CN115482882A - Pipeline corrosion data acquisition and pipeline corrosion rate prediction model construction method - Google Patents
Pipeline corrosion data acquisition and pipeline corrosion rate prediction model construction method Download PDFInfo
- Publication number
- CN115482882A CN115482882A CN202210987019.4A CN202210987019A CN115482882A CN 115482882 A CN115482882 A CN 115482882A CN 202210987019 A CN202210987019 A CN 202210987019A CN 115482882 A CN115482882 A CN 115482882A
- Authority
- CN
- China
- Prior art keywords
- corrosion
- pipeline
- data
- corrosion rate
- soil
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/10—Analysis or design of chemical reactions, syntheses or processes
-
- 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
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Analytical Chemistry (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Testing Resistance To Weather, Investigating Materials By Mechanical Methods (AREA)
Abstract
The invention relates to the technical field of pipeline corrosion data processing, in particular to a pipeline corrosion data acquisition and pipeline corrosion rate prediction model construction method, which comprises the steps of firstly acquiring a plurality of typical corrosive soils; then, for each of the typical corrosive soils, burying a plurality of pipeline samples; then, sequentially taking out pipeline samples from the plurality of typical corrosive soils according to a preset time node; finally, for each typical corrosive soil, respectively obtaining a maximum corrosion rate and a plurality of soil corrosion factors as pipeline corrosion data, wherein the maximum corrosion rate is obtained based on a plurality of pipeline samples taken from the typical corrosive soil. According to the embodiment of the invention, pipeline corrosion data are obtained through the sample and the corrosive soil, the corrosion curve is fitted according to the existing data, the obtained corrosion data can accurately reflect the relation between the dependent variable and the independent variable, and the obtained data is more accurate and reliable.
Description
Technical Field
The invention relates to the technical field of pipeline corrosion data processing, in particular to a method for acquiring pipeline corrosion data and constructing a pipeline corrosion rate prediction model.
Background
At present, the published data of the research on the service life of metal pipelines, particularly nodular cast iron pipelines, is less, and with the increase of the service life of buried nodular cast iron pipelines, the outer wall coating and the pipe wall of the buried nodular cast iron pipelines are corroded, slight corrosion has no influence on the use of the pipelines, severe corrosion has great influence on the pipelines, perforation and even breakage of the pipelines are often caused, and huge economic loss and social influence are caused. Therefore, it is very important to systematically study the corrosion resistance of the ductile cast iron pipe under different soil conditions and evaluate the expected service life.
On the other hand, users of ductile iron pipelines need to select the grade of ductile iron pipes meeting the requirement of engineering design life according to specific soil corrosion environment, so that the research and evaluation on the expected service life of the ductile iron pipeline system are urgent and important.
In the existing pipeline corrosion and service life research, most of the research focuses on the research on the service life of a steel pipe, the residual service life of the pipeline is calculated and evaluated according to the monitored corrosion rate after the steel pipe runs for years in the actual service corrosion environment, or the residual service life of the pipeline is calculated after the stress is calculated and analyzed by using finite element software according to the geometric dimension and shape of a corrosion pit on the outer wall of the steel pipe, and the service life research of the ductile iron pipeline is almost blank. The main reason is that the components of the ductile iron pipe and the steel pipe are different, and the corrosion rules are different.
Due to a number of factors affecting soil corrosivity, including soilSoil texture, oxygen content, soil resistivity, cl - 、SO 4 2- The corrosion condition of the pipeline under the condition of given soil physicochemical property parameters is calculated and evaluated by necessarily and systematically researching the corrosion factors of the soil and the corrosion behavior of the pipeline in the soil environment.
Based on this, it is necessary to develop and design a pipeline corrosion data acquisition method and a pipeline corrosion rate prediction method.
Disclosure of Invention
The embodiment of the invention provides a method for acquiring pipeline corrosion data and constructing a pipeline corrosion rate prediction model, which is used for solving the problem of poor corrosion modeling effect caused by inaccurate acquisition of the pipeline corrosion data in the prior art.
In a first aspect, an embodiment of the present invention provides a method for acquiring pipeline corrosion data, including:
obtaining a plurality of typical corrosive soils, wherein the typical corrosive soils have a corrosive effect on the pipes;
burying a plurality of pipe samples for each of said typical corrosive soils;
sequentially taking out pipeline samples from the typical corrosive soils according to a preset time node;
for each typical corrosive soil, respectively obtaining a maximum corrosion rate and a plurality of soil corrosion factors as pipeline corrosion data, wherein the soil corrosion factors characterize the factors of the pipeline laying ground influencing the pipeline corrosion rate, and the maximum corrosion rate is obtained based on a plurality of pipeline samples taken from the typical corrosive soil.
In one possible implementation, the burying, for each of the typical corrosive soils, a plurality of pipeline samples includes:
the pipeline sample material matrix is a metal pipe, a zinc coating is sprayed on the outer surface of the metal pipe, and an organic finishing coating is sprayed on the zinc coating;
and after the pipeline samples are subjected to edge sealing and drying, the pipeline samples are axially and horizontally placed at the bottom of the test pit according to the type sequence, the intervals between the pipeline samples and the edge of the test pit are larger than a preset value, and soil is paved on the pipeline samples, so that the soil around the pipeline samples is compact and has no gaps.
In one possible implementation, the intervals between the plurality of predetermined time nodes are set according to a gradually increasing rule;
obtaining a maximum corrosion rate from a plurality of pipe samples taken from the typical corrosive soil comprises:
for each pipeline sample, acquiring corrosion pit depth data of the deepest corrosion pit;
converting the plurality of etch pit depth data into maximum etch pit rate data corresponding to the time node according to a predetermined time node;
constructing a corrosion rate expectation function according to a plurality of time nodes and a plurality of maximum corrosion pit rate data corresponding to the time nodes, wherein the corrosion rate expectation function is as follows:
V＝a.t b
where V is the corrosion rate of a typical corrosive soil area, a is the forward constant, b is the backward constant, and t is the time.
In one possible implementation, after the obtaining the maximum corrosion rate and the plurality of soil corrosion factors for each of the typical corrosive soils respectively, the method includes:
acquiring a measured corrosion rate of an actually buried pipeline and a plurality of corrosion factors of actually buried soil;
normalizing according to the corrosion rate expectation function and the measured corrosion rate of the actually buried pipeline to obtain the corrosion rate of the actually buried pipeline;
the corrosion rate of the actually buried pipeline and a plurality of corrosion factors of soil of the actually buried pipeline are added to the pipeline corrosion data.
In one possible implementation, the corrosion rate of the actually buried pipeline is obtained by normalizing the corrosion rate according to the expected corrosion rate function and the measured corrosion rate of the actually buried pipeline
Normalizing according to the corrosion rate expectation function, a first formula and the measured corrosion rate of the actually buried pipeline to obtain the corrosion rate of the actually buried pipeline, wherein the first formula is as follows:
in the formula, V a For normalized corrosion rate, t, of the actual buried pipeline 1 Time node, V, for sampling of the actual buried pipeline 1 For the actual buried pipeline at t 1 Measurement of corrosion rate, V, at a time node 0 Is the long-term corrosion rate of areas with strongly corrosive soil.
In a second aspect, an embodiment of the present invention provides a method for constructing a pipeline corrosion rate prediction model, including:
obtaining a plurality of corrosion data, wherein the corrosion data comprises a long-term average corrosion rate and a plurality of soil corrosion factors, and the plurality of corrosion data are obtained by using the pipeline corrosion data obtaining method according to the first aspect or any one of the possible implementations of the first aspect;
constructing an initial model, wherein the initial model is constructed based on a neural network;
dividing the plurality of corrosion data into a training data set and a testing data set;
training: inputting a plurality of corrosion data of the training data set into the initial model to train the initial model until the output error of the initial model is lower than a threshold value;
and inputting a plurality of corrosion data of the test data set into the initial model, obtaining a test error of the initial model, and if the test error is higher than a threshold value, adjusting the structure of the initial model and skipping to the training step.
In one possible implementation, the neural network model is an Elman neural network model, including: the device comprises an input layer, a hidden layer, an output layer and a bearing layer;
the number of the nodes of the input layer is the same as the number of the plurality of soil corrosion factors of the corrosion data;
the output layer includes a long term average corrosion rate output node having an activation function of:
f(x)＝kx+d
wherein f (x) is an activation function of the long-term average corrosion rate output node, k is a weight coefficient, and d is a first bias;
determining the number of nodes of the hidden layer according to a first formula, the number of nodes of the input layer and the number of nodes of the output layer, wherein the first formula is as follows:
S＝(a 0 +b 0 )*0.4+c
wherein S is the number of hidden nodes, a 0 Is the number of nodes of the input layer, b 0 B is the number of output layer nodes, and c is a second bias;
the adjusting the structure of the initial model comprises: increasing or decreasing the number of nodes of the hidden layer.
In a third aspect, an embodiment of the present invention provides an operation and maintenance device for an electric power metering device, including:
a corrosion data acquisition module, configured to acquire a plurality of corrosion data, where the corrosion data includes a long-term average corrosion rate and a plurality of soil corrosion factors, and the plurality of corrosion data are acquired by using the pipeline corrosion data acquisition method according to the first aspect or any one of the possible implementations of the first aspect;
the initial model building module is used for building an initial model, wherein the initial model is built on the basis of a neural network;
a data classification module for classifying the plurality of corrosion data into a training data set and a test data set;
a training module for implementing the training steps: inputting a plurality of corrosion data of the training data set into the initial model to train the initial model until the output error of the initial model is lower than a threshold value;
and (c) a second step of,
and the testing module is used for inputting the plurality of corrosion data of the testing data set into the initial model, acquiring the testing error of the initial model, and if the testing error is higher than a threshold value, adjusting the structure of the initial model and skipping to the training step.
In a fourth aspect, the present invention provides an electronic device, which includes a memory and a processor, where the memory stores a computer program executable on the processor, and the processor executes the computer program to implement the steps of the method according to any one of the possible implementation manners of the second aspect or the second aspect.
In a fifth aspect, the present invention provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program implements the steps of the method according to the second aspect or any possible implementation manner of the second aspect.
Compared with the prior art, the implementation mode of the invention has the following beneficial effects:
the embodiment of the invention discloses a method for acquiring pipeline corrosion data, which comprises the steps of firstly acquiring a plurality of typical corrosive soils, wherein the typical corrosive soils have corrosive action on pipelines; then, for each of the typical corrosive soils, burying a plurality of pipeline samples; sequentially taking out pipeline samples from the plurality of typical corrosive soils according to a preset time node; finally, for each typical corrosive soil, respectively obtaining a maximum corrosion rate and a plurality of soil corrosion factors as pipeline corrosion data, wherein the soil corrosion factors characterize the factors of the pipeline laying ground influencing the pipeline corrosion rate, and the maximum corrosion rate is obtained based on a plurality of pipeline samples taken from the typical corrosive soil. According to the embodiment of the invention, the pipeline corrosion data is obtained through the sample and the corrosive soil, and the corrosion curve is fitted according to the existing data, so that the obtained corrosion data can accurately reflect the relation between the dependent variable and the independent variable, and the obtained data is more accurate and reliable.
The method for constructing the pipeline corrosion rate prediction model provided by the embodiment of the invention adopts the neural network model to construct, the data acquired by the pipeline corrosion data implementation mode of the invention can predict the long-term corrosion rate according to the corrosion factor of soil, and the service life of the metal pipeline can be more accurately estimated by combining the corrosion allowance, so that the pipeline meeting the requirements of users can be designed according to the soil property, and the requirements of the clients are met.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive labor.
FIG. 1 is a flow chart of a pipeline corrosion data acquisition method provided by an embodiment of the invention;
FIG. 2 is a flow chart of a method for constructing a pipeline corrosion rate prediction model according to an embodiment of the present invention;
fig. 3 is a basic structure diagram of an Elman neural network model provided by the embodiment of the invention;
FIG. 4 is a functional block diagram of a device for constructing a pipeline corrosion rate prediction model according to an embodiment of the present invention;
fig. 5 is a functional block diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
To make the objects, technical solutions and advantages of the present invention more apparent, the following description is given by way of embodiments with reference to the accompanying drawings.
The following is a detailed description of the embodiments of the present invention, which is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
Fig. 1 is a flowchart of a pipeline corrosion rate prediction model building method according to an embodiment of the present invention.
As shown in fig. 1, it shows a flowchart of an implementation of the method for constructing a pipeline corrosion rate prediction model provided in the first aspect of the embodiment of the present invention, which is detailed as follows:
in step 101, a plurality of typically corrosive soils are obtained, wherein the typically corrosive soils have a corrosive effect on the pipes.
In step 102, for each of the typical corrosive soils, a plurality of pipe coupons are buried.
In some embodiments, said burying for each of said typical corrosive soils a plurality of pipe samples comprises:
the pipeline sample material matrix is a metal pipe, a zinc coating is sprayed on the outer surface of the metal pipe, and an organic finishing coating is sprayed on the zinc coating;
and after the pipeline samples are subjected to edge sealing and drying, the pipeline samples are axially and horizontally placed at the bottom of the test pit according to the type sequence, the intervals between the pipeline samples and the edge of the test pit are larger than a preset value, and soil is paved on the pipeline samples, so that the soil around the pipeline samples is compact and has no gaps.
In some application scenes, D1, D2, D3, \8230, dm typical corrosive soil areas are selected as metal pipeline burying experimental sites, m is 4-20, and 4-15 test pits for burying samples are dug in each area to prepare the required metal pipe samples.
The metal pipeline sample material matrix is a metal pipe such as a ductile iron pipe or a steel pipe, and the metal pipe is firstly coated with a metal coatingArc spraying zinc coating on the outer surface of the road, wherein the unit weight of the zinc coating is 130g/m 2 ～200g/m 2 And then spraying an organic finishing coating on the zinc coating, wherein the thickness of the finishing coating is 70-150 microns. (the presence of a coating is not a sufficient requirement and is only one of the factors).
Then, the pipeline is processed into required sample size in a processing workshop, the sample size is DN100 multiplied by 100mm pipe sections, the number of parallel samples prepared in each coating period is 3-10, and the samples are subjected to edge sealing treatment. And drying the coating sample for one week, packaging the coating after the coating is completely dried, and transporting the packaged sample to the typical soil experiment stations. Removing the package of the sample, axially and horizontally placing the sample at the bottom of the test pit according to the type sequence, fixing the sample by soil, compacting the sample, ensuring that the interval between the sample and the edge of the test pit is more than 150mm, ensuring that the soil around the sample is compact without gaps, and then sequentially backfilling the excavated soil according to the original sequence.
In step 103, pipeline samples are sequentially taken from the plurality of typical corrosive soils according to a predetermined time node.
In step 104, for each of the typical corrosive soils, a maximum corrosion rate and a plurality of soil corrosion factors are respectively obtained as pipeline corrosion data, wherein the soil corrosion factors characterize the factors of the pipeline laying ground affecting the pipeline corrosion rate, and the maximum corrosion rate is obtained based on a plurality of pipeline samples taken from the typical corrosive soil.
In some embodiments, the intervals between each of the predetermined time nodes are set according to a gradually increasing rule;
obtaining a maximum corrosion rate from a plurality of pipe specimens taken from the typical corrosive soil comprises:
for each pipeline sample, obtaining the etch pit depth data of the deepest etch pit;
converting the plurality of etch pit depth data into maximum etch pit rate data corresponding to the time node according to a predetermined time node;
constructing a corrosion rate expectation function according to a plurality of time nodes and a plurality of maximum corrosion pit rate data corresponding to the time nodes, wherein the corrosion rate expectation function is as follows:
V＝a·t b
where V is the corrosion rate of a typical corrosive soil area, a is the forward constant, b is the backward constant, and t is the time.
Illustratively, after the pipeline sample is buried, T1, T2, T3, \8230, tn are excavated and detected in sequence according to a predetermined excavation year cycle sequence. The excavation period n can be set to be 3-12 periods, and the excavation period can be set to be 0.082, 0.164, 0.247 and 7 years according to the principle of first density and then sparse (the rule is that the interval period is gradually increased). Years 0.082, 0.164, 0.247, 0.493, 1 and 7 may also be mentioned. After the samples are dug out in each period, soil on the samples is carefully removed until the samples are completely removed, and when the soil is strong in adhesion and difficult to wipe off, a soft brush made of plastic can be used for soaking a proper amount of water to gently wash the soil and remove the soil. And (3) drying the sample in a drying box, taking out the sample after drying the sample for 24 hours, then processing and sampling the maximum pitting position, and detecting the maximum pitting depth data of the maximum pitting position by using an SEM scanning electron microscope.
Then, the maximum corrosion pit depth data of Tn samples in different periods T1, T2, T3, \8230, and the maximum corrosion rate data of the Tn samples in corresponding periods V1, V2, V3, \8230andVn are converted;
the maximum corrosion rate data of D1, D2, D3, \8230;, dm positions with different corrosion periods T1, T2, T3, \8230;, tn time are utilized to establish m shapes as follows: v = a · t b Wherein V is the etch rate, t is time, and a and b are constants.
Defining the soil environment with the maximum corrosion rate in each site under the condition of the longest period Tn as the most severe corrosion environment, and defining the corrosion rate model as a reference model, for example, defining the D3 site with the maximum corrosion rate under the condition of Tn as the reference model, defining the D3 site with the corrosion rate under the condition of the longest period Tn as V 0 。
In some embodiments, after step 104, comprising:
acquiring a measured corrosion rate of an actually buried pipeline and a plurality of corrosion factors of actually buried soil;
normalizing according to the corrosion rate expectation function and the measured corrosion rate of the actually buried pipeline to obtain the corrosion rate of the actually buried pipeline;
the corrosion rate of the actual buried pipeline and a plurality of corrosion factors of the soil of the actual buried pipeline are added to the pipeline corrosion data.
In some embodiments, said normalizing according to said expected corrosion rate function and the measured corrosion rate of the actually buried pipe to obtain the corrosion rate of the actually buried pipe comprises:
normalizing according to the corrosion rate expectation function, a first formula and the measured corrosion rate of the actually buried pipeline to obtain the corrosion rate of the actually buried pipeline, wherein the first formula is as follows:
in the formula, V a For normalized corrosion rate, t, of the actually buried pipeline 1 Time node, V, for sampling of the actual buried pipeline 1 For the actual buried pipeline at t 1 Measurement of corrosion rate, V, at a time node 0 Is the long-term corrosion rate of the areas with strong corrosive soil.
For example, in the pipeline corrosion data acquisition, the corrosion data obtained in the field can be applied to increase the number and provide necessary conditions for constructing a more precise corrosion model.
For example, let each place such as Jiadi t 1 Corrosion rate V for excavating and detecting pipelines buried in different ages in actual year 1 A standard normalization was performed to the etch rate data for the longest period Tn. The normalization method comprises the following steps:
V a for normalized corrosion rate, t, of the actual buried pipeline 1 Time node, V, for sampling of the actual buried pipeline 0 Corrosion rate, V, for the longest time node for a highly corrosive soil region 1 For the actual buried pipeline at t 1 Corrosion rate of time nodes.
And finally, adding the normalized corrosion rate and corrosion factors (such as physicochemical indexes) of the soil of the first land into the pipeline corrosion data as a part of the pipeline corrosion data.
As shown in fig. 2, in a second aspect, an embodiment of the present invention provides a method for constructing a pipeline corrosion rate prediction model, including:
In some embodiments, the neural network model is an Elman neural network model, comprising: the device comprises an input layer, a hidden layer, an output layer and a bearing layer;
the number of the nodes of the input layer is the same as the number of the plurality of soil corrosion factors of the corrosion data;
the output layer includes a long term average corrosion rate output node having an activation function of:
f(x)＝kx+d
wherein f (x) is an activation function of the long-term average corrosion rate output node, k is a weight coefficient, and d is a first bias;
determining the number of nodes of the hidden layer according to a first formula, the number of nodes of the input layer and the number of nodes of the output layer, wherein the first formula is as follows:
S＝(a 0 +b 0 )*0.4+c
wherein S is the number of hidden nodes, a 0 Is the number of nodes of the input layer, b 0 C is the number of output layer nodes, and c is the second bias;
the adjusting the structure of the initial model comprises: increasing or decreasing the number of nodes of the hidden layer.
Illustratively, soil erosion factors typically include physicochemical factors such as soil resistivity, oxidation-reduction potential, cl - 、SO 4 2- Water content, pH value and salt content, and the water content, the pH value and the salt content are used as dependent variables and input into a model, so that the expected long-term average corrosion rate can be obtained.
In some application scenarios, the prediction model is constructed based on an Elman neural network model, which is generally divided into four layers: an input layer, a hidden layer (intermediate layer), a handle layer and an output layer. As shown in fig. 3, the connection of the input layer, the hidden layer and the output layer is similar to a feedforward network, and the Elman neural network is a typical dynamic neural network, which is based on the basic structure of the BP network, and has a function of mapping dynamic characteristics by storing internal states, so that the system has the capability of adapting to time-varying characteristics. Because the training samples are fewer, the situation of relatively large errors in prediction is possible, and the situation can be avoided by increasing the sample size, eliminating error data in advance and the like.
In the embodiment of the invention, a network model net is established by a computer, and the net model hierarchical structure comprises an input layer, a hidden layer, a carrying layer and an output layer;
in the aspect of node setting, the number a of influencing factor nodes of an input layer is determined first 0 Such as 7 soil corrosion factors (soil resistivity, oxidation-reduction potential, cl) - 、SO 4 2- Water content, pH value, salt content) in which case a 0 =7, i.e. a is selected from soil corrosion factors 0 And a corresponding factor.
Then determining the node number b of the output layer 0 Average corrosion rate over a long period of time, e.g. for nodular cast iron pipes, when b 0 ＝1(b 0 To define a value, the output data port is referred to as 1, a fixed value).
Next, determining a hidden node number range, wherein the rule for determining the hidden node number is as follows:
S＝(a 0 +b 0 )*0.4+c
wherein S is the number of hidden nodes, a 0 As the number of nodes of the input layer, b 0 The constant c = 2-9 is the number of nodes of the output layer, and the range of the number of the calculated hidden nodes is [ m1, m2 ]]。
Then, establishing a network, debugging the training network, determining the number of hidden layer nodes as m1 and the hidden layer function as
Or
The output layer function is:
f(x)＝kx+d
wherein a' is a positive real number, and d and k values are determined, namely parameters are determined for the model.
Finally, setting the network learning rate to be 0.1-0.001, the network training times to be 1000-100000 and the error to be 0.1-0.0001, and then starting the network training.
And when the network training is finished after the set error is reached within the set training times, the network parameters can be stored after test verification, otherwise, the network establishing step is returned, the number of the hidden layer nodes is +1, the network, the hidden layer function and the output layer function are reestablished until the network convergence is successful, the network parameters are stored after test verification, and the stored network is named as net1.
Based on the corrosion prediction model, the life of the pipeline and the allowable working pressure of the pipeline in a specified age can be predicted, and the application of the model to three embodiments is described below.
Example 1:
firstly, four typical corrosive soil areas A, B, C and D in China are selected as nodular cast iron pipe burying experimental sites, and the maximum pitting depth data of excavation detection samples are carried out according to excavation periods of 1 month, 2 months, 3 months and 84 months.
After SEM detection, the corrosion rate of the D site in the A, B, C and D sites is the maximum, and the maximum value is 141.17 mu m/a and is marked as V 0 . Therefore, the D site is used as the most serious corrosion model to establish a corrosion rate prediction model, and the specific data are shown in Table 1
TABLE 1 site soil sample Corrosion Rate
The fitted model is:
V＝207.289t -0.229
the corrosion rates of 15 years of site-excavated pipeline E site, 16.2 years of site F site, 16.2 years of site G site and 6 years of site H site are respectively 19.80, 37.60, 33.13 and 1200.00 mu m/a, and V is utilized 1 *V 0 /(a*(t 1 ) b ) After the formula is normalized, the corrosion rates of the E site, the F site, the G site and the H site in 7 years are respectively 25.10 μm/a, 48.45 μm/a, 42.69 μm/a and 1231.81 μm/a.
The training data sequence is shown in table 2.
TABLE 2 training data sequence Listing
The test data series are shown in table 3.
Table 3 test data sequence table
Establishing a network model net by using a computer, wherein the model layer comprises an input layer, a hidden layer and an output layer;
determining the number of the nodes of the influence factors of the input layer to be 6, namely 6 soil corrosion factors (soil resistivity and Cl) - 、SO 4 2- Water content, pH value, salt content), in which case a 0 ＝6；
Determining the number of nodes b of the output layer 0 Average corrosion rate value of nodular cast iron pipeline in a long term, at this moment b 0 =1; the determination of the number of hidden nodes is another very important link in the neural network design, and due to the complexity of the problem, a good analytic expression is not found so far, and the example is according to the formula:
S＝(a 0 +b 0 )*0.4+c
the number of hidden nodes is determined to be between 3 and 12. Then, the number of hidden layer neurons is tested from 3 to 12 respectively, and the number of the hidden layer neurons corresponding to the optimal combination of the network training error and the network training frequency is selected through network training result comparison, as shown in table 4.
TABLE 4 iterative training situation table for different node numbers of neural prediction network hidden layers
As can be seen from table 1, as the number of hidden nodes increases, the average number of network iterations is significantly reduced. But as the number of nodes increases further, the average number of iterations of the network does not improve significantly. Considering that the number of hidden nodes is more, the generalization capability is poorer, namely, the identification capability of a new sample is poorer, and considering the convergence success number, the maximum iteration number and the minimum iteration number comprehensively, the number of hidden nodes of the neural network is selected to be S =4.
After multiple debugging tests, determining the hidden layer function as
Determining the output layer function as: f (x) =4x +6
And setting the network learning rate to be 0.05, the network training times to be 10000 and the error to be 0.01, and then starting the network training.
And after the error reaches 0.01 within 10000 training times, the network training is finished, and the network parameter is stored as net2 after test verification.
Inputting soil physicochemical property parameters such as soil resistivity of 44 Ω · cm and Cl under soil conditions of a given u place into the established network net2 - The content of SO is 1.6780% 4 2- The content was 0.1561%, the water content was 23.95%, the pH was 8.67, the salt content was 2.9660%, and the expected corrosion rate of the ductile iron pipe calculated by net2 was 15.29 μm/a.
The corrosion allowance calculation formula of the pipeline gives a mode of determining the corrosion allowance of the pipeline according to the minimum wall thickness of the pipeline, the actual allowable working pressure, the safety coefficient, the outer diameter and the tensile strength of a pipeline material, and the corrosion allowance calculation formula of the pipeline is as follows:
wherein h is the corrosion allowance of the pipeline, e min Of minimum wall thickness of the pipe, PFA Practice of For the actual allowable working pressure, SF is the safety factor for the actual allowable working pressure, DE is the outside diameter of the pipe, R m Is the tensile strength of the tubing material.
Calculating the corrosion allowance of the DN 100C-grade pipeline to be 2.58mm under the condition of the allowable working pressure of 10bar by using a corrosion allowance calculation formula of the pipeline;
the expected service life of the ductile iron pipe is L =2.58/0.01529=167.4 years;
the predetermined age allowable working pressure calculation formula gives the manner in which the predetermined age allowable working pressure is determined based on the expected long term average corrosion rate, the minimum wall thickness of the pipe, the tensile strength of the pipe material, and the outer diameter of the pipe, and the predetermined age allowable working pressure calculation formula is:
in the formula, p is T 0 Allowable operating pressure of age, e min Minimum wall thickness of pipe, V 3 To predict the average corrosion rate over a long period of time, R m The tensile strength of the pipe material and DE the outer diameter of the pipe.
Calculating the allowable working pressure p of the DN 100C-grade pipeline after 100 years; calculated by using a calculation formula of the allowable working pressure of a preset age
The final output DN100C class pipeline expected service life is 167.4 years, and the allowable working pressure p of the pipeline after 100 years is given to be 35.3bar.
Example 2:
firstly, four typical corrosive soil areas D1, D2, D3 and D4 in China are selected as ductile iron pipe burying experimental sites, and the maximum pit depth data of the excavation detection samples are excavated according to excavation periods of 0.082 year, 0.164 year, 0.247 year, 0.493 year, 1 year and 7 years.
After SEM detection, the corrosion rate of the D4 site in the D1, D2, D3 and D4 sites is the maximum, the maximum value is 141.17 mu m/a, and is marked as V 0 . Therefore, a corrosion rate prediction model is established by taking the D4 site as the most serious corrosion model, and specific data are as follows.
TABLE 5 D4 site bare tube (with pockmark) sample buried in soil corrosion speed
The fitting result is:
V＝205.731t -0.231
the corrosion rates of 15 years of D5 site, 16.2 years of D6 site, 16.2 years of D7 site and 6 years of D8 site of the pipeline excavated on site are respectively 19.80, 37.60, 33.13 and 1200.00 mu m/a, and V is utilized 1 *V 0 /(a*(t 1 ) b ) After the formula is normalized, the corrosion rates of the D5 site, the D6 site, the D7 site and the D8 site in 7 years are respectively 25.40 μm/a, 49.09 μm/a, 43.26 μm/a and 1245.60 μm/a.
Then summarizing, sorting and analyzing data of other field excavation pipelines, and taking data of 25 sites as training data and data of 5 sites as test data in total;
establishing a network model net by using a computer, wherein the model layer comprises an input layer, a hidden layer, a bearing layer and an output layer;
determining the number of influencing factor nodes of the input layer to be 7, namely 7 soil corrosion factors (soil resistivity, oxidation-reduction potential and Cl) - 、SO 4 2- Water content, pH value, salt content), in which case a 0 ＝7；
Determining the number of nodes b of the output layer 0 Average corrosion rate value of nodular cast iron pipeline in a long term, at this moment b 0 ＝1；
According to the formula:
S＝(a 0 +b 0 )*0.4+c
the number of hidden nodes is determined to be between 4 and 15. Then, respectively testing the number of the hidden layer neurons from 4 to 15, comparing network training results, and selecting the number of the 12 hidden layer neurons corresponding to the optimal combination of the network training error and the network training frequency;
after multiple debugging tests, determining the hidden layer function as
In the formula, e is a natural constant, and a' is a positive real number.
Determining the output layer function as:
f(x)＝3.5x+7
setting the network learning rate to be 0.03, the network training times to be 20000, and starting the network training after the error is 0.001.
And after the error reaches 0.001 within 20000 training times, the network training is finished, and the network parameter is stored as net3 after test verification.
Inputting soil physicochemical property parameters under soil conditions of given w places in the established network, such as soil resistivity of 59660 omega cm, oxidation-reduction potential of 558.0mV, cl - The content of the active ingredients is 0.0014 percent and SO 4 2- The content was 0.0010%, the water content was 11.3%, the pH was 6.9, the salt content was 0.02%, and the expected corrosion rate of the ductile iron pipe was calculated to be 27.73 μm/a using net3.
The corrosion allowance calculation formula of the pipeline gives a mode of determining the corrosion allowance of the pipeline according to the minimum wall thickness of the pipeline, the actual allowable working pressure, the safety coefficient, the outer diameter and the tensile strength of a pipeline material, and the corrosion allowance calculation formula of the pipeline is as follows:
wherein h is the corrosion allowance of the pipeline, e min Of minimum wall thickness of the pipe, PFA Practice of For the actual allowable working pressure, SF is the safety factor for the actual allowable working pressure, DE is the outside diameter of the pipe, R m Is the tensile strength of the pipe material.
Calculating the corrosion allowance of the DN 600K 9 grade pipeline under the allowable working pressure of 16bar to be 4.39mm by using a corrosion allowance calculation formula of the pipeline;
the expected service life of the ductile iron pipe is L =4.39/0.02773=158.3 years;
the predetermined age allowable working pressure calculation formula gives the manner in which the predetermined age allowable working pressure is determined based on the expected long term average corrosion rate, the minimum wall thickness of the pipe, the tensile strength of the pipe material, and the outer diameter of the pipe, and the predetermined age allowable working pressure calculation formula is:
in the formula, p is T 0 Allowable operating pressure of year, e min Is the minimum wall thickness of the pipe, V 3 To predict the average corrosion rate over a long period of time, R m The tensile strength of the pipe material and DE the outer diameter of the pipe.
Calculating the allowable working pressure of the DN 600K 9-grade pipeline after 60 years; the allowable working pressure of the DN 600K 9-grade pipeline after 60 years of working is obtained by calculation by using a calculation formula of the allowable working pressure in a preset year, namely 28.1bar.
Example 3
The soil physicochemical property parameters in the XX area soil environment provided by the customer of the X water company are as follows: the soil resistivity is 169820 omega cm, the oxidation-reduction potential is 156.8mV, and Cl - Content of 0.325% SO 4 2- The content of the water is 0.782%, the water content is 8.6%, the pH value is 8.3, the salt content is 0.916%, the allowable working pressure of the delivered water is 10bar, a DN100 nodular iron pipeline is selected, the expected working life is 100 years, and a proper pipe is selected for customers.
The method establishes a calculation and evaluation method net4, and the expected corrosion rate of the ductile iron pipeline is calculated to be 21.87 mu m/a.
The corrosion allowance calculation formula of the pipeline gives a mode of determining the corrosion allowance of the pipeline according to the minimum wall thickness of the pipeline, the actual allowable working pressure, the safety coefficient, the outer diameter and the tensile strength of a pipeline material, and the corrosion allowance calculation formula of the pipeline is as follows:
wherein h is the corrosion allowance of the pipeline, e min Of minimum wall thickness of the pipe, PFA Practice of For the actual allowable working pressure, SF for the safety factor of the actual allowable working pressure, DE for the outside diameter of the pipe, R m Is the tensile strength of the pipe material.
Calculating the corrosion allowance of the DN 100C-grade pipeline to be 2.58mm under the condition of allowable working pressure of 10bar by using a corrosion allowance calculation formula of the pipeline; the expected service life of the ductile iron pipe L =2.58/0.02187=117.9 years.
The calculation evaluation shows that the DN100C grade pipeline can meet the use requirement without using a thicker K9 grade pipeline, thereby saving metal resources, saving energy and reducing carbon, and meeting the requirement of users.
According to the embodiment of the method for acquiring the pipeline corrosion data, a plurality of typical corrosive soils are acquired at first, wherein the typical corrosive soils have corrosive effects on pipelines; then, for each of the typical corrosive soils, burying a plurality of pipeline samples; then, sequentially taking out pipeline samples from the plurality of typical corrosive soils according to a preset time node; finally, for each typical corrosive soil, respectively obtaining a maximum corrosion rate and a plurality of soil corrosion factors as pipeline corrosion data, wherein the soil corrosion factors characterize the factors of the pipeline laying ground influencing the pipeline corrosion rate, and the maximum corrosion rate is obtained based on a plurality of pipeline samples taken from the typical corrosive soil. According to the embodiment of the invention, the pipeline corrosion data is obtained through the sample and the corrosive soil, and the corrosion curve is fitted according to the existing data, so that the obtained corrosion data can accurately reflect the relation between the dependent variable and the independent variable, and the obtained data is more accurate and reliable.
The method for constructing the pipeline corrosion rate prediction model provided by the embodiment of the invention adopts the neural network model to construct, the data acquired by the pipeline corrosion data implementation mode of the invention can predict the long-term corrosion rate according to the corrosion factor of soil, and the service life of the metal pipeline can be more accurately estimated by combining the corrosion allowance, so that the pipeline meeting the requirements of users can be designed according to the soil property, and the requirements of the clients are met.
It should be understood that the sequence numbers of the steps in the above embodiments do not mean the execution sequence, and the execution sequence of each process should be determined by the function and the inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
The following are apparatus embodiments of the invention, and for details not described in detail therein, reference may be made to the corresponding method embodiments described above.
Fig. 4 is a functional block diagram of an operation and maintenance device for electric power metering equipment according to an embodiment of the present invention, and referring to fig. 4, the operation and maintenance device for electric power metering equipment 4 includes: an erosion data acquisition module 401, an initial model building module 402, a data classification module 403, a training module 404, and a testing module 405.
A corrosion data acquiring module 401, configured to acquire a plurality of corrosion data, where the corrosion data includes an average corrosion rate over a long period of time and a plurality of soil corrosion factors, and the plurality of corrosion data is acquired by the pipeline corrosion data acquiring method according to any one of claims 1 to 5;
an initial model building module 402, configured to build an initial model, where the initial model is built based on a neural network;
a data classification module 403, configured to classify the plurality of corrosion data into a training data set and a test data set;
a training module 404, configured to implement a training step: inputting a plurality of corrosion data of the training data set into the initial model to train the initial model until the output error of the initial model is lower than a threshold value;
a test module 405, configured to input the corrosion data of the test data set into the initial model, obtain a test error of the initial model, and adjust the structure of the initial model and skip to the training step if the test error is higher than a threshold.
Fig. 5 is a functional block diagram of an electronic device provided in an embodiment of the invention. As shown in fig. 5, the electronic apparatus 5 of this embodiment includes: a processor 500 and a memory 501, the memory 501 having stored therein a computer program 502 executable on the processor 500. When the processor 500 executes the computer program 502, the steps in the above-mentioned operation and maintenance methods and embodiments of the power metering device, such as steps 201 to 205 shown in fig. 2, are implemented.
Illustratively, the computer program 502 may be partitioned into one or more modules/units that are stored in the memory 501 and executed by the processor 500 to implement the present invention.
The electronic device 5 may be a desktop computer, a notebook, a palm computer, a cloud server, or other computing devices. The electronic device 5 may include, but is not limited to, a processor 500, and a memory 501. Those skilled in the art will appreciate that fig. 5 is merely an example of an electronic device 5 and does not constitute a limitation of the electronic device 5 and may include more or fewer components than shown, or some components may be combined, or different components, e.g., the electronic device may also include input-output devices, network access devices, buses, etc.
The Processor 500 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The storage 501 may be an internal storage unit of the electronic device 5, such as a hard disk or a memory of the electronic device 5. The memory 501 may also be an external storage device of the electronic device 5, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, provided on the electronic device 5. Further, the memory 501 may also include both an internal storage unit and an external storage device of the electronic device 5. The memory 501 is used for storing the computer program and other programs and data required by the electronic device. The memory 501 may also be used to temporarily store data that has been output or is to be output.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit, and the integrated unit may be implemented in a form of hardware, or may be implemented in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. For the specific working processes of the units and modules in the system, reference may be made to the corresponding processes in the foregoing method embodiments, which are not described herein again.
In the above embodiments, the description of each embodiment is focused on, and for parts that are not described or recited in a certain embodiment, reference may be made to the description of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the technical solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/electronic device and method can be implemented in other ways. For example, the above-described apparatus/electronic device embodiments are merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, all or part of the flow in the method according to the above embodiments may be implemented by a computer program, which may be stored in a computer readable storage medium and used by a processor to implement the steps of the method and apparatus embodiments. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, read-Only Memory (ROM), random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like.
The above embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may be modified or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. A method of obtaining pipeline corrosion data, comprising:
obtaining a plurality of typical corrosive soils, wherein the typical corrosive soils have a corrosive effect on the pipes;
burying a plurality of pipe samples for each of said typical corrosive soils;
sequentially taking out pipeline samples from the typical corrosive soils according to a preset time node;
for each typical corrosive soil, respectively acquiring a maximum corrosion rate and a plurality of soil corrosion factors as pipeline corrosion data, wherein the soil corrosion factors are used for representing the factors of the pipeline laying ground influencing the pipeline corrosion rate, and the maximum corrosion rate is acquired based on a plurality of pipeline samples taken from the typical corrosive soil.
2. The method for acquiring pipeline corrosion data according to claim 1, wherein the embedding a plurality of pipeline samples for each of the typical corrosive soils includes:
the pipeline sample material matrix is a metal pipe, a zinc coating is sprayed on the outer surface of the metal pipe, and an organic finishing coating is sprayed on the zinc coating;
and after the pipeline samples are subjected to edge sealing and drying, the pipeline samples are axially and horizontally placed at the bottom of the test pit according to the type sequence, the intervals between the pipeline samples and the edge of the test pit are larger than a preset value, and soil is paved on the pipeline samples, so that the soil around the pipeline samples is compact and has no gaps.
3. The pipeline corrosion data acquisition method according to claim 1 or 2, wherein intervals between a plurality of predetermined time nodes are set in a stepwise increasing rule;
obtaining a maximum corrosion rate from a plurality of pipe specimens taken from the typical corrosive soil comprises:
for each pipeline sample, acquiring corrosion pit depth data of the deepest corrosion pit;
converting the plurality of etch pit depth data into maximum etch pit rate data corresponding to the time node according to a predetermined time node;
constructing a corrosion rate expectation function according to a plurality of time nodes and a plurality of maximum corrosion pit rate data corresponding to the time nodes, wherein the corrosion rate expectation function is as follows:
V＝a·t b
where V is the corrosion rate of a typical corrosive soil area, a is the forward constant, b is the backward constant, and t is the time.
4. The method for acquiring corrosion data of pipelines according to claim 3, wherein after said acquiring a maximum corrosion rate and a plurality of soil corrosion factors for each of said typical corrosive soils respectively, comprises:
acquiring a measured corrosion rate of an actually buried pipeline and a plurality of corrosion factors of actually buried soil;
normalizing according to the corrosion rate expectation function and the measured corrosion rate of the actually buried pipeline to obtain the corrosion rate of the actually buried pipeline;
the corrosion rate of the actually buried pipeline and a plurality of corrosion factors of soil of the actually buried pipeline are added to the pipeline corrosion data.
5. The method of obtaining pipeline corrosion data according to claim 4, wherein said normalizing according to said expected corrosion rate function and a measured corrosion rate of the actual buried pipeline to obtain the corrosion rate of the actual buried pipeline comprises:
normalizing according to the corrosion rate expectation function, a first formula and the measured corrosion rate of the actually buried pipeline to obtain the corrosion rate of the actually buried pipeline, wherein the first formula is as follows:
in the formula, V a For normalized corrosion rate, t, of the actual buried pipeline 1 Time node, V, for sampling of the actual buried pipeline 1 For the actual buried pipeline at t 1 Measurement of the Corrosion Rate, V, of a time node 0 Is the long-term corrosion rate of the areas with strong corrosive soil.
6. A method for constructing a pipeline corrosion rate prediction model is characterized by comprising the following steps:
acquiring a plurality of corrosion data, wherein the corrosion data comprises a long-term average corrosion rate and a plurality of soil corrosion factors, and the plurality of corrosion data is acquired by using the pipeline corrosion data acquisition method according to any one of claims 1-5;
constructing an initial model, wherein the initial model is constructed based on a neural network;
dividing the plurality of corrosion data into a training data set and a testing data set;
training: inputting a plurality of corrosion data of the training data set into the initial model to train the initial model until the output error of the initial model is lower than a threshold value;
and inputting a plurality of corrosion data of the test data set into the initial model, obtaining a test error of the initial model, and if the test error is higher than a threshold value, adjusting the structure of the initial model and skipping to the training step.
7. The method for constructing the pipeline corrosion rate prediction model according to claim 6, wherein the neural network model is an Elman neural network model, and comprises the following steps: the device comprises an input layer, a hidden layer, an output layer and a receiving layer;
the number of the nodes of the input layer is the same as the number of the plurality of soil corrosion factors of the corrosion data;
the output layer includes a long term average corrosion rate output node having an activation function of:
f(x)＝kx+d
wherein f (x) is an activation function of the long-term average corrosion rate output node, k is a weight coefficient, and d is a first bias;
determining the number of nodes of the hidden layer according to a first formula, the number of nodes of the input layer and the number of nodes of the output layer, wherein the first formula is as follows:
S＝(a 0 +b 0 )*0.4+c
wherein S is the number of hidden nodes, a 0 As the number of nodes of the input layer, b 0 C is the number of output layer nodes, and c is the second bias;
the adjusting the structure of the initial model comprises: increasing or decreasing the number of nodes of the hidden layer.
8. A device for constructing a pipeline corrosion rate prediction model is characterized by comprising:
a corrosion data acquisition module for acquiring a plurality of corrosion data, wherein the corrosion data includes an average corrosion rate over a long period of time and a plurality of soil corrosion factors, and the plurality of corrosion data is acquired by the pipeline corrosion data acquisition method according to any one of claims 1 to 5;
the initial model building module is used for building an initial model, wherein the initial model is built on the basis of a neural network;
a data classification module for classifying the plurality of corrosion data into a training data set and a test data set;
a training module for implementing the training steps: inputting a plurality of corrosion data of the training data set into the initial model to train the initial model until the output error of the initial model is lower than a threshold value;
and (c) a second step of,
and the testing module is used for inputting a plurality of corrosion data of the testing data set into the initial model, acquiring the testing error of the initial model, and if the testing error is higher than a threshold value, adjusting the structure of the initial model and skipping to the training step.
9. An electronic device comprising a memory and a processor, the memory having stored therein a computer program operable on the processor, wherein the processor, when executing the computer program, implements the steps of the method as claimed in any of claims 6 to 7 above.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 6 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210987019.4A CN115482882A (en) | 2022-08-17 | 2022-08-17 | Pipeline corrosion data acquisition and pipeline corrosion rate prediction model construction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210987019.4A CN115482882A (en) | 2022-08-17 | 2022-08-17 | Pipeline corrosion data acquisition and pipeline corrosion rate prediction model construction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115482882A true CN115482882A (en) | 2022-12-16 |
Family
ID=84421937
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210987019.4A Pending CN115482882A (en) | 2022-08-17 | 2022-08-17 | Pipeline corrosion data acquisition and pipeline corrosion rate prediction model construction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115482882A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20230220957A1 (en) * | 2022-01-11 | 2023-07-13 | Saudi Arabian Oil Company | Digitalization and automation of corrosion coupon analysis with a predictive element |
-
2022
- 2022-08-17 CN CN202210987019.4A patent/CN115482882A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20230220957A1 (en) * | 2022-01-11 | 2023-07-13 | Saudi Arabian Oil Company | Digitalization and automation of corrosion coupon analysis with a predictive element |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Seghier et al. | Advanced intelligence frameworks for predicting maximum pitting corrosion depth in oil and gas pipelines | |
| Ji et al. | Probabilistic physical modelling of corroded cast iron pipes for lifetime prediction | |
| Mateo et al. | Profile or group discriminative techniques? Generating reliable species distribution models using pseudo‐absences and target‐group absences from natural history collections | |
| Shiri et al. | Estimation of daily suspended sediment load by using wavelet conjunction models | |
| Heidary et al. | A review of data-driven oil and gas pipeline pitting corrosion growth models applicable for prognostic and health management | |
| Blasone et al. | Parameter estimation in distributed hydrological modelling: comparison of global and local optimisation techniques | |
| CN113687433B (en) | Bi-LSTM-based magnetotelluric signal denoising method and system | |
| Wang et al. | Time dependent reliability analysis for cast iron pipes subjected to pitting corrosion | |
| Vantas et al. | Rainfall erosivity and its estimation: Conventional and machine learning methods | |
| CN115482882A (en) | Pipeline corrosion data acquisition and pipeline corrosion rate prediction model construction method | |
| Garg | Inductive group method of data handling neural network approach to model basin sediment yield | |
| CN110334478A (en) | Machinery equipment abnormality detection model building method, detection method and model | |
| Nor et al. | Radial basis function modeling of hourly streamflow hydrograph | |
| CN109632942B (en) | Inversion method of pipeline defect size based on ensemble learning | |
| Guo et al. | Simulation of a leak's growth process in water distribution systems based on growth functions | |
| Du et al. | Deeppipe: Theory-guided prediction method based automatic machine learning for maximum pitting corrosion depth of oil and gas pipeline | |
| Okeya et al. | Locating pipe bursts in a district metered area via online hydraulic modelling | |
| Xue et al. | Application of a probabilistic neural network for liquefaction assessment | |
| Javdanian | Field data-based modeling of lateral ground surface deformations due to earthquake-induced liquefaction | |
| CN115545112A (en) | Method for automatically identifying and processing large amount of sewage real-time automatic monitoring data | |
| CN115470693A (en) | Method and device for predicting service life of metal pipeline, electronic equipment and storage medium | |
| CN114091883A (en) | Method, device, medium and equipment for predicting well leakage risk horizon before drilling | |
| Tayfur et al. | Genetic algorithm-based discharge estimation at sites receiving lateral inflows | |
| Tran | Sensitivity of Markov model to different sampling sizes of condition data | |
| Xiang et al. | Corrosion Growth Modeling by Learning a Dynamic Bayesian Network From Multiple In-Line Inspection Data |
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 |