Method for processing running state data of gas shield welding equipment
By segmenting, encoding, compressing and principal component analysis the time series data of gas shielded welding equipment, the problem of zero value influence in gas shielded welding equipment data processing is solved, and efficient compression and accurate analysis of data are achieved.
Patent Information
- Application Number
- CN202510708326.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-29
- Publication Date
- 2025-09-30
AI Technical Summary
When processing large-scale, low-information-density time series data generated by gas shielded welding equipment, existing technologies are unable to accurately retain important data, resulting in reduced data analysis efficiency and accuracy.
By acquiring time series data, segmenting it based on continuous non-zero values, compressing it using run-length coding and constructing a two-dimensional matrix, principal component analysis is performed to determine the target operating status data.
Effectively remove zero-value data, improve information density, compress data size, while retaining data precision to ensure the accuracy of data analysis.
Smart Images
Figure CN120729326A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data processing, and in particular to a method for processing operating status data of gas shielded welding equipment. Background Art
[0002] During the real-time monitoring and maintenance of gas shielded welding equipment, it is usually necessary to collect and store a large amount of time series data, such as current, voltage, wire feed speed, shielding gas flow, etc.
[0003] Due to the unique nature of the welding process, this data often contains a large number of zero or near-zero values. Traditional storage and processing methods easily waste data storage space and consume a large amount of computing resources. Furthermore, existing data compression schemes often suffer from low compression ratios and poor processing efficiency when dealing with large-scale, low-information-density time series data. Furthermore, they are unable to accurately preserve important time series data during equipment operation, thus affecting the efficiency and accuracy of data analysis.
[0004] It can be seen that when processing the large-scale, low-information-density time series data generated by gas shielded welding equipment during operation, the existing technology cannot accurately retain the important time series data of the equipment during operation, thereby affecting the efficiency and accuracy of data analysis. Summary of the Invention
[0005] In view of this, it is necessary to provide a method for processing the operating status data of gas shielded welding equipment to solve the problem that the existing technology cannot accurately retain the important time series data of the equipment during operation when processing the large-scale, low information density time series data generated by the gas shielded welding equipment during operation, thereby affecting the efficiency and accuracy of data analysis.
[0006] In order to solve the above problems, in a first aspect, the present invention provides a method for processing operating status data of gas shielded welding equipment, comprising: Obtaining time series data of the operation of the gas shielded welding equipment, the time series data including continuous timestamps and the operating status data of the gas shielded welding equipment at each timestamp; Traversing the running status data based on continuous timestamps, and segmenting the running status data according to continuous non-zero values in the running status data to obtain multiple running status data segments; Run-length encoding is performed on each operating status data segment, and a two-dimensional matrix is constructed based on the compressed operating status data segment. Principal component analysis is performed on the two-dimensional matrix to determine target operating status data in the operating status data.
[0007] In a possible implementation, the running status data is segmented according to the continuous non-zero value running status data in the running status data to obtain multiple running status data segments, including: Determine a zero value judgment threshold for the running status data according to the type of the running status data; Continuous running status data of each type of running status data at the same time stamp that are all greater than the corresponding zero value judgment threshold are determined as a running status data segment, thereby obtaining a plurality of running status data segments.
[0008] In a possible implementation, after obtaining the time series data of the operation of the gas shielded welding equipment, the method includes: The running status data is divided into blocks according to the volume of the running status data to obtain a plurality of running status data blocks, wherein each running status data block can be divided into a plurality of running status data segments in parallel.
[0009] In a possible implementation, when the running status data is processed in blocks, the method further includes: determining the required zero value density in each operating status data block according to a preset compression ratio; The operation status data is processed in blocks according to the required zero value density to ensure that the zero value density of each operation status data block is greater than or equal to the required zero value density.
[0010] In one possible implementation, constructing a two-dimensional matrix based on the compressed operating status data segments includes: The running status data segments are sorted according to the timestamps corresponding to the running status data in the compressed running status data segments to generate a two-dimensional matrix, wherein each row in the two-dimensional matrix represents a type of running status data, and each data in the two-dimensional matrix includes the value of each running status data and the number of times the value appears.
[0011] In a possible implementation, performing principal component analysis on a two-dimensional matrix to determine target operating status data in the operating status data includes: Calculating the covariance matrix of each operating status data based on the two-dimensional matrix, and extracting the eigenvalues and eigenvectors of the operating status data according to the covariance matrix; The target operating status data is determined based on the eigenvector corresponding to the eigenvalue greater than the preset eigenvalue threshold and each operating status data in the compressed operating status data segment.
[0012] In one possible implementation, determining target operating status data based on a feature vector corresponding to a feature value greater than a preset feature value threshold and the operating status data includes: performing standardization processing on each operating status data in the compressed operating status data segment to obtain standardized operating status data; A low-dimensional space is constructed based on the eigenvectors corresponding to eigenvalues greater than a preset eigenvalue threshold, and the eigenvectors are used to project the standardized operating status data into the low-dimensional space to obtain the target operating status data.
[0013] The beneficial effects of the present invention are as follows: the method for processing the operating status data of the gas shielded welding equipment provided by the present invention traverses the operating status data according to the timestamps in the acquired time series data of the operation of the gas shielded welding equipment, and segments the operating status data according to the continuous non-zero values in the operating status data to obtain multiple operating status data segments, which can effectively remove the zero-value data in the operating status data and improve the information density of the operating status data. By performing run-length encoding compression on each operating status data segment, the volume of the operating status data can be further compressed and the scale of the operating status data can be reduced, while retaining the accuracy of the operating status data. By constructing a two-dimensional matrix based on the compressed operating status data segments and performing principal component analysis on the two-dimensional matrix, the target operating status data in the operating status data is determined, and the important operating status data in the operating status data that plays a decisive role in data analysis can be accurately located to ensure the accuracy of the data analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.
[0015] Figure 1 A flow chart of a method for processing operating status data of gas shielded welding equipment provided by an embodiment of the present invention; Figure 2 A schematic diagram of a flow chart of a data segmentation method provided in an embodiment of the present invention; Figure 3 A schematic diagram of a data segmentation method according to an embodiment of the present invention; Figure 4 A flowchart of a method for determining target operating status data provided by an embodiment of the present invention; Figure 5 A schematic diagram of a flow chart of a data dimensionality reduction method provided by an embodiment of the present invention; Figure 6 This is an architecture diagram of a welding data acquisition hardware device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0016] The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein the accompanying drawings constitute a part of the present invention and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.
[0017] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0018] Real-time monitoring and maintenance of gas shielded welding equipment typically requires the collection and storage of large amounts of time-series data, such as current, voltage, wire feed speed, and shielding gas flow. Due to the unique nature of the welding process, this data often contains a large number of zero or near-zero values. Traditional storage and processing methods can lead to a waste of data storage space and high consumption of computing resources. Furthermore, existing data compression schemes often suffer from low compression ratios and poor processing efficiency when dealing with large-scale, low-information-density time-series data. They also fail to accurately retain important time-series data from the equipment during operation, impacting the efficiency and accuracy of data analysis.
[0019] In order to solve the above technical problems, Figure 1 As shown, the present invention provides a method for processing operating status data of gas shielded welding equipment, comprising: S101, obtaining time series data of the operation of the gas shielded welding equipment, where the time series data includes continuous time stamps and operation status data of the gas shielded welding equipment at each time stamp.
[0020] In the embodiment of the present invention, time series data refers to data that records the continuous operating status data generated by the gas shielded welding equipment during welding in the order of timestamps, including continuous timestamps and the operating status data of the gas shielded welding at each timestamp. A variety of sensors can be used to collect key parameters of the gas shielded welding equipment during the welding process, such as current, voltage, temperature, gas flow, welding speed, etc. Specifically, as shown in Table 1, a set of time series data of the operation process of the gas shielded welding equipment provided by the embodiment of the present invention is shown. In actual applications, when the gas shielded welding equipment is in use, the time series data generated is massive, generally Level data, and due to the characteristics of the operation of the gas shielded welding equipment, there will be a large amount of zero-value data in the middle, such as the 3rd, 5th, and 7th rows in the table. These zero-value data greatly reduce the information density of the data and increase the data volume, which will affect the analysis of the gas shielded welding operation status data. The present invention will process it later.
[0021] Table 1: Time series data of gas shielded welding equipment
[0022] S102 , traversing the running status data based on continuous timestamps, and segmenting the running status data according to continuous non-zero values in the running status data to obtain a plurality of running status data segments.
[0023] In an embodiment of the present invention, as in the aforementioned embodiment, in order to eliminate the influence of zero-value data on the analysis of the gas shielded welding operation status data, it is necessary to traverse the time series data of the gas shielded welding operation to remove the zero-value data. Specifically, the operation status data can be traversed according to the continuous timestamps in the time series data, and the size of the operation status data and the zero value can be compared one by one. The operation status data is segmented according to the comparison result, and the continuous non-zero value operation status data is used as an operation status data segment to obtain the operation status data segment. For example, referring to Table 1, 7 data segments can be obtained by segmenting the operation status data in Table 1, which are: Data segment 1: welding current [150, 148], welding voltage [24, 23.8], wire feed speed [5, 4.9], gas flow rate [20, 19.5], (time interval: 1704067200 to 1704067201); Data segment 2: welding current
[155] , welding voltage [24.2], wire feed speed [5.1], gas flow rate [20.5] (time interval: 1704067203); Data segment 3: welding current
[160] , welding voltage [24.5], wire feed speed [5.2], gas flow rate
[21] (time interval: 1704067205); Data segment 4: welding current [165, 152, 149], welding voltage [24.7, 24.1, 23.9], wire feed speed [5.3, 5.0, 4.8], gas flow rate [21.5, 20.3, 19.8] (time interval: 1704067207 to 1704067209); Data segment 5: welding current
[156] , welding voltage [24.3], wire feed speed [5.2], gas flow rate [20.6] (time interval: 1704067211); Data segment 6: welding current
[161] , welding voltage [24.6], wire feed speed [5.3], gas flow rate [21.1] (time interval: 1704067213); Data segment 7: welding current
[166] , welding voltage [24.8], wire feed speed [5.4], gas flow rate [21.6] (time interval: 1704067215).
[0024] Of course, the above embodiment is only a simple embodiment for ease of description. The time series generated during the specific implementation process will be much more complicated, and the length of each data segment is also variable, and needs to be divided according to actual conditions.
[0025] S103 , performing run-length encoding compression on each running status data segment, constructing a two-dimensional matrix based on the compressed running status data segment, performing principal component analysis on the two-dimensional matrix, and determining target running status data in the running status data.
[0026] In an embodiment of the present invention, run-length coding compression is a lossless compression method that can represent the same data in the form of data value + number of repetitions, and can retain data accuracy to the greatest extent. After the operation status data segment is compressed using run-length coding compression, a two-dimensional matrix is constructed based on the compressed operation status data segment. The constructed two-dimensional matrix can retain the structure and data value of the operation status data with high fidelity, which is convenient for subsequent principal component analysis. The specific construction method of the two-dimensional matrix will be described in detail later in the present invention. For the constructed two-dimensional matrix, the target operation status data in the operation status data can be determined through the principal component analysis algorithm. The specific process of the principal component analysis will be described in detail later in the present invention.
[0027] The method for processing the operating status data of gas shielded welding equipment provided by the present invention traverses the operating status data according to the timestamps in the acquired time series data of the operation of the gas shielded welding equipment, and segments the operating status data according to the continuous non-zero values in the operating status data to obtain multiple operating status data segments. The method can effectively remove zero-value data in the operating status data and improve the information density of the operating status data. By performing run-length encoding compression on each operating status data segment, the volume of the operating status data can be further compressed and the scale of the operating status data can be reduced while retaining the accuracy of the operating status data. By constructing a two-dimensional matrix based on the compressed operating status data segments and performing principal component analysis on the two-dimensional matrix, the target operating status data in the operating status data is determined. The important operating status data in the operating status data that plays a decisive role in data analysis can be accurately located to ensure the accuracy of the data analysis.
[0028] In some embodiments of the present invention, Figure 2 As shown, the running status data is segmented according to the continuous non-zero value running status data in the running status data to obtain multiple running status data segments, including: S201: Determine a zero value judgment threshold of the running status data according to the type of the running status data.
[0029] In an embodiment of the present invention, the types of operating status data include current, voltage, wire feed speed, shielding gas flow rate, etc. When determining whether the operating status data is zero, due to the presence of disturbances, the numerical value of the operating status data is not directly compared with 0. Instead, a zero-value judgment threshold that does not conform to the rules of the type of operating status data is determined based on the type of operating status data. For example, for welding current, because the welding current is generally around 160A, the zero-value judgment threshold can be 10A or other current values much smaller than 160A, so as to eliminate the influence of smaller disturbance currents. Furthermore, all zero-value judgment thresholds can be set to 0.01, which can quickly remove zero-value data while reducing the influence of disturbances.
[0030] S202 : Determine continuous running status data of each type of running status data at the same timestamp that is greater than the corresponding zero value judgment threshold as one running status data segment, and obtain multiple running status data segments.
[0031] In an embodiment of the present invention, after determining the zero value judgment threshold for each type of operating status data, the numerical value of each operating status data is compared with the zero value judgment threshold, the operating status data that is less than or equal to the zero value judgment threshold is eliminated, and the continuous non-zero value operating status data is determined as an operating status data segment, thereby obtaining multiple operating status data segments.
[0032] The embodiment of the present invention can greatly reduce the impact of disturbances on data screening and improve data accuracy by determining the corresponding zero-value judgment threshold according to the type of operating status data.
[0033] In some embodiments of the present invention, after obtaining the time series data of the operation of the gas shielded welding equipment, the method includes: The running status data is divided into blocks according to the volume of the running status data to obtain a plurality of running status data blocks, wherein each running status data block can be divided into a plurality of running status data segments in parallel.
[0034] In the embodiment of the present invention, since the gas shielded welding equipment generates a large amount of data during welding, in order to improve the data processing efficiency, the operating status data can be divided into blocks and processed in parallel. For example, according to the volume of the operating status data, the operating status data is divided into multiple operating status data blocks. For example, a group of operating status data has data, you can The running status data is divided into 100 data blocks for parallel processing. Each data block includes , which can greatly improve data processing efficiency.
[0035] In some embodiments of the present invention, Figure 3 As shown, when the running status data is processed in blocks, the following steps are also included: S301, determining the required zero value density in each operating status data block according to a preset compression ratio; S302 , performing block processing on the operating status data according to the required zero value density, so as to ensure that the zero value density of each operating status data block is greater than or equal to the required zero value density.
[0036] In an embodiment of the present invention, when a large amount of operating status data is processed in blocks, in order to ensure the compression efficiency of each subsequent data block, it is necessary to ensure the compression ratio of each data block. The compression ratio can be determined based on the zero value density in the data block. For example, for a data block with a zero value density of 0% to 50%, the compression ratio is 2 times, for a data block with a zero value density of 51% to 80%, the compression ratio is 5 times, and for a data block with a zero value density of 81% to 100%, the compression ratio is 10 times. Specifically, the corresponding required zero value density can be determined based on the preset compression ratio requirement. The zero value density refers to the ratio of the number of zero value data in a running status data block to the total number of data. After determining the required zero value density, the running status data is processed in blocks based on the required zero value density to ensure that the zero value density of each running status data block is greater than or equal to the required zero value density.
[0037] The embodiment of the present invention divides the running status data into blocks according to a preset compression ratio, ensuring that the compression ratio of each running status data block meets the requirements, thereby improving data compression efficiency.
[0038] In some embodiments of the present invention, constructing a two-dimensional matrix based on the compressed running status data segments includes: The running status data segments are sorted according to the timestamps corresponding to the running status data in the compressed running status data segments to generate a two-dimensional matrix, wherein each row in the two-dimensional matrix represents a type of running status data, and each data in the two-dimensional matrix includes the value of each running status data and the number of times the value appears.
[0039] In an embodiment of the present invention, continuing from the above embodiment, the compressed operating status data segments are shown in Table 2. Each type of operating status data in each segment is represented by a numerical value and the number of occurrences of the numerical value. The compressed operating status data is sorted according to the timestamp corresponding to each operating status data. One type of operating status data is used as a row of a two-dimensional matrix to generate a two-dimensional matrix.
[0040] Table 2: Compressed operating status data table
[0041] The embodiment of the present invention generates a two-dimensional matrix by sorting the compressed running status data segments, which facilitates the determination of subsequent target running status data.
[0042] In some embodiments of the present invention, Figure 4 As shown, principal component analysis is performed on the two-dimensional matrix to determine the target operating status data in the operating status data, including: S401, calculating a covariance matrix of each operating status data based on a two-dimensional matrix, and extracting eigenvalues and eigenvectors of the operating status data according to the covariance matrix; S402 : Determine target operating status data based on a feature vector corresponding to a feature value greater than a preset feature value threshold and each piece of operating status data in the compressed operating status data segment.
[0043] In the embodiment of the present invention, in order to ensure the accuracy of feature extraction, principal component analysis is used to first determine the threshold formula:
[0044] in, is the principal component analysis threshold, The parameter is reserved for information density, is the zero-value density, where the information density retention parameter is inversely correlated with the zero-value density. The higher the zero-value density, the smaller the information density retention parameter. The value range of the information density retention parameter is 0~1. For example, if the zero-value density of a data segment is 60%, then λ=0.8(1-0.6)=0.32, and the principal components with eigenvalues greater than 0.32 are retained.
[0045] Assume that the covariance matrix C is based on n samples and m new sample data blocks are added , the updated covariance matrix is:
[0046] in, is the new covariance matrix, is the mean of old data, represents the jth sample in the new data block.
[0047] The data processing flow is to first calculate the mean and covariance matrix of the old data based on the first batch of data; then, every time a new data block Xnew is received, the mean and covariance matrix are updated according to the above formula; finally, the eigenvalues and eigenvectors are recalculated regularly to maintain real-time performance.
[0048] In order to further reduce the impact of disturbance data on data analysis, the eigenvalues and eigenvectors of the operating status data can be determined through the covariance matrix, and the final target operating status data can be determined through the eigenvalues and eigenvectors. Specifically, the covariance matrix of the operating status data is first calculated using formula (1):
[0049] Where C is the covariance matrix, Refers to the value of the i-th operating status data, Refers to the frequency of the i-th running status data, k is the length of the compressed data segment, refers to the weighted mean, and n refers to the total number of samples in the data segment.
[0050] Based on this, the eigenvalues and eigenvectors of the covariance matrix can be determined by solving equation (2): (2) Where C is the covariance matrix, is the eigenvalue, I is the unit matrix, and by solving equation (2), the eigenvalue of the covariance matrix can be determined , then the eigenvalue Multiplying with the identity matrix I, we can get the eigenvector V.
[0051] Furthermore, because the eigenvalue of the covariance represents the variance of each eigenvector, that is, the distance of each eigenvector from the expected value, the larger the eigenvalue, the more the eigenvector can indicate the data characteristics. Therefore, the eigenvectors corresponding to the first k largest eigenvalues among the eigenvalues are selected, and combined with the compressed operating status data, the target operating status data can be determined.
[0052] In some embodiments of the present invention, Figure 5 As shown, determining target operating status data based on the eigenvector corresponding to the eigenvalue greater than the preset eigenvalue threshold and the operating status data includes: S501, performing standardization processing on each running status data in the compressed running status data segment to obtain standardized running status data; S502 , constructing a low-dimensional space based on eigenvectors corresponding to eigenvalues greater than a preset eigenvalue threshold, and projecting the standardized operating status data into the low-dimensional space using the eigenvectors to obtain target operating status data.
[0053] In the embodiment of the present invention, the operating status data needs to be standardized first, and the Z-score standardization method can be used, as shown in formula (3): (3) in, , is the standardized operating status data, is the average value of each operating status data, is the standard deviation of each operating status data, Refers to the value of the i-th operating status data, It refers to the frequency of the i-th running status data, and k is the length of the compressed data segment.
[0054] The standardized data Projecting into the new low-dimensional space composed of the selected eigenvectors, the principal components and residual components are obtained. The projection formula is shown in formula (4): (4) Among them, Y is the low-dimensional data after projection, and V is the feature vector.
[0055] In an embodiment of the present invention, after the standardized operating status data is projected into a low-dimensional space, the obtained low-dimensional data is the target operating status data. By analyzing the target status data, the operating status of the gas shielded welding equipment can be determined, which is convenient for management and maintenance.
[0056] The embodiment of the present invention compresses and reduces the dimension of the original gas shielded welding operation status data, thereby reducing data complexity while retaining important information, lowering data transmission and storage costs, and providing a guarantee for the accuracy of data analysis.
[0057] The present invention also provides a welding data acquisition hardware device, such as Figure 6As shown in the figure, the hardware device includes a welding data acquisition card and various sensor modules. The data acquisition card includes an STM32 data acquisition board and a MEGMEET welding data acquisition device. The STM32 acquisition card connects to voltage and current sensors via ADC and RS485 communication protocol to collect welding voltage, current, gas flow, and other data. It then uploads this data to an MQTT server via a serial port connection to a 4G module. The MEGMEET welding data acquisition device directly connects to the welding line via USB protocol to obtain data such as wire feed speed and uploads this data to a server via a built-in 4G module, enabling real-time collection and storage of crane welding process parameters. The main control chip is the STM32F407ZGT6, based on the Cortex-M4 core and running at a main frequency of 168MHz. This chip is equipped with a variety of peripheral interfaces to meet the needs of remote data transmission and monitoring of crane welding process parameters. The MEGMEET welding data acquisition device uses a DC arc welding power supply and has a rich set of peripheral interfaces, including RS232, RS485, CAN, and USB communication. It uses built-in sensors to collect real-time welding process data and uploads this data to a cloud server via Wi-Fi and 4G communication modules. The voltage sensor features a DC range of 0-150V, high accuracy of 0.1%, and strong anti-interference capabilities. It operates from a 24V supply voltage and outputs either a 4-20mA current signal or a 0-5V voltage signal. The current sensor uses a split-type Hall effect current transmitter with a DC range of 0-600A, a 24V supply voltage, and a 4-20mA current output signal. Its response time is <20µs, and its accuracy reaches 1% when the current exceeds 50A. The flow sensor uses the MF5708 high-precision gas flow meter with a range of 100L / min, a 12-24V supply voltage, and an accuracy of ±(2.0+0.5FS)%. It uses RS485 communication. The 4G module uses the Hezhou Air724UG-4G module, which supports 4G communication and offers a wide range of external interfaces, including UART, IIC, SPI, GPIO, and a SIM card slot. It also features a variety of low-power designs and efficient data transmission capabilities. It supports AT command control and TLS / SSL encryption protocols, making it widely used in smart devices, IoT terminals, and remote monitoring. The welder used is a Megmeet welder, which boasts a stable arc, real-time adjustment of welding current, voltage, and waveform, and supports high-frequency pulse welding mode to reduce spatter, making it particularly suitable for welding HG70 thin plates or complex structures. The welder also features flexible communication interfaces, including Device Net, CAN Open, MEGMEET CAN, RS485, Ethernet, and IP. It supports push-pull welding guns, with the welder directly driving the push-pull gun motor, maintaining torque and speed synchronization with the wire feeder motor. An STM32 data acquisition card receives voltage and 4-20mA current signals from the Hall effect sensor via an ADC, acquiring gas flow data via the RS485 communication protocol.The MEGMEET welding data logger connects to the welding machine via USB to obtain wire feed speed information. The 4G module connected via the serial port aggregates and uploads the data to the MQTT server.
[0058] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed by the present invention should be covered by the scope of protection of the present invention.
Claims
1. A method for processing operating status data of gas shielded welding equipment, characterized in that: include: Acquire time series data of the operation of the gas shielded welding equipment, wherein the time series data includes continuous time stamps and operation status data of the gas shielded welding equipment at each time stamp; Traversing the running status data based on continuous timestamps, and segmenting the running status data according to continuous non-zero values in the running status data to obtain a plurality of running status data segments; Run-length encoding is performed on each of the operating status data segments, and a two-dimensional matrix is constructed based on the compressed operating status data segments. Principal component analysis is performed on the two-dimensional matrix to determine target operating status data in the operating status data.
2. The method for processing operating status data of gas shielded welding equipment according to claim 1, characterized in that: The step of segmenting the running status data according to the continuous non-zero value running status data in the running status data to obtain a plurality of running status data segments includes: determining a zero value judgment threshold of the operating status data according to the type of the operating status data; Continuous running status data of each type of running status data at the same time stamp that are all greater than the corresponding zero value judgment threshold are determined as a running status data segment, thereby obtaining a plurality of running status data segments.
3. The method for processing operating status data of gas shielded welding equipment according to claim 1, characterized in that: After obtaining the time series data of the gas shielded welding equipment operation, the method includes: The running status data is divided into blocks according to the volume of the running status data to obtain a plurality of running status data blocks, wherein each of the running status data blocks can be divided into a plurality of running status data segments in parallel.
4. The method for processing operating status data of gas shielded welding equipment according to claim 3, characterized in that: When the operation status data is processed in blocks, the method further includes: determining the required zero value density in each operating status data block according to a preset compression ratio; The operation status data is processed in blocks according to the required zero value density to ensure that the zero value density of each operation status data block is greater than or equal to the required zero value density.
5. The method for processing operating status data of gas shielded welding equipment according to claim 1, characterized in that: The constructing of a two-dimensional matrix based on the compressed running status data segment includes: The compressed running status data segments are sorted according to the timestamps corresponding to the running status data in each segment to generate a two-dimensional matrix, wherein each row in the two-dimensional matrix represents a type of running status data, and each data in the two-dimensional matrix includes a value of each running status data and the number of times the value appears.
6. The method for processing operating status data of gas shielded welding equipment according to claim 5, characterized in that: The performing principal component analysis on the two-dimensional matrix to determine target operating status data in the operating status data includes: Calculating a covariance matrix of each of the operating status data based on the two-dimensional matrix, and extracting eigenvalues and eigenvectors of the operating status data according to the covariance matrix; Target operating status data is determined based on a feature vector corresponding to the feature value greater than a preset feature value threshold and each piece of operating status data in the compressed operating status data segment.
7. The method for processing operating status data of gas shielded welding equipment according to claim 6, characterized in that: The determining target operating status data based on the eigenvector corresponding to the eigenvalue greater than the preset eigenvalue threshold and the operating status data includes: performing standardization processing on each operating status data in the compressed operating status data segment to obtain standardized operating status data; A low-dimensional space is constructed based on the eigenvector corresponding to the eigenvalue greater than a preset eigenvalue threshold, and the standardized operating status data is projected into the low-dimensional space using the eigenvector to obtain the target operating status data.