Pipeline automatic welding fault diagnosis method, device, equipment and medium

Abstract

The invention discloses an automatic pipeline welding fault diagnosis method which comprises the following steps: acquiring fault sample data, and performing fault classification on the fault sample data through a classification algorithm; collecting operation data of a plurality of sensors arranged on the automatic pipeline welding machine part; performing data preprocessing and standardization processing on the operation data to generate standardized data; classifying the standardized data based on the category corresponding to the fault classification, and performing model training on the classified standardized data and a fault classification result through a random forest algorithm to generate a fault judgment model; and diagnosing faults of the welding machine parts in real time based on the fault judgment model. The invention further discloses an automatic pipeline welding fault diagnosis device and equipment and a medium. According to the scheme, the faults of the multiple parts of the automatic welding machine can be diagnosed online in real time, the difficulty of manually identifying the faults of all the units of the pipeline automatic welding machine by technicians is lowered, and the fault removal efficiency and the positioning precision are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic welding fault analysis, and in particular to a pipeline automatic welding fault diagnosis method, device, equipment and readable medium. Background Art

[0002] The pipeline automatic welding machine is a complex and high-precision equipment system composed of a welding trolley, welding track, remote control unit, handheld device, signal processing unit, drive unit, main control unit, welding power supply control unit, welding power supply, etc. It is the key to ensuring the quality of pipeline welding. Once a fault occurs, it will affect the normal welding of the pipeline.

[0003] At present, on-site service uses skilled technicians to provide technical support for automatic welding equipment. If multiple units have equipment failures at the same time, service delays will occur. Existing fault diagnosis requires technicians to tediously distinguish each unit of the automatic welding machine and then make a diagnosis, which not only consumes a lot of manpower and energy, but also due to technical limitations, the fault cannot be accurately located.

[0004] Therefore, there is a need in the prior art for improving the automatic pipeline welding fault diagnosis method based on machine learning. Summary of the invention

[0005] In view of this, the purpose of the embodiments of the present invention is to propose a pipeline automatic welding fault diagnosis method, device, equipment and readable medium. The present invention can accurately locate the faults of each unit of automatic welding, provide a practical diagnosis plan, and realize the intelligence of pipeline automation.

[0006] Based on the above purpose, an embodiment of the present invention provides a pipeline automatic welding fault diagnosis method, comprising the following steps:

[0007] Obtain fault sample data of the pipeline automatic welding machine, and classify the fault sample data through a classification algorithm;

[0008] Collecting operation data of multiple sensors installed on the components of the pipeline automatic welding machine;

[0009] Perform data preprocessing and standardization on the operation data to generate standardized data;

[0010] Classifying the standardized data based on the categories corresponding to the fault classification, and performing model training on the classified standardized data and the results of the fault classification by using a random forest algorithm to generate a fault judgment model;

[0011] Real-time diagnosis of welding machine component faults based on the fault judgment model.

[0012] In some implementations, classifying the fault sample data by a classification algorithm includes:

[0013] The fault sample data is classified into communication faults, system faults and status faults through KNN classification algorithm.

[0014] Among them, communication failures include abnormal communication between the main controller and the motion controller, abnormal communication between the motion controller and the drive module, abnormal communication between the main controller and the handheld device, and abnormal communication between the main controller and the power module; system failures include abnormal system reading and writing database, and system crash; status failures include abnormal trolley movement speed, abnormal wire feeding speed, abnormal swing width, abnormal welding angle, abnormal welding voltage, abnormal welding current, abnormal arc stability, abnormal temperature, abnormal gas flow, abnormal arc starting state, and abnormal arc ending state.

[0015] In some embodiments, data preprocessing includes data classification, filtering, data outlier removal, frequency matching, and encoding.

[0016] In some implementations, classifying the standardized data based on the category corresponding to the fault classification includes:

[0017] The standardized data is classified based on the categories corresponding to the fault classification, and PCA is used to reduce the feature dimension after classification.

[0018] In some implementations, performing model training on the classified standardized data and the results of the fault classification by using a random forest algorithm to generate a fault judgment model includes:

[0019] The classified standardized data and the results of the fault classification are divided into a training set and a test set. The classified standardized data are trained on the model using the random forest algorithm in the training set. The model is evaluated on the test set using confusion matrix, precision, and recall. The robustness of the model is evaluated using F1-score. The hyperparameters of the decision tree are adjusted based on the evaluation results to optimize the model.

[0020] In some embodiments, the multiple sensors include voltage sensors, current sensors, angle sensors, air flow sensors, temperature sensors and height sensors, and the multiple sensors are deployed on the welding carriage, welding track, remote control unit, handheld device, signal processing unit, drive unit, main control unit, welding power supply control unit, and welding power supply to collect operation data.

[0021] In some implementations, diagnosing a fault of a welding machine component in real time based on a fault judgment model includes:

[0022] The fault judgment model is stored in a cloud server or a local database, and the fault judgment model in the cloud server or the local database is retrieved to perform real-time diagnosis of the fault of the welding machine component.

[0023] Another aspect of the embodiment of the present invention further provides a pipeline automatic welding fault diagnosis device, comprising:

[0024] A fault classification module, the fault classification module is configured to obtain fault sample data of the pipeline automatic welding machine, and classify the fault sample data through a classification algorithm;

[0025] A data acquisition module, the data acquisition module is configured to collect operating data of a plurality of sensors disposed on components of the pipeline automatic welding machine;

[0026] A data processing module, the data processing module is configured to perform data preprocessing and standardization processing on the operation data to generate standardized data;

[0027] A model training module, the model training module is configured to classify the standardized data based on the categories corresponding to the fault classification, and perform model training on the classified standardized data and the results of the fault classification through a random forest algorithm to generate a fault judgment model;

[0028] The fault judgment module is configured to diagnose the faults of welding machine components in real time based on the fault judgment model.

[0029] According to another aspect of the embodiments of the present invention, there is provided a computer device, comprising: at least one processor; and a memory, wherein the memory stores computer instructions executable on the processor, and the instructions implement the steps of the above method when executed by the processor.

[0030] According to another aspect of the embodiments of the present invention, a computer-readable storage medium is provided, which stores a computer program that implements the above method steps when executed by a processor.

[0031] The present invention has at least the following beneficial technical effects:

[0032] The present invention provides a pipeline automatic welding fault diagnosis method, device, equipment and readable medium. By performing cluster analysis on automatic welding fault samples, the operation data of multiple components are obtained to train a fault judgment model, and the faults of multiple components of the automatic welding machine are diagnosed in real time online, which reduces the difficulty of manual identification of faults of various units of the pipeline automatic welding machine by technical service personnel, improves fault elimination efficiency and positioning accuracy, and provides necessary technical guarantees for the stability and promotion of pipeline automatic welding. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these drawings without paying creative work.

[0034] Figure 1 A schematic diagram of an embodiment of a pipeline automatic welding fault diagnosis method provided by the present invention;

[0035] Figure 2 A schematic diagram of an embodiment of the fault classification provided by the present invention;

[0036] Figure 3 A schematic diagram of an embodiment of a pipeline automatic welding fault diagnosis device provided by the present invention;

[0037] Figure 4 A schematic diagram of an embodiment of a computer device provided by the present invention;

[0038] Figure 5 A schematic diagram of an embodiment of a computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION

[0039] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the embodiments of the present invention are further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0040] It should be noted that all expressions using "first" and "second" in the embodiments of the present invention are for distinguishing two non-identical entities with the same name or non-identical parameters. It can be seen that "first" and "second" are only for the convenience of expression and should not be understood as limitations on the embodiments of the present invention. The subsequent embodiments will not explain this one by one.

[0041] Based on the above objectives, a first aspect of an embodiment of the present invention provides an embodiment of a pipeline automatic welding fault diagnosis method. Figure 1 FIG. 2 is a schematic diagram of an embodiment of a pipeline automatic welding fault diagnosis method provided by the present invention. Figure 1 As shown, the pipeline automatic welding fault diagnosis method according to the embodiment of the present invention comprises the following steps:

[0042] S1. Obtain fault sample data of the pipeline automatic welding machine, and classify the fault sample data through a classification algorithm;

[0043] S2, collecting operation data of multiple sensors arranged on the components of the pipeline automatic welding machine;

[0044] S3, performing data preprocessing and standardization on the operation data to generate standardized data;

[0045] S4. Classify the standardized data based on the categories corresponding to the fault classification, and perform model training on the classified standardized data and the results of the fault classification by using a random forest algorithm to generate a fault judgment model;

[0046] S5. Diagnose the faults of welding machine components in real time based on the fault judgment model.

[0047] Furthermore, in S1, classifying the fault sample data by a classification algorithm includes:

[0048] The KNN and other algorithms are used to cluster the welding equipment fault sample data set, and professional technicians are combined to classify and reclassify according to the equipment module, type, function, etc. of the welding machine fault. According to the long-term use of pipeline automatic welding, the fault types and general distribution are summarized, and the fault categories are divided into communication faults, system faults, and status faults. The fault classification is as follows: Figure 2 shown.

[0049] Among them, communication faults include abnormal communication between the main controller and the motion controller, abnormal communication between the motion controller and the drive module, abnormal communication between the main controller and the handheld device, and abnormal communication between the main controller and the power module;

[0050] System failures include abnormalities in reading and writing databases and system crashes;

[0051] Status-related faults include abnormal trolley movement speed, abnormal wire feeding speed, abnormal swing width, abnormal welding angle, abnormal welding voltage, abnormal welding current, abnormal arc stability, abnormal temperature, abnormal gas flow, abnormal arc starting state, and abnormal arc ending state.

[0052] Further, in S2, raw operation data is collected by sensors deployed on the welding carriage, welding track, main control unit, handheld device, signal processing unit, drive unit, touch screen, welding power supply control unit, welding power supply, etc., including: current, voltage, motor speed, acceleration, gas flow, inter-module communication, welding machine fault data, etc. In some embodiments, multiple sensors include voltage sensors, current sensors, angle sensors, gas flow sensors, temperature sensors, and height sensors. The specific data obtained include: welding layer code, handheld device button code, signal processing unit on-off information, position information (angle), welding speed, swing width, side stop time, motor speed, given current of welding equipment, real-time current, given voltage, real-time voltage, given wire feeding speed, wire feeding speed, gas flow, temperature, altitude, etc. The above-mentioned deployed sensor data is collected to the control system through a multi-channel data collector.

[0053] Furthermore, in S3, due to the different acquisition frequencies of the sensors, the collected data are processed by down-conversion and resampling of current, voltage, motor speed, acceleration, etc., wherein the data preprocessing includes data classification, filtering, data outlier removal, frequency synchronization, and encoding. Specifically:

[0054] Data classification: Some of the collected raw data are continuous values ​​and some are Boolean values. According to the type of value, they are divided into two types: continuous and Boolean.

[0055] Data outlier removal: For continuous value types, algorithms such as normal distribution are used to remove noise and abnormal data in the collected data; specifically, the mathematical expectation and standard deviation of the normal population are calculated, appropriate probability data are selected to be retained, and other values ​​are removed.

[0056] Filtering: Since the collected signals of electrical signals such as current, voltage, motor speed, acceleration, etc. are affected by various noises, filtering is performed; FC1:6 and FC2:7 bandpass filters are used to filter and remove noises.

[0057] Frequency synchronization: Since the acquisition frequencies of each sensor are different, all data must be processed at the same frequency, and the appropriate frequency is selected for the acquired data for frequency synchronization. Most sensor data such as current, voltage, motor speed, acceleration, temperature, and air flow are collected at high frequencies, and resampling and other measures are used for frequency synchronization; the acquisition frequencies of data such as welding layer codes, handheld device button codes, and signal processing unit on / off information are processed using high-frequency sampling measures.

[0058] One-Hot coding: Most pipeline automatic welding faults are discrete data. In order to meet the algorithm requirements, one-hot coding is performed on each fault according to the quantity distribution of the fault state.

[0059] Data standardization processing: After the collected data is frequency-homogenized, it is standardized with a mean of 0 and a standard deviation of 1, so that all collected data are under the same standard.

[0060] In some embodiments of the present invention, classifying the standardized data based on the categories corresponding to the fault classification includes: classifying the standardized data based on the categories corresponding to the fault classification, and performing feature dimension reduction using PCA after classification.

[0061] Furthermore, in S4, model training, evaluation and hyperparameter tuning are completed, and the standardized data and the results of fault classification are divided into training sets and test sets. Since faults are divided into three categories: communication, system and status, a multi-level classification strategy is adopted to first distinguish the communication class, system class and status class. This application uses PCA to reduce the dimension of features and separate them. The random forest algorithm is used to train the fault model within the major categories. The pros and cons of the model are evaluated on the test set through confusion matrix, precision rate and recall rate. The robustness of the model is evaluated by F1-score. Through the above evaluation parameters, adjust the hyperparameters such as n_estimators (number of trees), max_depth (depth of the tree), max_leaf_nodes (maximum number of leaf nodes) to optimize the model. After repeated testing and optimization, a suitable model is trained.

[0062] Further, in S5, diagnosing the fault of the welding machine component in real time based on the fault judgment model includes:

[0063] The fault judgment model is stored in a cloud server or a local database, and the fault judgment model of the cloud server or the local database is retrieved to perform real-time diagnosis of the faults of welding machine components. In some embodiments, the real-time monitoring data of pipeline automatic welding is uploaded to a network server or a cloud platform through an interface for remote fault diagnosis, which can realize fault detection of multiple welding machines at the same time. In other embodiments, the model can be directly called and fault diagnosis can be performed by local deployment on the device.

[0064] The specific implementation of the present invention is further described below based on specific examples.

[0065] In this embodiment, first, the pipeline automatic welding fault classification is used to obtain key information. First, based on the common fault transmission mechanism, professional and technical personnel are divided into communication class, system class, and state class. For the large class, the welding machine equipment sample data set is clustered by algorithms such as KNN. Specifically, in the communication class, the K value in the KNN algorithm defaults to 6. In the system class, the K value in the KNN algorithm defaults to 4. The state class is divided into motion class and process class. The K value in the KNN algorithm in the motion class defaults to 7. The K value in the KNN algorithm in the process class defaults to 10. The CH method is used to evaluate the above clustering model, and the fault classification is obtained after optimization.

[0066] Furthermore, sensor data is obtained, and voltage and current sensors with appropriate ranges are selected without other interference to measure voltage and current. The magnetoelectric method is used to measure the movement position, speed and distance of the walking motor, swing motor and wire feeding motor. The angle sensor measures the welding gun position. The gas flow sensor is used to measure the gas flow. The temperature sensor and height sensor are used to measure the temperature and height values. The above-mentioned deployed sensor data are collected into the control system through a multi-channel data collector.

[0067] Furthermore, the data is preprocessed and standardized. Data preprocessing includes filtering, data outlier removal, frequency matching, one-hot encoding, data standardization, etc.

[0068] Current, voltage, etc. are collected at high frequency (200000), and high-frequency filtering is performed on the current and voltage signals based on Shannon sampling theorem. In this embodiment, a bandpass filter with parameters FC1:6 and FC2:7, structure: direct FIR, and design method is selected to perform filtering to remove clutter.

[0069] In the process of eliminating data outliers, this embodiment adopts normal distribution to calculate the mathematical expectation μ and standard deviation σ, and eliminates data greater than μ+3σ or less than μ-3σ.

[0070] This embodiment performs frequency synchronization processing on data collected by sensors of different frequencies, collects data with high frequency, and adopts measures such as resampling to perform frequency synchronization processing; the data collection frequency of welding layer code, handheld device button code, signal processing unit on / off information, etc. is processed by high-frequency sampling measures.

[0071] The above data are then standardized with a mean of 0 and a standard deviation of 1 to make the collected data under the same standard. The data are dimensionless to make the various data under the same specifications.

[0072] Furthermore, the above data are used for model training, evaluation and hyperparameter tuning. First, the standardized data is divided into a training set and a test set. According to the correspondence between the above collected data and the welding machine fault sample set, a welding machine state fault sample set is formed: D = {(y|x1, x2, x3, x4, ..., xn)}, where x = {x1, x2, x3, x4, ..., xn} is the preprocessed data set and y is the fault.

[0073] Furthermore, the PCA feature dimensionality reduction method is used to classify the faults into three categories: communication, system, and status. The random forest algorithm is used to train the fault model within the major categories. The confusion matrix, precision, and recall rate are used to evaluate the quality of the model on the test set, and 0.9 and 0.9 are taken here. The F1-score is used to evaluate the robustness of the model, and 0.8 is taken here. According to the above evaluation parameters, n_estimators (the number of trees) is adjusted to 4, max_depth (the depth of the tree) is adjusted to 4, and max_leaf_nodes (the maximum number of leaf nodes) is adjusted to 12.

[0074] In this embodiment, the trained model is deployed to a back-end network server or a cloud platform, and the real-time monitoring data of pipeline automatic welding is uploaded to the back-end network server or a cloud platform through an interface for remote fault diagnosis.

[0075] It should be pointed out in particular that the various steps in the various embodiments of the above-mentioned pipeline automatic welding fault diagnosis method can be cross-linked, replaced, added, and deleted. Therefore, these reasonable permutations and combinations of the pipeline automatic welding fault diagnosis method should also belong to the protection scope of the present invention, and the protection scope of the present invention should not be limited to the embodiments.

[0076] Based on the above purpose, a second aspect of an embodiment of the present invention provides a pipeline automatic welding fault diagnosis device. Figure 3 FIG. 1 is a schematic diagram of an embodiment of a pipeline automatic welding fault diagnosis device provided by the present invention. Figure 3 As shown, the pipeline automatic welding fault diagnosis device according to the embodiment of the present invention includes the following modules:

[0077] A fault classification module 011 is configured to obtain fault sample data of the pipeline automatic welding machine and classify the fault sample data through a classification algorithm;

[0078] A data acquisition module 012, the data acquisition module 012 is configured to collect operating data of a plurality of sensors disposed on the components of the pipeline automatic welding machine;

[0079] The data processing module 013 is configured to perform data preprocessing and standardization processing on the operation data to generate standardized data;

[0080] The model training module 014 is configured to classify the standardized data based on the categories corresponding to the fault classification, and perform model training on the classified standardized data and the results of the fault classification by using a random forest algorithm to generate a fault judgment model;

[0081] The fault judgment module 015 is configured to diagnose the faults of the welding machine components in real time based on the fault judgment model.

[0082] Based on the above purpose, a third aspect of an embodiment of the present invention provides a computer device. Figure 3 FIG. 2 is a schematic diagram of an embodiment of a computer device provided by the present invention. Figure 4 As shown, the computer device of the embodiment of the present invention includes the following devices: at least one processor 021; and a memory 022, the memory 022 stores computer instructions 023 that can be executed on the processor, and the instructions implement the steps of the above method when executed by the processor.

[0083] The present invention also provides a computer-readable storage medium. Figure 5 FIG. 2 is a schematic diagram of an embodiment of a computer-readable storage medium provided by the present invention. Figure 5As shown, the computer-readable storage medium 031 stores a computer program 032 that performs the above method when executed by a processor.

[0084] Finally, it should be noted that a person skilled in the art can understand that all or part of the processes in the above-mentioned embodiment method can be implemented by instructing the relevant hardware through a computer program, and the program of the pipeline automatic welding fault diagnosis method can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium of the program can be a disk, an optical disk, a read-only storage memory (ROM) or a random access memory (RAM), etc. The above-mentioned computer program embodiment can achieve the same or similar effect as any of the corresponding aforementioned method embodiments.

[0085] In addition, the method disclosed in the embodiment of the present invention can also be implemented as a computer program executed by a processor, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the above functions defined in the method disclosed in the embodiment of the present invention are performed.

[0086] In addition, the above method steps and system units may also be implemented using a controller and a computer-readable storage medium for storing a computer program that enables the controller to implement the above steps or unit functions.

[0087] It will also be appreciated by those skilled in the art that various exemplary logic blocks, modules, circuits and algorithm steps described in conjunction with the disclosure herein can be implemented as electronic hardware, computer software or a combination of the two. In order to clearly illustrate this interchangeability of hardware and software, a general description has been given to the functions of various schematic components, blocks, modules, circuits and steps. Whether this function is implemented as software or hardware depends on specific applications and the design constraints imposed on the entire system. Those skilled in the art can implement the function in various ways for each specific application, but this implementation decision should not be interpreted as causing a departure from the disclosed scope of the embodiments of the present invention.

[0088] In one or more exemplary designs, the function can be implemented in hardware, software, firmware or any combination thereof. If implemented in software, the function can be stored on a computer-readable medium or transmitted by a computer-readable medium as one or more instructions or codes. Computer-readable media include computer storage media and communication media, and the communication media include any media that helps to transmit a computer program from one location to another. The storage medium can be any available medium that can be accessed by a general or special computer. As an example and not limiting, the computer-readable medium may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage devices, magnetic disk storage devices or other magnetic storage devices, or any other medium that can be used to carry or store the required program code in the form of an instruction or data structure and can be accessed by a general or special computer or a general or special processor. In addition, any connection can be appropriately referred to as a computer-readable medium. For example, if a coaxial cable, optical fiber cable, twisted pair, digital subscriber line (D0L) or wireless technologies such as infrared, radio and microwaves are used to send software from a website, server or other remote source, the above-mentioned coaxial cable, optical fiber cable, twisted pair, D0L or wireless technologies such as infrared, radio and microwaves are all included in the definition of the medium. As used herein, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk, Blu-ray disc, wherein disks usually reproduce data magnetically, while optical discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0089] The above are exemplary embodiments disclosed in the present invention, but it should be noted that various changes and modifications may be made without departing from the scope disclosed in the embodiments of the present invention as defined in the claims. The functions, steps and / or actions of the method claims according to the disclosed embodiments described herein do not need to be performed in any particular order. In addition, although the elements disclosed in the embodiments of the present invention may be described or required in individual form, they may also be understood as multiple unless explicitly limited to the singular.

[0090] It should be understood that, as used herein, the singular forms "a", "an" are intended to include the plural forms as well, unless the context clearly supports an exception. It should also be understood that, as used herein, "and / or" refers to any and all possible combinations including one or more of the associated listed items.

[0091] The serial numbers of the embodiments disclosed in the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0092] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0093] A person skilled in the art should understand that the discussion of any of the above embodiments is only exemplary and is not intended to imply that the scope of the disclosure of the embodiments of the present invention (including the claims) is limited to these examples; under the concept of the embodiments of the present invention, the technical features in the above embodiments or different embodiments can also be combined, and there are many other changes in different aspects of the embodiments of the present invention as above, which are not provided in detail for the sake of simplicity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the protection scope of the embodiments of the present invention.

Claims

1. A pipeline automatic welding fault diagnosis method, characterized in that: The following steps are involved: Acquire fault sample data of the pipeline automatic welding machine, and classify the fault sample data by using a classification algorithm; Collecting operation data of multiple sensors installed on the components of the pipeline automatic welding machine; Performing data preprocessing and standardization on the operation data to generate standardized data; Classifying the standardized data based on the categories corresponding to the fault classification, and performing model training on the classified standardized data and the results of the fault classification by using a random forest algorithm to generate a fault judgment model; The faults of welding machine components are diagnosed in real time based on the fault judgment model.

2. The pipeline automatic welding fault diagnosis method according to claim 1 is characterized in that: Classifying the fault sample data by a classification algorithm includes: The fault sample data is classified into communication faults, system faults and status faults by using a KNN classification algorithm; Among them, the communication type failures include abnormal communication between the main controller and the motion controller, abnormal communication between the motion controller and the drive module, abnormal communication between the main controller and the handheld device, and abnormal communication between the main controller and the power module; the system type failures include abnormal system reading and writing database, and system crash; the status type failures include abnormal trolley movement speed, abnormal wire feeding speed, abnormal swing width, abnormal welding angle, abnormal welding voltage, abnormal welding current, abnormal arc stability, abnormal temperature, abnormal gas flow, abnormal arc starting state, and abnormal arc ending state.

3. The pipeline automatic welding fault diagnosis method according to claim 1 is characterized in that: The data preprocessing includes data classification, filtering processing, data outlier elimination, frequency synchronization processing, and encoding processing.

4. The pipeline automatic welding fault diagnosis method according to claim 1 is characterized in that: Classifying the standardized data based on the category corresponding to the fault classification includes: The standardized data is classified based on the categories corresponding to the fault classification, and after classification, PCA is used to perform feature dimension reduction.

5. The pipeline automatic welding fault diagnosis method according to claim 1 is characterized in that: Performing model training on the classified standardized data and the result of the fault classification by using a random forest algorithm to generate a fault judgment model includes: The classified standardized data and the results of the fault classification are divided into a training set and a test set. The classified standardized data are trained on a model using a random forest algorithm in the training set. The model is evaluated on the test set using confusion matrix, precision, and recall. The robustness of the model is evaluated using F1-score. The hyperparameters of the decision tree are adjusted based on the evaluation results to optimize the model.

6. The pipeline automatic welding fault diagnosis method according to claim 1 is characterized in that: The multiple sensors include voltage sensors, current sensors, angle sensors, air flow sensors, temperature sensors and height sensors. The multiple sensors are deployed on the welding carriage, welding track, remote control unit, handheld device, signal processing unit, drive unit, main control unit, welding power supply control unit and welding power supply to collect operation data.

7. The pipeline automatic welding fault diagnosis method according to claim 1 is characterized in that: Real-time diagnosis of welding machine component faults based on the fault judgment model includes: The fault judgment model is stored in a cloud server or a local database, and the fault judgment model in the cloud server or the local database is retrieved to perform real-time diagnosis of the fault of the welding machine component.

8. A pipeline automatic welding fault diagnosis device, characterized in that: include: A fault classification module, wherein the fault classification module is configured to obtain fault sample data of the pipeline automatic welding machine and classify the fault sample data through a classification algorithm; A data acquisition module, wherein the data acquisition module is configured to collect operating data of a plurality of sensors disposed on components of the pipeline automatic welding machine; A data processing module, wherein the data processing module is configured to perform data preprocessing and standardization processing on the operation data to generate standardized data; A model training module, wherein the model training module is configured to classify the standardized data based on the categories corresponding to the fault classification, and perform model training on the classified standardized data and the results of the fault classification by using a random forest algorithm to generate a fault judgment model; A fault judgment module is configured to diagnose faults of welding machine components in real time based on the fault judgment model.

9. A computer device, characterized in that: include: at least one processor; as well as A memory storing computer instructions executable on the processor, wherein the instructions, when executed by the processor, implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images