CN117991288B - Carriage detection early warning device and method based on laser radar - Google Patents
Carriage detection early warning device and method based on laser radar Download PDFInfo
- Publication number
- CN117991288B CN117991288B CN202410396259.6A CN202410396259A CN117991288B CN 117991288 B CN117991288 B CN 117991288B CN 202410396259 A CN202410396259 A CN 202410396259A CN 117991288 B CN117991288 B CN 117991288B
- Authority
- CN
- China
- Prior art keywords
- carriage
- detection
- data
- crack
- result
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 287
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000004458 analytical method Methods 0.000 claims abstract description 18
- 238000007781 pre-processing Methods 0.000 claims abstract description 18
- 230000002159 abnormal effect Effects 0.000 claims abstract description 10
- 238000010276 construction Methods 0.000 claims abstract description 4
- 238000012545 processing Methods 0.000 claims description 32
- 238000009499 grossing Methods 0.000 claims description 17
- 230000009467 reduction Effects 0.000 claims description 14
- 238000012937 correction Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 8
- 238000009434 installation Methods 0.000 claims description 8
- 230000005856 abnormality Effects 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 4
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000007637 random forest analysis Methods 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 16
- 238000006722 reduction reaction Methods 0.000 description 11
- 238000004422 calculation algorithm Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000010606 normalization Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The application discloses a carriage detection early warning device and method based on a laser radar, comprising the following steps: model construction module: constructing carriage detection channels of carriages of different types; the data acquisition module to be detected: detecting a target carriage to obtain data to be detected; the detection data preprocessing module is used for: preprocessing data to be detected to obtain standard detection data; the first car detection result acquisition module: if the carriage deformation coefficient and the carriage crack coefficient fall into a qualified carriage deformation interval and a qualified carriage crack interval of the carriage detection channel, a target carriage detection channel is obtained, and standard detection data are input into the target carriage detection channel for detection to obtain a first carriage detection result; the abnormal component result acquisition module: performing heterogeneous part analysis according to the first carriage detection result to obtain a heterogeneous part result; the carriage detection result judging module: and comprehensively evaluating the first carriage detection result and the heterogeneous piece result to obtain a final carriage detection result which is qualified or unqualified.
Description
Technical Field
The application relates to the technical field of carriage detection and early warning devices, in particular to a carriage detection and early warning device and method based on a laser radar.
Background
In the field of railway transportation, it is very important to ensure the safety of trains and cars. The traditional carriage detection method cannot meet the high requirement of modern railway transportation on safety. Therefore, more advanced technical means are needed to improve the accuracy and efficiency of detection. Along with the continuous increase of the traffic volume of trains, the requirements for carriage detection are also increasing. The traditional detection method can not meet the detection requirements of large batch and high efficiency. Therefore, there is a need to employ more efficient, more accurate detection methods to improve transportation efficiency. Along with the continuous development of intelligent technology, the intelligent requirements for carriage detection are also higher and higher.
Disclosure of Invention
The carriage detection early warning device and method based on the laser radar can detect and early warn carriages in a large scale with high efficiency and high accuracy.
The technical scheme adopted for solving the technical problems is as follows: a carriage detection early warning device based on laser radar includes:
Model construction module: extracting qualified carriage deformation intervals and qualified carriage crack intervals of carriages of different types from historical carriage detection data, and constructing carriage detection channels of the carriages of different types, wherein the types of processing data detected by the carriage detection channels of each type are the same but the detection indexes are different;
The data acquisition module to be detected: detecting a target carriage by adopting a laser radar device to obtain data to be detected, wherein the data to be detected comprises carriage deformation coefficients and carriage crack coefficients of the target carriage;
The detection data preprocessing module is used for: preprocessing data to be detected to obtain standard detection data;
The first car detection result acquisition module: judging whether the carriage deformation coefficient and the carriage crack coefficient of the standard detection data fall into a qualified carriage deformation interval and a qualified carriage crack interval of the carriage detection channel, if so, obtaining a target carriage detection channel, inputting the standard detection data into the target carriage detection channel for detection to obtain a first carriage detection result, and if not, outputting carriage detection abnormality;
the abnormal component result acquisition module: performing heterogeneous part analysis according to the first carriage detection result to obtain a heterogeneous part result of the target carriage, wherein the heterogeneous part result reflects the loading state of the target carriage;
The carriage detection result judging module: and comprehensively evaluating the first carriage detection result and the heterogeneous piece result to obtain a final carriage detection result which is qualified or unqualified.
Preferably, the laser radar device can scan each part of the target carriage, the installation positions, the number and the installation angles of the laser radar devices of different types of carriages are set according to the type of the carriage, and the laser radar device sets proper detection parameters according to the characteristics of the target carriage required to be detected. These parameters may include the frame rate of the lidar, the scanning frequency, the laser wavelength, etc. Determining the size and shape of a carriage to be detected, calculating the coverage of a required radar, calculating the maximum detection distance, scanning angle, angular resolution, measurement accuracy and the like of the radar, simulating by using a software tool through the data to obtain the optimal installation position, angle and number of the radar, installing the radar according to the obtained data, firmly fixing the radar device, and setting the scanning mode of the radar device such as continuous scanning, triggering scanning and the like and frequency. Ensuring that the data acquisition system is able to receive, store and process the raw output of the radar. After the setting is completed, the laser radar device is started to perform testing, and normal operation of the laser radar device is ensured.
The application also provides a carriage detection early warning method based on the laser radar, which comprises the following steps:
S1: extracting qualified carriage deformation intervals and qualified carriage crack intervals of carriages of different types from historical carriage detection data, and constructing carriage detection channels of the carriages of different types, wherein the types of processing data detected by the carriage detection channels of each type are the same but the detection indexes are different;
s2: detecting a target carriage by adopting a laser radar device to obtain data to be detected, wherein the data to be detected comprises carriage deformation coefficients and carriage crack coefficients of the target carriage;
S3: the data to be detected is preprocessed to obtain standard detection data, and effective data are screened from the data to be detected in preprocessing, so that abnormal values, noise data and the like are eliminated. This may be achieved by setting a threshold, filtering the data, etc. Data normalization: and carrying out normalization processing on the screened effective data to eliminate the influence of data dimension. The data to be detected may be scaled to fall within a smaller interval, such as [0,1] or [ -1,1]. And (3) data smoothing: and smoothing the normalized data to remove noise and fluctuation. It may be implemented by using a filter, a smoothing algorithm, etc. Data normalization: and carrying out standardization processing on the smoothed data so as to enable comparability among different data. Each data may be subtracted by its mean value and divided by the standard deviation to yield normalized data to be detected. Data correction: and correcting the standardized data to be detected according to the pretreatment index. For example, if the preprocessing index includes temperature correction, the temperature factor may be taken into consideration to perform corresponding correction on the data to be detected. Outputting standard detection data: after data preprocessing, the obtained standard detection data can be used for further analysis and processing. These parameters may be used to evaluate the quality of the car, detect anomalies, and the like. Through the step, the data to be detected can be subjected to data preprocessing according to the preprocessing index, and standard detection data can be obtained. The standard detection data can be better used for subsequent analysis and processing, and the accuracy and reliability of carriage detection are improved;
S4: judging whether the carriage deformation coefficient and the carriage crack coefficient of the standard detection data fall into a qualified carriage deformation interval and a qualified carriage crack interval of the carriage detection channel, if so, obtaining a target carriage detection channel, inputting the standard detection data into the target carriage detection channel for detection to obtain a first carriage detection result, and if not, outputting carriage detection abnormality;
S5: performing heterogeneous part analysis according to the first carriage detection result to obtain a heterogeneous part result of the target carriage, wherein the heterogeneous part result reflects the loading state of the target carriage;
S6: and comprehensively evaluating the first carriage detection result and the heterogeneous piece result to obtain a final carriage detection result which is qualified or unqualified.
Preferably, the carriage detection channel comprises an input layer, a processing layer and an output layer, wherein the input layer receives collected data and transmits the collected data to the processing layer, the processing layer processes and predicts the data, and the output layer outputs a detection result.
Preferably, the car types include cars of different load, freight, passenger, and special function types.
Preferably, the preprocessing in step S3 includes:
a1: carrying out data filtering on the data to be detected, and removing noise and interference to obtain anti-interference parameters;
A2: carrying out data noise reduction on the anti-interference parameters, and further removing noise and abnormal values to obtain noise reduction parameters;
a3: carrying out data smoothing treatment on the noise reduction parameters to obtain smoothing parameters;
A4: and carrying out data correction on the smoothing parameters to obtain standard detection data.
Preferably, determining whether the car deformation coefficient of the standard detection data falls within the acceptable car deformation zone of the car detection channel includes: and inputting the carriage deformation coefficient into a height detection operator and a width detection operator of the carriage detection channel, judging whether the height detection result of the target carriage falls into a height threshold value or not, and judging whether the width detection result of the target carriage falls into a width threshold value or not.
Preferably, determining whether the car crack coefficient of the standard detection data falls within the acceptable car crack interval of the car detection channel includes: inputting the carriage crack coefficient into a crack length detection operator, a crack width detection operator, a crack depth detection operator and a crack expansion rate detection operator of a carriage detection channel, judging whether a crack length detection result of a target carriage falls into a crack length threshold value, judging whether the crack width detection result of the target carriage falls into a crack width threshold value, judging whether the crack depth detection result of the target carriage falls into a crack depth threshold value, and judging whether the crack expansion rate detection result of the target carriage falls into a crack expansion rate threshold value.
Preferably, the components of the target carriage are compared and analyzed to obtain the difference and the association between different components to judge the loading state of the target carriage, wherein the loading state comprises overload or unbalanced load.
The application has the following substantial effects:
1. The carriage detection early warning device based on the laser radar obtains data to be detected after data acquisition processing is carried out by adopting the laser radar device suitable for a target carriage, the data to be detected is detected through a target carriage detection channel to obtain a first carriage detection result, first detection is carried out, then heterogeneous piece analysis is carried out according to the first carriage detection result to obtain a heterogeneous piece result of the target carriage, the first carriage detection result and the heterogeneous piece result are comprehensively evaluated to carry out secondary detection, and the final carriage detection result is qualified or unqualified, so that the detection requirements of mass and high efficiency of railway transportation carriage detection are met, and the transportation efficiency and the intelligent level are improved;
2. The carriage detection early warning method based on the laser radar detects and analyzes the size, the crack and the heterogeneous part of the target carriage, comprehensively obtains more accurate carriage detection results, and meets the high requirements of modern railway transportation on safety.
Drawings
FIG. 1 is a block diagram of a device according to a first embodiment of the present application;
fig. 2 is a flowchart of the steps of a second embodiment of the present application.
Detailed Description
The technical scheme of the application is further specifically described by the following specific examples.
Example 1
A carriage detection early warning device based on laser radar, as shown in figure 1, includes:
Model construction module: extracting qualified carriage deformation intervals and qualified carriage crack intervals of carriages of different types from historical carriage detection data, and constructing carriage detection channels of the carriages of different types, wherein the types of processing data detected by the carriage detection channels of each type are the same but the detection indexes are different;
the data acquisition module to be detected: adopt laser radar device to detect the target carriage, obtain to wait to detect data, wait to detect the carriage deformation coefficient and the carriage crack coefficient of data including the target carriage, laser radar device can scan each part in target carriage, and laser radar device's in different grade type carriage mounted position, quantity and installation angle set up according to the carriage type, when setting up the laser radar device in every type carriage, need confirm radar detection parameter, set for the detection parameter: and setting proper detection parameters according to the characteristics of the target carriage to be detected. These parameters may include the frame rate of the lidar, the scanning frequency, the laser wavelength, etc. Determining the size and shape of a carriage to be detected, calculating the coverage of a required radar, calculating the maximum detection distance, scanning angle, angular resolution, measurement accuracy and the like of the radar, simulating by using a software tool through the data to obtain the optimal installation position, angle and number of the radar, installing the radar according to the obtained data, firmly fixing the radar device, and setting the scanning mode of the radar device such as continuous scanning, triggering scanning and the like and frequency. Ensuring that the data acquisition system can receive, store and process the original output of the radar;
The detection data preprocessing module is used for: preprocessing data to be detected to obtain standard detection data;
The first car detection result acquisition module: judging whether the carriage deformation coefficient and the carriage crack coefficient of the standard detection data fall into a qualified carriage deformation interval and a qualified carriage crack interval of the carriage detection channel, if so, obtaining a target carriage detection channel, inputting the standard detection data into the target carriage detection channel for detection to obtain a first carriage detection result, and if not, outputting carriage detection abnormality;
the abnormal component result acquisition module: performing heterogeneous part analysis according to the first carriage detection result to obtain a heterogeneous part result of the target carriage, wherein the heterogeneous part result reflects the loading state of the target carriage;
The carriage detection result judging module: and comprehensively evaluating the first carriage detection result and the heterogeneous piece result to obtain a final carriage detection result which is qualified or unqualified.
The carriage detection early warning device based on the laser radar obtains data to be detected after data acquisition processing is carried out by adopting the laser radar device suitable for a target carriage, the data to be detected is detected through a target carriage detection channel to obtain a first carriage detection result, primary detection is carried out, then heterogeneous piece analysis is carried out according to the first carriage detection result, a heterogeneous piece result of the target carriage is obtained, the first carriage detection result and the heterogeneous piece result are comprehensively evaluated to carry out secondary detection, the final carriage detection result is obtained to be qualified or unqualified, the detection requirements of mass and high efficiency of railway transportation carriage detection are met, and the transportation efficiency and the intelligent level are improved.
Example two
A carriage detection early warning method based on laser radar, as shown in figure 2, comprises the following steps:
S1: extracting qualified carriage deformation intervals and qualified carriage crack intervals of different types of carriages from historical carriage detection data, and constructing carriage detection channels of the different types of carriages, wherein the types of processing data detected by each type of carriage detection channel are the same but the detection indexes are different, and the types of carriages comprise carriages with different loads, freight transportation, passenger transportation and special function types: the historical carriage detection data can be obtained through automatic detection equipment or manual detection, a qualified carriage deformation section is a safety range for representing the shape change of the carriage, the deformation condition of the carriage can be identified from detection records through edge detection, feature point matching or a three-dimensional reconstruction method in image processing, a qualified carriage crack section is a safety range for representing the occurrence of cracks in the carriage surface or structure, the cracks comprise the length, the width, the depth, the expansion rate and the like of the cracks, an image segmentation algorithm can be used for identifying crack areas, and the length and the width of the cracks can be calculated by measuring the pixel number or the morphological characteristics of the areas. For the acquisition of depth information, it is possible to use three-dimensional scanning or laser ranging techniques. The method comprises the steps of using a support vector machine, a random forest or a convolutional neural network as an initial model, training the initial model through an obtained qualified carriage deformation interval and a qualified carriage crack interval, and constructing a carriage detection channel, wherein the carriage detection channel comprises an input layer, a processing layer and an output layer, the input layer receives collected data, the data after pretreatment is transmitted to the processing layer, the initial model is integrated in the processing layer, data processing and prediction are carried out, and the output layer outputs a detection result. Finally, a carriage detection channel is obtained. The car detection channel is applied to all cars.
S2: the laser radar device is adopted to detect a target carriage, data to be detected are obtained, and the data to be detected comprise carriage deformation coefficients and carriage crack coefficients of the target carriage: firstly, ensuring that the installation position and the angle of a laser radar device can ensure that each part of a target carriage can be correctly scanned, generating point cloud data, and analyzing the point cloud data: and analyzing the generated point cloud data, and extracting to-be-detected data related to the target carriage. The data to be detected includes information of the shape, size, position, etc. of the target compartment.
S3: preprocessing the data to be detected to obtain standard detection data: the preprocessing needs to screen effective data from the data to be detected, and eliminates abnormal values, noise data and the like. This may be achieved by setting a threshold, filtering the data, etc. Specific:
A1: firstly, data filtering is carried out on the data to be detected, noise and interference are removed, anti-interference parameters are obtained, and the method can be realized by adopting a digital filter, a smoothing algorithm and the like. In this process, appropriate filter parameters may be set to optimize the filtering effect, such as selecting an appropriate filter type, adjusting the filter order, etc.;
A2: the anti-interference parameters are subjected to data noise reduction, noise and abnormal values are further removed to obtain noise reduction parameters, and the noise reduction parameters can be realized by adopting a statistical method, a clustering algorithm, a deep learning-based method and the like. In the process, a proper noise reduction method can be selected by combining the actual application scene and the data characteristics, and the noise reduction effect is evaluated and adjusted;
A3: and carrying out data smoothing treatment on the noise reduction parameters to obtain smoothing parameters so as to reduce fluctuation and mutation of the data, wherein the smoothing parameters can be realized by adopting methods such as a moving average method, a low-pass filter, wavelet transformation and the like. In this process, parameters of the smoothing process may be adjusted to obtain an optimal processing effect, such as selecting an appropriate window size, filter type, etc.;
a4: and carrying out data correction on the smoothing parameters to obtain more accurate and reliable standard detection data. Can be realized by adopting a model-based method, regression analysis, artificial intelligence and other technologies. In the process, a correction model can be constructed, training is performed by using priori knowledge and historical data, and correction results are verified and adjusted. Standard detection data are obtained: after data filtering, noise reduction, smoothing and correction, the obtained standard detection parameters can be used for subsequent carriage detection and analysis.
The parameters can reflect the actual conditions of the carriages and provide important reference bases for safe operation of the train. Through the steps, data filtering, noise reduction, smoothing and correction can be performed on the detection data to be processed, so that more accurate and reliable standard detection data can be obtained. The data can be better used for carriage detection and analysis, and the safety and reliability of train operation are improved.
S4: judging whether the carriage deformation coefficient and the carriage crack coefficient of the standard detection data fall into a qualified carriage deformation interval and a qualified carriage crack interval of the carriage detection channel, if so, obtaining a target carriage detection channel, inputting the standard detection data into the target carriage detection channel to detect to obtain a first carriage detection result, and if not, outputting carriage detection abnormality: specifically, standard detection data is input into a carriage detection channel as input data of carriage detection, whether a height detection result of a target carriage falls into a height threshold value is judged through a height detection operator and a width detection operator of the carriage detection channel, whether a width detection result of the target carriage falls into a width threshold value is judged, whether a crack length detection result of the target carriage falls into a crack length threshold value is judged through a crack length detection operator, a crack width detection operator, a crack depth detection operator and a crack expansion rate detection operator of the carriage detection channel, whether a crack width detection result of the target carriage falls into a crack width threshold value is judged, whether a crack depth detection result of the target carriage falls into a crack depth threshold value is judged, whether a crack expansion rate detection result of the target carriage falls into a crack expansion rate threshold value is judged, if yes, the target carriage detection channel is obtained, and the standard detection data is input into the target carriage detection channel to detect to obtain a first carriage detection result, if no, and carriage detection abnormality is output.
The target car detection channel further analyzes and processes the standard detection data. The first car detection result may include a quality assessment of the car, an abnormal situation, and the like. Outputting a detection report: and summarizing and sorting the generated detection results to form a first car detection result report. The report should include information such as the number of the target car, the detection time, the detection parameters, the detection results, etc. These results can be used to evaluate and control the quality of the car to ensure safe operation of the train. Meanwhile, the detection result can be further analyzed and processed as required, such as comparison with historical data, trend prediction and the like.
It should be noted that: the height detection operator, the width detection operator, the crack length detection operator, the crack width detection operator, the crack depth detection operator, and the crack propagation rate detection operator may be a preset algorithm or model for determining whether the monitored data of the carriage reach a preset threshold interval.
S5: performing heterogeneous part analysis according to the first carriage detection result to obtain a heterogeneous part result of the target carriage, wherein the heterogeneous part result reflects the loading state of the target carriage: the heterocomponent analysis refers to comparing and analyzing the various components of the cabin to understand the differences and associations between the different components. In this process, key parameters in the first car detection result, such as car height, width, weight, etc., may be extracted and sorted, and these parameters are classified and compared. Heterogeneous results were obtained: by performing the heterogeneous analysis on the target car, heterogeneous results of the target car can be obtained. These results may include differences between different components, anomalies, and so forth. Judging whether overload or unbalanced load is satisfied: based on the heterogeneous product result, the loading state of the vehicle cabin can be further obtained. The loading state can reflect the cargo distribution, weight and the like in the carriage. In this process, it can be judged whether the loading state of the car satisfies the condition of overload or unbalanced load. Overload refers to the fact that the weight of goods in a carriage exceeds a specified limit, and unbalanced load refers to the fact that the goods in the carriage are unevenly distributed to cause the inclination or the deviation of the carriage. Judging the safety coefficient of the carriage: according to the carriage loading state, the safety coefficient of the carriage can be further judged. The safety factor may reflect the stability and reliability of the car in the loaded state. If the loading state of the carriage is poor, the safety coefficient of the carriage may be reduced, and the running risk of the train is increased.
S6: comprehensively evaluating the first carriage detection result and the heterogeneous piece result to obtain a final carriage detection result which is qualified or unqualified: the first carriage detection result and the heterogeneous part result are respectively assigned with weights, the weights are different according to different carriage types, the final carriage detection result is output as being qualified or unqualified after comprehensive evaluation, and when the carriage detection is unqualified, an early warning instruction can be executed and braking processing is carried out so as to ensure the running safety of the train. Sending out an early warning instruction: when the carriage detection result is unqualified, the system can automatically or manually send out an early warning instruction. The early warning instruction can be prompted in various modes such as acousto-optic and the like so that related personnel can find and process problems in time. Executing a braking process: the system may automatically or manually initiate the braking process while the warning command is issued. The braking process may include emergency braking or retarded braking of the car to reduce the train speed and ensure that the train does not continue into an unacceptable car. Notifying relevant personnel: when the detection result is unqualified, the system can automatically or manually inform related personnel, such as train drivers, maintenance personnel and the like, so that the related personnel can take corresponding measures in time for processing. Record and report: the system may automatically record car detection results and corresponding processes and generate reports for subsequent review and analysis. The report may include information on the number of failed cars, time of detection, manner of processing, etc. Through the steps, when the carriage detection result is unqualified, the early warning instruction can be timely executed and braking treatment measures can be taken to ensure the running safety of the train. Meanwhile, related personnel can be informed in time and take corresponding treatment measures so as to avoid accidents.
The above-described embodiment is only a preferred embodiment of the present application, and is not limited in any way, and other variations and modifications may be made without departing from the technical aspects set forth in the claims.
Claims (7)
1. Carriage detects early warning device based on laser radar, its characterized in that includes:
Model construction module: extracting qualified carriage deformation intervals and qualified carriage crack intervals of carriages of different types from historical carriage detection data, and constructing carriage detection channels of the carriages of different types, wherein the types of processing data detected by the carriage detection channels of each type are the same but the detection indexes are different;
The data acquisition module to be detected: detecting a target carriage by adopting a laser radar device to obtain data to be detected, wherein the data to be detected comprises carriage deformation coefficients and carriage crack coefficients of the target carriage;
The detection data preprocessing module is used for: preprocessing data to be detected to obtain standard detection data;
The first car detection result acquisition module: judging whether the carriage deformation coefficient and the carriage crack coefficient of the standard detection data fall into a qualified carriage deformation interval and a qualified carriage crack interval of a carriage detection channel, if so, obtaining a target carriage detection channel, inputting the standard detection data into the target carriage detection channel for detection to obtain a first carriage detection result, if not, outputting carriage detection abnormality, using a support vector machine, a random forest or a convolutional neural network as an initial model, training the initial model through the obtained qualified carriage deformation interval and the qualified carriage crack interval, constructing the carriage detection channel, wherein the carriage detection channel comprises an input layer, a processing layer and an output layer, the input layer receives acquired data, transmits the preprocessed data to the processing layer, integrates the initial model into the processing layer for data processing and prediction, and the output layer outputs the detection result;
The abnormal component result acquisition module: carrying out isomerism analysis according to a first carriage detection result to obtain an isomerism result of a target carriage, wherein the isomerism result reflects the loading state of the target carriage, the isomerism analysis refers to comparing and analyzing each component part of the carriage so as to know the difference and the association between different components, extracting and sorting key parameters including the height, the width and the weight of the carriage in the first carriage detection result, and classifying and comparing the parameters to obtain an isomerism result;
The carriage detection result judging module: and comprehensively evaluating the first carriage detection result and the heterogeneous piece result to obtain a final carriage detection result which is qualified or unqualified.
2. The laser radar-based car detection and early warning device according to claim 1, wherein the laser radar device can scan each part of the target car, and the installation positions, the number and the installation angles of the laser radar devices of different types of cars are set according to the type of car.
3. The laser radar-based carriage detection and early warning method for carrying out carriage detection and early warning by the laser radar-based carriage detection and early warning device as claimed in claim 1 is characterized by comprising the following steps:
S1: extracting qualified carriage deformation intervals and qualified carriage crack intervals of carriages of different types from historical carriage detection data, and constructing carriage detection channels of the carriages of different types, wherein the types of processing data detected by the carriage detection channels of each type are the same but the detection indexes are different;
s2: detecting a target carriage by adopting a laser radar device to obtain data to be detected, wherein the data to be detected comprises carriage deformation coefficients and carriage crack coefficients of the target carriage;
s3: preprocessing data to be detected to obtain standard detection data;
S4: judging whether the carriage deformation coefficient and the carriage crack coefficient of the standard detection data fall into a qualified carriage deformation interval and a qualified carriage crack interval of a carriage detection channel, if so, obtaining a target carriage detection channel, inputting the standard detection data into the target carriage detection channel for detection to obtain a first carriage detection result, if not, outputting carriage detection abnormality, using a support vector machine, a random forest or a convolutional neural network as an initial model, training the initial model through the obtained qualified carriage deformation interval and the qualified carriage crack interval, constructing the carriage detection channel, wherein the carriage detection channel comprises an input layer, a processing layer and an output layer, the input layer receives acquired data, transmits the preprocessed data to the processing layer, integrates the initial model into the processing layer for data processing and prediction, and the output layer outputs the detection result;
S5: carrying out isomerism analysis according to a first carriage detection result to obtain an isomerism result of a target carriage, wherein the isomerism result reflects the loading state of the target carriage, the isomerism analysis refers to comparing and analyzing each component part of the carriage so as to know the difference and the association between different components, extracting and sorting key parameters including the height, the width and the weight of the carriage in the first carriage detection result, and classifying and comparing the parameters to obtain an isomerism result;
S6: and comprehensively evaluating the first carriage detection result and the heterogeneous piece result to obtain a final carriage detection result which is qualified or unqualified.
4. The laser radar based car detection and early warning method according to claim 3, wherein the car types include cars of different load, freight, passenger and special function types.
5. The laser radar-based car detection and early warning method according to claim 4, wherein the method comprises the following steps of: the preprocessing in step S3 includes:
a1: carrying out data filtering on the data to be detected, and removing noise and interference to obtain anti-interference parameters;
A2: carrying out data noise reduction on the anti-interference parameters, and further removing noise and abnormal values to obtain noise reduction parameters;
a3: carrying out data smoothing treatment on the noise reduction parameters to obtain smoothing parameters;
A4: and carrying out data correction on the smoothing parameters to obtain standard detection data.
6. The laser radar-based car detection and early warning method according to claim 4, wherein determining whether the car deformation coefficient of the standard detection data falls within the acceptable car deformation interval of the car detection channel comprises: and inputting the carriage deformation coefficient into a height detection operator and a width detection operator of the carriage detection channel, judging whether the height detection result of the target carriage falls into a height threshold value or not, and judging whether the width detection result of the target carriage falls into a width threshold value or not.
7. The laser radar-based car detection and early warning method according to claim 3 or 6, wherein determining whether the car crack coefficient of the standard detection data falls within the acceptable car crack interval of the car detection channel comprises: inputting the carriage crack coefficient into a crack length detection operator, a crack width detection operator, a crack depth detection operator and a crack expansion rate detection operator of a carriage detection channel, judging whether a crack length detection result of a target carriage falls into a crack length threshold value, judging whether the crack width detection result of the target carriage falls into a crack width threshold value, judging whether the crack depth detection result of the target carriage falls into a crack depth threshold value, and judging whether the crack expansion rate detection result of the target carriage falls into a crack expansion rate threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410396259.6A CN117991288B (en) | 2024-04-03 | 2024-04-03 | Carriage detection early warning device and method based on laser radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410396259.6A CN117991288B (en) | 2024-04-03 | 2024-04-03 | Carriage detection early warning device and method based on laser radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117991288A CN117991288A (en) | 2024-05-07 |
| CN117991288B true CN117991288B (en) | 2024-06-14 |
Family
ID=90887668
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410396259.6A Active CN117991288B (en) | 2024-04-03 | 2024-04-03 | Carriage detection early warning device and method based on laser radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117991288B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118565274B (en) * | 2024-07-31 | 2024-10-01 | 浙江桦芯国创科技有限公司 | Electronic detonator production detection method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109816937A (en) * | 2019-03-29 | 2019-05-28 | 淮阴工学院 | A kind of the gas leakage early warning system and method for tank car |
| AU2021217745A1 (en) * | 2020-02-06 | 2022-09-01 | Inventio Ag | Method and device for determining estimated real dimensions of an elevator car |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104820733B (en) * | 2015-04-17 | 2018-11-27 | 中车青岛四方机车车辆股份有限公司 | A kind of bullet train demand meta-model method for building up and device |
| ES2818654T3 (en) * | 2016-03-21 | 2021-04-13 | Railpod Inc | Combined active and passive procedures and systems to detect and measure internal faults within metal rails |
| CN206348155U (en) * | 2016-12-19 | 2017-07-21 | 北京卓成世纪科技发展有限公司 | Vehicle safety intelligent detecting prewarning system |
| FR3085656B1 (en) * | 2018-09-11 | 2023-04-28 | Renault Sas | ROAD DETECTION METHOD FOR A MOTOR VEHICLE PROVIDED WITH A LIDAR SENSOR |
| CN110376595B (en) * | 2019-06-28 | 2021-09-24 | 湖北大学 | Vehicle measuring system of automatic loading machine |
| DE102021211352B3 (en) * | 2021-10-07 | 2023-02-23 | Cargobeamer Ag | Method for carrying out a wagon technical inspection of a freight train and inspection device for carrying out the method, goods handling method and goods handling device |
| KR102390204B1 (en) * | 2021-10-08 | 2022-04-25 | (주)삼문이엔아이 | On-board system for detecting deformation of railway structures |
| CN117329971B (en) * | 2023-12-01 | 2024-02-27 | 海博泰科技(青岛)有限公司 | Compartment balance detection method and system based on three-dimensional laser radar |
| CN117805848A (en) * | 2024-02-06 | 2024-04-02 | 煤炭科学研究总院有限公司 | Train loading condition detection system and method |
-
2024
- 2024-04-03 CN CN202410396259.6A patent/CN117991288B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109816937A (en) * | 2019-03-29 | 2019-05-28 | 淮阴工学院 | A kind of the gas leakage early warning system and method for tank car |
| AU2021217745A1 (en) * | 2020-02-06 | 2022-09-01 | Inventio Ag | Method and device for determining estimated real dimensions of an elevator car |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117991288A (en) | 2024-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117991288B (en) | Carriage detection early warning device and method based on laser radar | |
| CN110261436B (en) | Rail fault detection method and system based on infrared thermal imaging and computer vision | |
| EP3783356A1 (en) | Train component crack damage monitoring method and system | |
| EP2093561B1 (en) | Device and method for rapid imaging and inspecting of a moving target | |
| US9214016B2 (en) | Pantograph monitoring system and method | |
| KR101469714B1 (en) | System for checking tire uneven wear and method of the same | |
| KR102017870B1 (en) | Real-time line defect detection system | |
| CN111610191B (en) | Road detection and repair system | |
| CN115479949B (en) | Bridge safety monitoring and early warning method and system based on big data | |
| CN113175982A (en) | Railway loading metering method and system based on 3D radar level scanner | |
| CN1854724A (en) | Method for classifying casting defects within the framework of an X-ray analysis | |
| CN117078018A (en) | Highway entry unmanned on duty super management system based on control discernment | |
| CN117006953A (en) | Vehicle overload detection early warning method and system | |
| CN117213392A (en) | Dry bulk cargo carrying capacity measuring method based on optical scanning method | |
| CN112801110B (en) | Target detection method and device for image distortion correction of linear array camera of rail train | |
| Asplund et al. | Assessment of the data quality of wayside wheel profile measurements. | |
| CN118015000B (en) | Surface defect detection method for guide rail based on image processing | |
| CN112013895B (en) | Method for monitoring and evaluating quality condition of railway track basic equipment | |
| CN112012060B (en) | Railway track basic equipment quality condition prediction and judgment method | |
| CN118371734B (en) | Additive manufacturing defect detection system and method based on deep learning | |
| RU2821813C1 (en) | System for assessing technical condition of railway rolling stock | |
| CN118129878A (en) | Overweight detection method and system based on gravity sensor | |
| CN112016180B (en) | Quality condition analysis method for railway track foundation equipment | |
| JP2002257538A (en) | Shape discrimination device and shape discrimination method | |
| Scalera et al. | Automatic Assessment of Rail Squat Defects by Sensor Fusion |
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 |