CN114330131B - Method for rapidly determining abrasive particle water jet preventive grinding steel rail depth - Google Patents
Method for rapidly determining abrasive particle water jet preventive grinding steel rail depth Download PDFInfo
- Publication number
- CN114330131B CN114330131B CN202111657513.6A CN202111657513A CN114330131B CN 114330131 B CN114330131 B CN 114330131B CN 202111657513 A CN202111657513 A CN 202111657513A CN 114330131 B CN114330131 B CN 114330131B
- Authority
- CN
- China
- Prior art keywords
- data
- rail
- steel rail
- loss
- preventive
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Machines For Laying And Maintaining Railways (AREA)
Abstract
The invention discloses a method for rapidly determining the depth of a grinding particle water jet preventive grinding steel rail, which comprises the following steps: the method comprises the steps that bearing data of a track to be polished are obtained through a remote operation and maintenance system of the track; acquiring steel rail loss data of abrasive particle water jet preventive grinding steel rails; establishing a track bearing and rail loss data set according to the track bearing data and the rail loss data; calculating a track bearing and steel rail loss data set by adopting a convolutional neural network algorithm and obtaining a determination model of the track bearing and steel rail loss data; collecting steel rail loss and preventive polishing depth samples, and establishing steel rail loss data and preventive polishing depth data sets; calculating the data set obtained in the previous step by adopting a convolutional neural network algorithm and obtaining a determination model of the track loss degree and the preventive grinding depth of the steel rail; outputting the preventive polishing depth of the steel rail; the method is easy to realize, and the pre-polishing depth can be accurately determined, so that the steel rail is polished and repaired once in place.
Description
Technical Field
The invention belongs to the technical field of grinding particle water jet steel rail grinding, and particularly relates to a method for quickly determining the depth of a grinding particle water jet preventive grinding steel rail.
Background
The railway transportation industry in China is very developed, and the running speed and the loading capacity of the train are continuously improved. The railway transportation brings great economic benefit to China, meanwhile, the problems of fatigue, abrasion and the like of the steel rail are increasingly serious, the service life of the steel rail is continuously shortened due to huge passenger load and cargo load pressure, the train wheel rail and the steel rail are in a composite contact state of rolling contact and sliding contact, and the stress of the steel rail is complex and variable; however, the rail polishing repair can improve the service life of the rail, ensure the running quality of the train, and ensure the safety of the rail while reducing the running cost of the rail.
Compared with the traditional grinding wheel grinding mode, the high-pressure abrasive particle water jet grinding and repairing method has the advantages of high processing efficiency, low cost and good working environment, belongs to a green rail grinding mode, and is hopeful to replace the traditional multi-grinding wheel enveloping rail grinding method. The high-pressure abrasive particle water jet technology is to mix high-pressure water with a certain quantity and quality of abrasive materials after being pressurized by a pump and then jet the mixture from a specific nozzle to form a high-speed high-energy jet beam. Rail grinding is divided into three modes, pre-grinding, preventive grinding and repair grinding. The preventive polishing is to polish the steel rail periodically to repair the depth of the rail head and prevent the loss of contact fatigue, wave abrasion, cracks and the like of the steel rail.
The existing method for determining the preventive grinding depth of the steel rail mainly comprises expert experience determination, instrument scanning determination, template determination, machine vision determination and the like, for example, the surface quality of the steel rail of a section to be preventive ground is detected in real time in the field through a profile detection device, or a worker visually observes the surface quality, and after the data of the surface steel rail quality are obtained, the expert can determine the preventive grinding depth through experience; although providing great convenience for determining the pre-sanding depth, the problems of low precision and low efficiency still remain.
In addition, CN201721007888.7 proposes a rapid determination instrument for the polishing amount of railway rails, which detects rail profile information through probes, and compares the rail profile information with a standard rail profile library to obtain the polishing amount of rails; however, in the patent, the polishing amount is determined by measuring by an instrument, so that the same position of the instrument for clamping the steel rail each time cannot be ensured, and the long section steel rail is not easy to measure by adopting a magnet to connect the steel rail; CN201511019882.7 mentions a method and a device for calculating the polishing depth of a steel rail polishing head, and repairs the steel rail by presetting the polishing depth and then polishing, but does not mention the origin of the steel rail pre-polishing depth data. CN109840907B mentions a rail abrasion detection method based on deep learning, measures rail abrasion through a sensor, trains a test sample by using a deep learning frame, builds a model, but does not mention polishing depth values after determining the rail abrasion amount, and does not relate to the utilization of bearing data of the rail.
Therefore, a method for rapidly determining the depth of the abrasive particle water jet preventive grinding steel rail is urgently needed, and the problems that in the prior art, the surface quality of the steel rail in the grinding process is checked by manpower, and the pre-grinding depth of the steel rail is determined empirically, so that the accuracy is low and the efficiency is low are solved.
Disclosure of Invention
Aiming at the defects or improvement demands of the prior art, the invention provides a method for rapidly determining the depth of a grinding particle water jet preventive grinding steel rail, and a rail bearing and steel rail loss data sample library is established by collecting the steel rail loss degree in the preventive grinding period of a rail section and bearing data samples of the section of the rail; training data in a track bearing and steel rail loss data sample library by a convolutional neural network learning method to obtain a determining model of the track bearing and steel rail loss degree; the method comprises the steps of establishing a steel rail loss and preventive polishing depth data sample base by collecting polishing depth data corresponding to steel rails with different wear degrees in the preventive polishing process of the steel rails on a track section, training a data set in the steel rail loss and preventive polishing depth data sample base by a convolutional neural network learning method, and obtaining a determination model of the track loss degree and the preventive polishing depth of the steel rails; the method is easy to realize by obtaining two calculation models through two times of convolutional neural network analysis, obtains track bearing in the polishing period by inquiring road section information of the steel rail and combining a remote operation and maintenance system of the track under the daily maintenance period of the steel rail, inputs a determination model of the track bearing and the steel rail loss degree to obtain the loss degree of the steel rail, inputs a determination model of the track loss degree and the steel rail preventive polishing depth to output and obtain the steel rail preventive polishing depth; the invention provides a new thought for determining the depth of the high-pressure abrasive particle water jet preventive grinding steel rail, the grinding efficiency of the steel rail can be further improved under the use condition of the high-pressure abrasive particle water jet preventive grinding steel rail, when the sample cases in the sample database are enough, the accuracy of a model obtained through training can be further improved, the loss degree of the steel rail can be accurately output, and the pre-grinding depth is determined, so that the grinding repair of the steel rail can be completed at one time without grinding; the method can solve the problems that in the prior art, the surface quality of the steel rail is checked and polished by manpower, and the pre-polishing depth of the steel rail is determined empirically, so that the accuracy is low and the efficiency is low.
In order to achieve the above object, the present invention provides a method for rapidly determining the depth of a steel rail for abrasive particle water jet preventive grinding, comprising the steps of: the method comprises the following steps:
s1: position information of a rail where the rail to be prophylactically polished is located is determined, and rail bearing data of the rail are inquired and obtained through a remote operation and maintenance system of the rail;
s2: acquiring rail loss data in the process of preventive grinding of the rails on the track by a high-pressure abrasive particle water jet;
s3: establishing a track bearing and rail loss data set according to the track bearing data and the rail loss data;
s4: calculating the track bearing and steel rail loss data set by adopting a convolutional neural network algorithm, and obtaining a determination model of the track bearing and steel rail loss data;
s5: repeating the steps S1-S4, collecting rail loss data and preventive polishing depth samples of the track, and establishing a rail loss data and preventive polishing depth data set;
s6: performing data division, model training, parameter adjustment, effect verification and initialization output on the steel rail loss data and the preventive grinding depth data set by adopting a convolutional neural network algorithm, and finally obtaining a determination model of the rail loss degree and the preventive grinding depth of the steel rail;
s7: and outputting the preventive polishing depth of the track steel rail.
Further, the structure of the convolutional neural network comprises an input layer, a hidden layer and an output layer;
the number of layers of the input layer is determined by the type of steel rail bearing data in the data set of the rail bearing and steel rail loss, and the number of hidden neurons in the hidden layer is determined by the factor level in the data set of the rail bearing and steel rail loss.
Further, the obtaining of the determining model of the track bearing and rail loss data in step S4 further includes the following steps:
s41: dividing the track bearing and steel rail loss database into training set data, test set data and verification set data;
s42: training the training set data to obtain an initialization model of track bearing and steel rail loss data, continuously training the initialization model through the training set data, and outputting a preliminary result model through a classifier after the iteration number is reached;
s43: and carrying out parameter adjustment on the preliminary result model through the Loss function curve of the test set data, and using the verification set data to verify the effect and performance of the model after the adjustment is optimal in effect, so as to finally obtain a determination model of the track bearing and steel rail Loss data.
Further, the parameters of the initialization model in step S42 include the number of hidden layers, the number of hidden layer neurons, the activation function, the optimization function, and the number of neurons of the designated input/output.
Further, the proportions of the training set data, the test set data and the verification set data in the step S41 are divided into 6:1:3.
further, the activation function in step S42 is a ReLU function.
Further, the determining model of the track loss degree and the steel rail preventive grinding depth obtained in the step S6 further includes: and (6): 1:3 dividing the steel rail loss data and the preventive grinding depth data set into training set data, test set data and verification set data according to the proportion; the method of steps S41 to S43 is repeated.
Further, the track bearing data in step S1 includes a train type, a carrying mass, a train running speed, a passing train number, a brake section, and an acceleration section.
Further, the rail loss data in step S2 includes flaking, fatigue crack, crushing, wave grinding, light band condition, side grinding, fatliquoring, wave grinding, fish scale loss.
In general, the above technical solutions conceived by the present invention, compared with the prior art, enable the following beneficial effects to be obtained:
(1) According to the method for rapidly determining the abrasive particle water jet preventive grinding steel rail depth, a rail bearing and steel rail loss data sample library is established by collecting steel rail loss degree in a preventive grinding period of a rail section and bearing data samples of the section of rail; training data in a track bearing and steel rail loss data sample library by a convolutional neural network learning method to obtain a determining model of the track bearing and steel rail loss degree; the method comprises the steps of establishing a steel rail loss and preventive polishing depth data sample base by collecting polishing depth data corresponding to steel rails with different wear degrees in the preventive polishing process of the steel rails on a track section, training a data set in the steel rail loss and preventive polishing depth data sample base by a convolutional neural network learning method, and obtaining a determination model of the track loss degree and the preventive polishing depth of the steel rails; the method is easy to realize by obtaining two calculation models through two times of convolutional neural network analysis, obtains track bearing in the polishing period by inquiring road section information of the steel rail and combining a remote operation and maintenance system of the track under the daily maintenance period of the steel rail, inputs a determination model of the track bearing and the steel rail loss degree to obtain the loss degree of the steel rail, inputs a determination model of the track loss degree and the steel rail preventive polishing depth to output and obtain the steel rail preventive polishing depth; the invention provides a new thought for determining the depth of the high-pressure abrasive particle water jet preventive grinding steel rail, the grinding efficiency of the steel rail can be further improved under the use condition of the high-pressure abrasive particle water jet preventive grinding steel rail, when the sample cases in the sample database are enough, the accuracy of a model obtained through training can be further improved, the loss degree of the steel rail can be accurately output, and the pre-grinding depth is determined, so that the grinding repair of the steel rail can be completed at one time without grinding; the method can solve the problems that in the prior art, the surface quality of the steel rail is checked and polished by manpower, and the pre-polishing depth of the steel rail is determined empirically, so that the accuracy is low and the efficiency is low.
(2) According to the method for rapidly determining the depth of the abrasive particle water jet preventive grinding steel rail, a convolutional neural network analysis method is utilized to conduct research and analysis on the relationship between the track bearing and the track loss in the preventive grinding process of the steel rail and the relationship between the track loss and the preventive grinding depth, a device or workers do not need to conduct field detection to obtain surface loss data of the steel rail, a sample library is established, a model is generated through convolutional neural network analysis training, and the loss degree of the steel rail can be obtained by inputting the track bearing; the method omits the detection work of the surface quality of the steel rail in the polishing process, saves manpower and material resources, simultaneously takes a large number of data samples as data support, and has higher accuracy of the analyzed result.
Drawings
FIG. 1 is a flow chart of a method for rapidly determining the depth of a rail for abrasive particle water jet preventive grinding in accordance with an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a process of track-bearing sample establishment according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart of rail loss sample establishment in an embodiment of the present invention;
FIG. 4 is a schematic flow chart of the track bearing and rail loss sample data set establishment according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a convolutional neural network according to an embodiment of the present invention;
FIG. 6 is a schematic flow chart of the creation of a rail loss and preventative grinding depth sample data set according to an embodiment of the present invention;
fig. 7 is a schematic view of the preventive grinding depth of the rail according to the embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention. In addition, the technical features of the embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
As shown in fig. 1-7, the present invention provides a method for rapidly determining the depth of a steel rail for abrasive particle water jet preventive grinding, comprising the steps of:
s1: position information of a rail where the rail to be prophylactically polished is located is determined, and rail bearing data of the rail are inquired and obtained through a remote operation and maintenance system of the rail; specifically, position information of a track where a rail to be prophylactically polished is located is clarified, track bearing data of the section of track in the polishing period is queried from a remote operation and maintenance system of the track, and the track bearing data comprises train type, carrying quality, train running speed, number of passing trains, braking sections, accelerating sections and the like (shown in fig. 2);
s2: acquiring rail loss data in the process of preventive grinding of the rails on the track by a high-pressure abrasive particle water jet; specifically, steel rail loss data corresponding to the bearing of the rail in the process of collecting high-pressure abrasive particle water jet preventive grinding of the steel rail (shown in figure 3); the steel rail loss data comprise flaking, fatigue crack, crushing, wave grinding, light band condition, side grinding, fat edge, wave grinding, fish scale loss and the like;
s3: establishing a track bearing and rail loss data set according to the track bearing data and the rail loss data; specifically, the track bearing data and the steel rail loss data are used as bearing and loss samples of the steel rail, and a track bearing and steel rail loss data set is established (shown in fig. 4); as shown in fig. 3 and 4, where m is the train mass, v is the train speed, and a is the train acceleration;
s4: calculating the track bearing and steel rail loss data set by adopting a CNN convolutional neural network algorithm, and obtaining a determination model of the track bearing and steel rail loss data; the method specifically comprises the following steps:
s41: dividing the track bearing and steel rail loss database into training set data, test set data and verification set data; specifically, a CNN convolutional neural network algorithm is employed at 6:1:3 dividing the track bearing and steel rail loss database into training set data, test set data and verification set data according to the proportion; the convolutional neural network comprises an input layer, a hidden layer and an output layer (shown in fig. 5), wherein the input layer number is determined by the type of bearing information of steel rails in a data set, the number of hidden neurons is determined by the factor level in a data sample, and an activation function selected by the method is a ReLU function, and the formula is as follows:
f ReLU ＝max(0，z) (1)
wherein f ReLU To activate the function, z is an input dataset parameter.
When the convolutional neural network conducts forward, the calculation formula of the neuron is as follows:
y k ＝f(u k -b k ) (4)
wherein X is i Represents the i-th value, W, in the input data Ki Representing a weight associated with the i-th input quantity; u (u) k Representing a weighted sum of all input variables; b k Is a threshold value; f (·) is the activation function; y is k Is the output of the neural network;
s42: training by using the training set data to obtain an initialization model of track bearing and steel rail loss data; the initialization model parameters comprise the number of hidden layers, the number of hidden layer neurons, an activation function, an optimization function, the number of neurons of appointed input and output and the like; specifically, through training the training set model, after the number of iterations is reached, model training is completed, at this time, the weights among neurons are fixed, a result is output through a classifier softmax, a preliminary result model is formed, at this time, the class of the model with the largest weight belongs to, and the loss function of the classifier softmax is as shown in formula (5):
Loss＝-∑y i ln a i (5)
wherein the classifier is a regression model softmax, y i For output, a i Is a function characteristic parameter;
s43: and adjusting model parameters of the preliminary result model through a function loss curve of the test set data, and verifying the effect and performance of the model by using the verification set data after adjusting to the optimal effect, so as to finally obtain a determination model of the track bearing and steel rail loss data.
S5: repeating the steps S1-S4, and collecting polishing depth data corresponding to the rails with different abrasion degrees in the rail preventive polishing process to serve as rail loss data and preventive polishing depth samples of the rail, and establishing a rail loss data and preventive polishing depth data set (shown in figure 6);
s6: performing data division, model training, parameter adjustment, effect verification and initialization output on the steel rail loss data and the preventive grinding depth data set by adopting a convolutional neural network algorithm, and finally obtaining a determination model of the rail loss degree and the preventive grinding depth of the steel rail; specifically, at 6:1:3 dividing the steel rail loss data and the preventive grinding depth data set into training set data, test set data and verification set data according to the proportion; repeating the method of the steps S41-S43, and obtaining a determination model of the track loss degree and the preventive grinding depth of the steel rail after model training, parameter adjustment, effect verification and initialization output;
s7: and outputting the preventive grinding depth of the track steel rail (shown in figure 7).
The invention provides a principle of a method for rapidly determining the depth of a grinding steel rail by abrasive particle water jet preventive grinding: the invention provides a new thought for determining the depth of the high-pressure abrasive particle water jet preventive grinding steel rail, and a rail bearing and steel rail loss data sample library is established by collecting the steel rail loss degree in the preventive grinding period of a rail section and the bearing data sample of the section of the rail; training data in a track bearing and steel rail loss data sample library by a convolutional neural network learning method to obtain a determining model of the track bearing and steel rail loss degree; the method comprises the steps of establishing a steel rail loss and preventive polishing depth data sample base by collecting polishing depth data corresponding to steel rails with different wear degrees in the preventive polishing process of the steel rails on a track section, training a data set in the steel rail loss and preventive polishing depth data sample base by a convolutional neural network learning method, and obtaining a determination model of the track loss degree and the preventive polishing depth of the steel rails; the method is easy to realize by obtaining two calculation models through two times of convolutional neural network analysis, obtains track bearing in the polishing period by inquiring road section information of the steel rail and combining a remote operation and maintenance system of the track under the daily maintenance period of the steel rail, inputs a determination model of the track bearing and the steel rail loss degree to obtain the loss degree of the steel rail, inputs a determination model of the track loss degree and the steel rail preventive polishing depth to output and obtain the steel rail preventive polishing depth;
the method comprises the steps of determining a section of steel rail pre-polished by high-pressure abrasive particle water jet, obtaining position information of the section of rail, inquiring rail bearing data in a certain section of preventive polishing period by combining with a remote operation and maintenance system of a railway, inputting the rail bearing data into a trained determination model of rail bearing and rail loss degree, analyzing the rail bearing information in a period of time, inputting a neural network training model, obtaining the rail loss degree through sample comparison analysis and empirical expression analysis calculation, and then comparing with a rail loss and preventive polishing depth database to further determine the rail preventive polishing depth; according to the invention, under the condition of polishing the steel rail by the high-pressure abrasive particle water jet, the steel rail polishing efficiency can be further improved, when the number of sample cases in the sample database is enough, the accuracy of a model obtained through training can be further improved, the loss degree of the steel rail can be accurately output, and the pre-polishing depth is determined, so that the steel rail can be polished and repaired in place once without polishing; the invention has a certain innovation in the field of steel rail polishing, the relation between the track bearing and the track loss in the process of preventive polishing of the steel rail and the relation between the track loss and the preventive polishing depth are researched and analyzed by using a convolutional neural network analysis method, the surface loss data of the steel rail is acquired without the need of devices or workers for in-situ detection, a model is generated by establishing a sample library and analyzing and training through the convolutional neural network, and the loss degree of the steel rail can be obtained by inputting the track bearing; the method omits the detection work of the surface quality of the steel rail in the polishing process, saves manpower and material resources, simultaneously takes a large number of data samples as data support, and has higher accuracy of the analyzed result.
It will be readily appreciated by those skilled in the art that the foregoing description is merely a preferred embodiment of the invention and is not intended to limit the invention, but any modifications, equivalents, improvements or alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (7)
1. A method for rapidly determining the depth of a steel rail for abrasive particle water jet preventive grinding, comprising the steps of:
s1: position information of a rail where the rail to be prophylactically polished is located is determined, and rail bearing data of the rail are inquired and obtained through a remote operation and maintenance system of the rail;
s2: acquiring rail loss data in the process of preventive rail grinding by high-pressure abrasive particle water jet;
s3: establishing a track bearing and rail loss data set according to the track bearing data and the rail loss data;
s4: calculating the track bearing and steel rail loss data set by adopting a convolutional neural network algorithm, and obtaining a determination model of the track bearing and steel rail loss data;
s5: acquiring rail loss data and preventive polishing depth samples of a rail, and establishing a rail loss data and preventive polishing depth data set;
s6: performing data division, model training, parameter adjustment, effect verification and initialization output on the steel rail loss data and the preventive grinding depth data set by adopting a convolutional neural network algorithm, and finally obtaining a determination model of the rail loss degree and the preventive grinding depth of the steel rail;
s7: under the maintenance period of the steel rail, obtaining the track bearing in the current polishing period by inquiring the road section information of the steel rail and combining a remote operation and maintenance system of the track, inputting a determination model of the track bearing and steel rail loss data to obtain the loss degree of the steel rail, and inputting a determination model of the track loss degree and the steel rail preventive polishing depth to obtain the steel rail preventive polishing depth;
the track bearing data in the step S1 comprises a train type, a carrying quality, a train running speed, the number of passing trains, a brake section and an acceleration section;
the rail loss data in step S2 includes flaking, fatigue cracking, crushing, wave grinding, light band conditions, side grinding, fatliquoring, wave grinding, fish scale loss.
2. The method for rapidly determining the depth of a steel rail for abrasive particle water jet preventive grinding according to claim 1, wherein the structure of the convolutional neural network in the step S4 comprises an input layer, a hidden layer and an output layer;
the number of layers of the input layer is determined by the type of steel rail bearing data in the data set of the rail bearing and steel rail loss, and the number of hidden neurons in the hidden layer is determined by the factor level in the data set of the rail bearing and steel rail loss.
3. A method for rapid determination of abrasive particle water jet preventive grinding rail depth according to claim 2, wherein the obtaining of the determination model of rail bearing and rail loss data in step S4 further comprises the steps of:
s41: dividing the track bearing and steel rail loss data set into training set data, test set data and verification set data;
s42: training the training set data to obtain an initialization model of track bearing and steel rail loss data, continuously training the initialization model through the training set data, and outputting a preliminary result model through a classifier after the iteration number is reached;
s43: and carrying out parameter adjustment on the preliminary result model through the Loss function curve of the test set data, and using the verification set data to verify the effect and performance of the model after the adjustment is optimal in effect, so as to finally obtain a determination model of the track bearing and steel rail Loss data.
4. A method for rapidly determining a depth of a rail for abrasive grain water jet preventive grinding according to claim 3, wherein the parameters of the initialization model in step S42 include the number of hidden layers, the number of hidden layer neurons, an activation function, an optimization function, and the number of neurons for designated input and output.
5. A method for rapidly determining abrasive grain water jet preventive grinding rail depth according to claim 3 or 4, wherein the ratio of the training set data, the test set data and the verification set data in step S41 is divided into 6:1:3.
6. a method of rapidly determining the depth of a rail for abrasive grain water jet preventive grinding according to claim 3 or 4, wherein the activation function in step S42 is a ReLU function.
7. A method for rapidly determining the depth of a rail for abrasive particle water jet preventive grinding according to claim 3 or 4, wherein the step S6 of obtaining a model of the degree of track loss and the depth of rail preventive grinding further comprises: and (6): 1:3 dividing the steel rail loss data and the preventive grinding depth data set into training set data, test set data and verification set data according to the proportion;
training the training set data to obtain an initialization model of steel rail loss data and preventive grinding depth data, continuously training the initialization model through the training set data, and outputting a preliminary result model through a classifier after the number of iterations is reached;
and carrying out parameter adjustment on the preliminary result model through the Loss function curve of the test set data, and using the verification set data to verify the effect and performance of the model after the adjustment is optimal in effect, so as to finally obtain a determination model of the steel rail Loss data and the preventive grinding depth data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111657513.6A CN114330131B (en) | 2021-12-30 | 2021-12-30 | Method for rapidly determining abrasive particle water jet preventive grinding steel rail depth |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111657513.6A CN114330131B (en) | 2021-12-30 | 2021-12-30 | Method for rapidly determining abrasive particle water jet preventive grinding steel rail depth |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114330131A CN114330131A (en) | 2022-04-12 |
| CN114330131B true CN114330131B (en) | 2023-07-07 |
Family
ID=81018373
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111657513.6A Active CN114330131B (en) | 2021-12-30 | 2021-12-30 | Method for rapidly determining abrasive particle water jet preventive grinding steel rail depth |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114330131B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114481727A (en) * | 2022-04-18 | 2022-05-13 | 中铁第四勘察设计院集团有限公司 | Intelligent control system and method for unmanned water jet steel rail grinding wagon |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112836272A (en) * | 2021-01-20 | 2021-05-25 | 北京交通大学 | Neural network model-based high-speed railway steel rail profile optimization design method |
| CN113239599A (en) * | 2021-06-15 | 2021-08-10 | 江苏理工学院 | Intelligent tire wear life estimation method and device based on BP neural network |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111733639A (en) * | 2020-06-23 | 2020-10-02 | 中国铁道科学研究院集团有限公司铁道建筑研究所 | Method for replacing damaged sleeper of ballastless track |
| CN112613234A (en) * | 2020-12-24 | 2021-04-06 | 中铁第四勘察设计院集团有限公司 | Pantograph slide plate abrasion trend estimation and service life prediction method and device |
-
2021
- 2021-12-30 CN CN202111657513.6A patent/CN114330131B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112836272A (en) * | 2021-01-20 | 2021-05-25 | 北京交通大学 | Neural network model-based high-speed railway steel rail profile optimization design method |
| CN113239599A (en) * | 2021-06-15 | 2021-08-10 | 江苏理工学院 | Intelligent tire wear life estimation method and device based on BP neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114330131A (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bevan et al. | Development and validation of a wheel wear and rolling contact fatigue damage model | |
| CN114330131B (en) | Method for rapidly determining abrasive particle water jet preventive grinding steel rail depth | |
| CN102253049B (en) | Method for accurately detecting surface quality on line in production process of band steel | |
| CN107179101A (en) | A kind of detection and the evaluation method of wheel face roughness and abrasive particle distribution | |
| CN111809463B (en) | Intelligent steel rail polishing system based on AI method and corresponding polishing method | |
| CN104732003A (en) | Grinding process assessment method based on reliability | |
| CN108274187A (en) | A kind of complex curved surface parts defect repair system and restorative procedure | |
| CN115758289B (en) | Rail wave mill identification method based on multitask learning neural network | |
| CN103630564A (en) | Measure method for residual stress of high-speed train body | |
| Fan et al. | Wheel wear prediction of high-speed train using NAR and BP neural networks | |
| CN114565314B (en) | Digital twinning-based hot rolled steel coil end face quality control system and method | |
| CN117173151B (en) | Visual identification device, method and system for defects of outer surface of long steel | |
| CN111809464B (en) | Intelligent polishing control method for steel rail | |
| Beretta et al. | RAAI Project: Life-prediction and prognostics for railway axles under corrosion-fatigue damage | |
| Kalentiev et al. | Prediction of abrasive weight wear rate using machine learning methods | |
| WO2023124583A1 (en) | Steel rail repair post-treatment strategy making method | |
| CN112464397A (en) | Railway turnout steel rail polishing profile optimization method | |
| RU2502572C2 (en) | Classification of flaws and grinding of forming rolls | |
| Wang et al. | A combined computer vision and image processing method for surface coverage measurement of shot peen forming | |
| Saleh et al. | Detection of air bubbles from tire shearography images | |
| Asplund et al. | Assessment of the data quality of wayside wheel profile measurements. | |
| Protsenko et al. | The wear resistance research of the rail contact surface depending on the grinding process | |
| Liu et al. | Load spectrum compiling and fatigue life estimation of the automobile wheel hub | |
| CN112945976A (en) | Method and device for detecting contact fatigue crack of steel rail | |
| CN117235507B (en) | Method for identifying abnormal state of grinding stone of steel rail grinding wagon |
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 |