization |
CN109766990A (en) * | 2019-01-09 | 2019-05-17 | 西安建筑科技大学 | A kind of submerged pipeline corrosion rate prediction technique based on AGA-ELM hybrid algorithm |
Non-Patent Citations (2)
Title |
---|
基于IGA-ELM网络的滚动轴承故障诊断;皮骏;马圣;贺嘉诚;孔庆国;林家泉;刘光才;;航空学报(09);全文 * |
基于遗传算法优化BP神经网络预测CO_2/H_2S环境中套管钢的腐蚀速率;万里平;徐友红;冯兆阳;孔斌;杨兵;;腐蚀与防护(09);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN110889155A (en) | 2020-03-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110889155B (en) | Steel bridge deck corrosion prediction model and construction method | |
Li et al. | Automated decision making in highway pavement preventive maintenance based on deep learning | |
Wang et al. | Prediction of highway asphalt pavement performance based on Markov chain and artificial neural network approach | |
CN113344305B (en) | Rapid prediction method for rainstorm waterlogging event | |
Qiao et al. | Application of grey theory in pollution prediction on insulator surface in power systems | |
Chen et al. | Evaluation and prediction for effect of conductive gussasphalt mixture on corrosion of steel bridge deck | |
Trofimenko et al. | Modeling of influence of climate change character on the territory of the cryolithozone on the value of risks for the road network | |
CN114611759A (en) | Optimization method of maintenance scheme of asphalt pavement for major and medium maintenance of highway | |
Mahmood et al. | A unified artificial neural network model for asphalt pavement condition prediction | |
Wu et al. | Comprehensive evaluation of ice disaster risk of the Ningxia–Inner Mongolia Reach in the upper Yellow River | |
Jin et al. | Integration of an improved transformer with physical models for the spatiotemporal simulation of urban flooding depths | |
CN108199895B (en) | Intelligent card port optimal arrangement method and device | |
Ye et al. | Optimizing weight and threshold of BP neural network using SFLA: applications to nonlinear function fitting | |
CN116167121A (en) | Safety performance assessment method for bridge construction support system based on Gao Sibei phyllus network | |
Qian | Road pavement performance evaluation model based on hybrid genetic algorithm neural network | |
CN113541211A (en) | Method and system for determining toughness of alternating current-direct current power distribution network | |
Kazakov et al. | Measurement of highway maintenance patrol efficiency: model and factors | |
Zhang et al. | Efficient reliability assessment method for bridges based on Markov Chain Monte Carlo (MCMC) with Metropolis-Hasting Algorithm (MHA) | |
CN114861458B (en) | Bridge traffic safety rapid evaluation method based on mapping relation proxy model | |
Jinlei et al. | Insulator Contamination Prediction Model Based on BP Neural Network Optimized by Genetic Algorithm | |
Sood et al. | Pavement deterioration modeling in India | |
Han et al. | Pavement roughness level classification based on logistic and decision tree machine learnings | |
Marsili et al. | Explainable artificial intelligence for understanding the ageing classes of reinforced concrete bridge components | |
Wu et al. | A new reliability model of bridge fatigue based on mind evolutionary algorithm | |
Chen et al. | A Resilience Enhancement Model for Complex Distribution Network Coupling with Human Resources and Traffic Network |
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 |