Real-time acquisition method and system for welding heat input

Abstract

The embodiment of the application provides a real-time acquisition method and system of welding heat input, the method comprises the following steps: perceiving displacement movement of a welding torch or a welding wire through a diffuse reflection photoelectric switch array installed on the surface of a pipeline, and generating a corresponding pulse trigger signal; perceiving current data and voltage data of a welding circuit through a Hall sensor system, and generating a corresponding analog electric signal; synchronously acquiring the pulse trigger signal and the analog electric signal through a multi-channel data acquisition device, and converting the analog electric signal into a digital signal with a time stamp; calculating real-time welding speed data according to the time sequence of the pulse trigger signal and the preset spatial geometric relationship of a plurality of diffuse reflection photoelectric switches; performing multi-stage filtering processing on the current data and the voltage data; determining a unified time stamp, and performing time sequence matching on the filtered current data, the voltage data and the welding speed data based on the unified time stamp; and calculating a welding heat input value.

Description

Technical Field

[0001] This application relates to the technical field of mechanical design and manufacturing, specifically to a method and system for real-time acquisition of welding heat input. Background Technology

[0002] Welding heat input is a crucial parameter in the welding process, directly affecting the quality and mechanical properties of the weld. It is calculated from three main welding process parameters: welding current, welding voltage, and welding speed. During the welding of pipe butt joint circumferential welds, the steel pipe is in a fixed position, while the weld pool moves circumferentially. Due to the influence of gravity and changes in bevel dimensions, the welding heat input varies across different weld passes and positions in the pipe butt joint circumferential weld. Understanding the real-time welding heat input of the pipe butt joint circumferential weld is beneficial for monitoring whether the heat input exceeds the range specified in the welding procedure. Furthermore, real-time heat input data is helpful for production and research work related to welding process optimization and microstructure properties.

[0003] Currently, automated welding technology has successfully achieved real-time acquisition of welding process parameters for pipe butt ring welds. However, manual welding technology involves human control of the movement of the welding torch or electrode, and the welding current, welding voltage, and welding speed are not controllable. Therefore, it is necessary to develop a professional welding process parameter data acquisition and processing system to achieve real-time acquisition of welding heat input for manual welding of pipe butt ring welds.

[0004] Therefore, there is an urgent need for a solution to the above problems. Summary of the Invention

[0005] This application proposes a method and system for real-time acquisition of welding heat input to address the deficiencies of the prior art.

[0006] According to a first aspect of the embodiments of this application, a method for real-time acquisition of welding heat input is provided, comprising: The displacement of the welding torch or welding wire is sensed by a diffuse reflection photoelectric switch array installed on the surface of the pipe, and a corresponding pulse trigger signal is generated. The current and voltage data of the welding circuit are sensed by the Hall effect sensor system, and corresponding analog electrical signals are generated. Using a multi-channel data acquisition device with a unified hardware clock as the time reference, the pulse trigger signal and the analog electrical signal are acquired synchronously, and the analog electrical signal is converted into a digital signal with a timestamp. Based on the time sequence of the pulse trigger signal and the preset spatial geometric relationship of multiple diffuse reflection photoelectric switches, the real-time welding speed data is calculated. The current data and the voltage data are subjected to multi-stage filtering to eliminate noise interference; A unified timestamp is determined, and the filtered current data, voltage data, and welding speed data are time-matched based on the unified timestamp; The welding heat input value is calculated based on the matched current data, voltage data, and welding speed data.

[0007] In some embodiments, the diffuse reflection photoelectric switch array is mounted on the surface of the pipe via a chain-type fixing base, and the photosensitive part of each diffuse reflection photoelectric switch is directly facing the movement path of the welding gun or the welding wire.

[0008] In some embodiments, prior to sensing the displacement of the welding torch or welding wire via a diffuse reflective photoelectric switch array mounted on the pipe surface, the following is included: The position of each diffuse reflection photoelectric switch is adjusted by multiple three-dimensional fine-tuning gimbals to make the laser spot coincide with the center line of the weld. The distance between the front end of each diffuse reflection photoelectric switch and the surface of the pipe is measured using a gauge block, and the Z-axis of the multiple three-dimensional fine-tuning gimbals is adjusted so that the distance reaches the preset working range of the sensor. The angle between the central axis of each diffuse reflection photoelectric switch and the normal to the surface of the pipe is measured using a digital angle ruler, and the Y-axis and X-axis of the multiple three-dimensional fine-tuning gimbals are finely adjusted to ensure that the angle is 90°. Locking device for locking the three-dimensional fine-tuning gimbal.

[0009] In some implementations, the step of synchronously acquiring the pulse trigger signal and the analog electrical signal using a multi-channel data acquisition device with a unified hardware clock as the time reference includes: Using a multi-channel high-speed data acquisition card and data transfer terminals, the pulse trigger signal and the analog electrical signal are synchronously acquired with the hardware clock of the multi-channel high-speed data acquisition card as the absolute time reference. The sampling frequency ranges from 0 to 250 kHz.

[0010] In some implementations, the calculation of real-time welding speed data includes: The angle difference between the first and last diffuse reflection photoelectric switches is obtained by using an inclination meter, and the arc length is calculated based on the pipe diameter. Based on the time sequence of the pulse trigger signal, determine the time difference between the triggering of adjacent photoelectric switches by the welding torch or the welding wire; Based on the arc length and the time difference, calculate the real-time welding speed data.

[0011] In some embodiments, the multi-stage filtering process for the current data and the voltage data includes: The current data and the voltage data are processed sequentially by a Butterworth low-pass filter, a Chebyshev band-stop filter, and a Savitsky-Gore smoothing filter, respectively. The Butterworth low-pass filter is used for low-pass filtering, and its cutoff frequency is adaptively adjusted based on background noise spectrum analysis; the Chebyshev band-stop filter is used for band-stop filtering, and its stopband center frequency automatically tracks the power frequency and its harmonics; the Savitsky-Gore smoothing filter is used for smoothing filtering.

[0012] In some implementations, determining the uniform timestamp includes: The hardware clock of the data acquisition board is used as the absolute time reference for the entire system. Configure a hardware timing engine for analog input sampling, digital input counting, and digital output triggering; The current acquisition task is started at a fixed sampling rate, and the data of each sampling batch includes a sampling timestamp generated by hardware. The value of the encoder counter is read at each sampling interval and bound to the absolute timestamp of the time it is read; When the triggering conditions are met, the hardware timing engine performs a digital output operation to accurately record the pulse generation time.

[0013] In some embodiments, the method further includes: Armored shielded cables are used to transmit sensor signals, and an RC low-pass filter circuit is added at the hardware level to achieve anti-interference processing.

[0014] In some implementations, after calculating the welding heat input value, the following is included: Output and store the collected data; The collected data includes the timestamp, current data, voltage data, welding speed data, and heat input value, and is stored in a structured CSV file format.

[0015] According to a second aspect of this application, a real-time acquisition system for welding heat input is provided, comprising: The displacement sensing and signal triggering module is used to sense the displacement of the welding torch or welding wire through a diffuse reflection photoelectric switch array installed on the surface of the pipe, and generate a corresponding pulse trigger signal. The analog electrical signal generation module is used to sense the current and voltage data of the welding circuit through the Hall sensor system and generate corresponding analog electrical signals. The digital signal acquisition module is used to synchronously acquire the pulse trigger signal and the analog electrical signal through a multi-channel data acquisition device, using a unified hardware clock as the time reference, and convert the analog electrical signal into a digital signal with a timestamp. The welding speed determination module is used to calculate real-time welding speed data based on the time sequence of the pulse trigger signal and the preset spatial geometric relationship of multiple diffuse reflection photoelectric switches. A multi-stage filtering module is used to perform multi-stage filtering on the current data and the voltage data to eliminate noise interference; The timing matching module is used to determine a unified timestamp and perform timing matching between the filtered current data, the voltage data and the welding speed data based on the unified timestamp. The input value determination module is used to calculate the welding heat input value based on the matched current data, voltage data, and welding speed data.

[0016] The beneficial effects of the real-time acquisition method and system for welding heat input in the embodiments of this application include at least the following: This application's embodiments effectively solve the problem of inaccurate welding speed measurement due to human operation during manual welding by employing non-contact optical sensing technology. Its photoelectric switch array, based on the principle of diffuse reflection, reliably captures the welding torch displacement and generates pulse signals with high timing accuracy, providing a stable data source for speed calculation. Hall effect sensing technology enables safe and accurate acquisition of welding current and voltage, and its electrical isolation characteristics completely avoid the risk of high-voltage intrusion, ensuring the safety of the system and operators. The hardware synchronous acquisition architecture ensures the timing consistency of multi-source heterogeneous signal acquisition, and a unified hardware clock reference provides accurate timestamps for all data, laying a solid foundation for subsequent multi-parameter fusion calculations. The speed calculation method based on pulse time series and spatial geometric relationships cleverly solves the displacement measurement problem on curved circumferential welds, outputting a true reflection of the instantaneous speed value of the welding torch movement through real-time calculation of arc length and time difference. By innovatively employing a multi-level digital filter chain to process the electrical signal, high-frequency noise and power frequency interference are suppressed sequentially, and the waveform is smoothed. While removing noise, the true signal characteristics are preserved to the maximum extent, ensuring the quality of the original data used for thermal input calculations. By using a timestamp matching algorithm to precisely align data from different physical sources in the time dimension, it ensures that each set of current, voltage, and speed data strictly corresponds to the same welding moment, eliminating calculation errors caused by transmission delays. Finally, based on the strictly matched multi-source data, the welding heat input value is calculated in real time, realizing precise quantitative monitoring of the manual welding process. This provides reliable data support for welding quality assessment and process optimization, effectively solving the long-standing problem of real-time heat input monitoring in this field. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the real-time acquisition method for welding heat input according to an embodiment of this application. Figure 2 This is a flowchart illustrating a specific implementation of the real-time acquisition method for welding heat input according to an embodiment of this application. Figure 3 This is a schematic diagram of the structure of a real-time acquisition system for welding heat input according to an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed embodiments of the present application, but merely to illustrate selected embodiments of the present application. Other embodiments obtained by those skilled in the art based on the embodiments of the present application without inventive effort are all within the scope of protection of the embodiments of the present application.

[0020] It can be noted that similar reference numerals and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it will not be further defined and explained in subsequent figures according to the embodiments of this application.

[0021] This application discloses a real-time acquisition method for welding heat input, which is executed based on a real-time welding heat input acquisition system. This method is applicable to the manual welding process of pipe butt circumferential welds. By developing this real-time acquisition system for the manual welding process of pipe butt circumferential welds, real-time acquisition of welding heat input can be achieved, which is beneficial for production and scientific research work such as monitoring heat input during the welding process, optimizing welding processes, and studying microstructure properties. (See attached figure) Figure 1 As shown, the method includes steps 110-170.

[0022] Step 110: The displacement of the welding torch or welding wire is sensed by a diffuse reflection photoelectric switch array installed on the surface of the pipe, and a corresponding pulse trigger signal is generated.

[0023] Among them, the diffuse reflection photoelectric switch array installed on the surface of the pipe is the core sensing component for realizing non-contact real-time measurement of welding speed. Its purpose is to convert the displacement motion of the welding torch or welding wire on the circumferential weld into a series of pulse trigger signals that can be accurately timed through the optical sensing principle.

[0024] See attached document Figure 2 As shown, the diffuse reflection photoelectric switch array can be mounted on an object capturing mechanism, which includes a chain-type fixing base and multiple three-dimensional fine-tuning gimbals. The chain-type fixing base consists of two high-strength engineering plastic chains (different lengths can be selected according to the pipe diameter), a quick-lock tensioner, and a rigid connecting plate. The chains wrap around the pipe and are tightened by the quick-lock tensioner, thus rigidly fixing the entire system to the pipe. Multiple three-dimensional fine-tuning gimbals are connected to the rigid connecting plate, providing linear fine-tuning slides in the X (horizontal), Y (vertical), and Z (telescopic) directions, as well as rotational adjustment (Pan) around the Z-axis. Each degree of freedom is equipped with a precision lead screw and scale, achieving an adjustment accuracy of ±0.1mm.

[0025] In some embodiments, the diffuse reflection photoelectric switch array is mounted on the surface of the pipe via a chain-type fixing base, and the photosensitive part of each diffuse reflection photoelectric switch is directly facing the movement path of the welding gun or the welding wire.

[0026] In some embodiments, prior to sensing the displacement of the welding torch or welding wire using an array of diffuse reflection photoelectric switches mounted on the pipe surface, the process includes: adjusting the position of each diffuse reflection photoelectric switch using multiple three-dimensional fine-tuning gimbals to align the laser spot with the weld centerline; measuring the distance between the front end of each diffuse reflection photoelectric switch and the pipe surface using a gauge block, and adjusting the Z-axis of the multiple three-dimensional fine-tuning gimbals to bring the distance within a preset working range of the sensor; measuring the angle between the central axis of each diffuse reflection photoelectric switch and the normal to the pipe surface using a digital angle gauge, and fine-tuning the Y-axis of the multiple three-dimensional fine-tuning gimbals. A shaft is used to ensure that the included angle is 90°; a locking device is used to lock the fine-tuning gimbal.

[0027] For example, the above-mentioned adjustment of the position of each diffuse reflection photoelectric switch by multiple three-dimensional fine-tuning gimbals is used to make the laser spot coincide with the center line of the weld, including the generation of three-dimensional coordinates of the welding gun or welding wire and the mapping and matching of multiple three-dimensional fine-tuning gimbals with the weld.

[0028] For example, the three-dimensional fine-tuning gimbal provides linear fine-tuning in the X, Y, and Z directions and rotational adjustment around the Z-axis. Each degree of freedom is equipped with a precision lead screw and scale, and the adjustment accuracy reaches ±0.1mm.

[0029] For example, the generation of the three-dimensional coordinates of a welding torch or welding wire can be assumed, for instance, to be primarily in a two-dimensional plane. The system moves along a preset weld seam trajectory on the Y-plane, while simultaneously acquiring real-time height (Z-axis) information via a Z-axis linear encoder or laser rangefinder. Coordinate update algorithm (for...) The -Y plane can be shown as follows: ; ; ; This can be achieved by reading Z-axis sensor values ​​or by incremental calculation using the Z-axis encoder. This is the angle between the current welding torch movement direction and the X-axis. This angle can be provided in real time by the motion control system, or approximately calculated based on the coordinate changes at two consecutive time points. ).

[0030] For example, the mapping and matching of multiple 3D fine-tuning gimbals with weld seams can be configured, for instance, by creating an ideal 3D path model of a pipe weld seam, which can be represented as a series of discrete point coordinates. A set. Real-time matching algorithms, for example, can calculate the current welding torch coordinates. The Euclidean distances to all points along the preset weld path are used to find the closest point. Among them, distance The following formula is used to calculate: ; Additionally, if the distance If the deviation exceeds the tolerance range (e.g., exceeding 2.0 mm), an early warning can be issued or the motion control system can be notified to correct the deviation, so as to ensure the welding quality and the effectiveness of the monitoring position.

[0031] In some embodiments, the sensor mounting bracket is fixed to the photoelectric switch in this application. To ensure that the light spot of the photoelectric switch (or laser positioner) is accurately aligned with the weld centerline and maintains the optimal working distance and angle, this application follows the standardized installation and calibration process below. The specific installation and calibration process includes: initial fixing, coarse adjustment - aligning with the weld centerline, fine adjustment - ensuring distance and angle, and system calibration and verification.

[0032] For example, initial fixation includes: wrapping the chain around the pipe to be welded, ensuring that the photoelectric switch is roughly facing the weld area; manually tightening the chain and locking it with a quick-lock tensioner to make the main body of the system stable and without shaking.

[0033] For example, coarse adjustment-alignment of the weld centerline involves activating the positioning laser (or using a separate calibration laser) with a photoelectric switch, which creates a visible spot on the pipe surface; manually observing the position of the spot relative to the weld bevel centerline; loosening the X-axis and Pan locking knobs of the fine adjustment gimbal and adjusting its position until the laser spot coincides with the weld centerline.

[0034] For example, fine-tuning – ensuring distance and angle calibration includes distance calibration and angle calibration. Distance calibration involves using a gauge block or dedicated thickness gauge to measure the distance D between the front end of the photoelectric switch and the pipe surface, and adjusting the Z-axis (telescopic) fine-tuning slide until distance D reaches the sensor's optimal operating range (e.g., 20 ± 2 mm as required in the product manual). Angle calibration involves using a digital angle gauge to measure the angle between the photoelectric switch's central axis and the normal to the pipe surface at that point, and adjusting the Y-axis (vertical) and X-axis (horizontal) fine-tuning of the pan-tilt unit to ensure this angle is 90° (i.e., the sensor is directly facing the pipe surface). The purpose is to ensure the accuracy of the measured distance and the precision of the trigger position.

[0035] For example, system calibration and verification include: after completing the above mechanical adjustments, locking all the locking devices of the fine-tuning gimbals; starting the system software and entering "calibration mode"; moving the welding torch past the photoelectric switch and observing whether a trigger signal is generated in the software interface the instant the welding torch passes the light spot; moving the welding torch parallel to both sides of the weld centerline to verify the repeatability of the trigger point. Ideally, the trigger point should always be on the weld centerline with an error of less than ±0.5mm.

[0036] This application employs non-contact measurement, eliminating the need for any additional devices on the welding torch and avoiding any interference with the welding operation, thus ensuring the flexibility of manual welding. It features a highly anti-interference design, utilizing modulated laser and synchronous detection technology to effectively resist the effects of strong arc light, electromagnetic interference, and fumes at the welding site, ensuring high reliability. It is applicable to complex curved surfaces; the flexible chain design allows the sensor array to perfectly conform to the curved surfaces of pipes of different diameters, solving the problem that planar measurement systems cannot be used for circumferential welds. Furthermore, the photoelectric switch's response time is typically in the millisecond or even microsecond range, accurately capturing the instantaneous position of the welding torch and providing a high-precision pulse signal for calculating instantaneous velocity. The sensors typically have a high protection rating (e.g., IP67), and an anti-spatter quartz glass protective plate can be added in front, enabling this application to adapt to harsh working conditions at the welding site.

[0037] Step 120: The current and voltage data of the welding circuit are sensed by the Hall sensor system, and the corresponding analog electrical signals are generated.

[0038] The Hall effect sensor system of this application is a core component for achieving high-precision acquisition of welding electrical parameters. Its technical feature lies in non-contact sensing of current and voltage data in the welding circuit and converting them into analog electrical signals that can be processed by subsequent data acquisition systems.

[0039] This Hall sensor system operates based on the Hall effect principle. When a current-carrying conductor or semiconductor is placed in a magnetic field, and the direction of the current is perpendicular to the direction of the magnetic field, a measurable potential difference, known as the Hall voltage, is generated in the direction perpendicular to both the current and the magnetic field. This voltage signal is proportional to the magnetic field strength (and consequently, to the current that generates the magnetic field). By measuring this voltage, the original current or voltage parameters can be measured indirectly and in isolation.

[0040] In some implementations, the Hall effect sensing system consists of two independent but coordinated units: the sensing and conversion of welding current and the sensing and conversion of arc voltage.

[0041] For example, welding current can be sensed using a Hall effect current sensor. This sensor is typically a ring or open structure, containing a Hall element and a magnetic core. It can be mounted, for example, directly clipped onto the welding ground cable. When the welding current (a strong current) flows through the cable, it generates a magnetic field in the sensor core with a strength proportional to the current. The signal generation can be achieved, for example, by detecting this magnetic field through the Hall element and outputting a weak analog voltage signal. This signal is amplified and normalized by the signal conditioning circuitry (such as an amplifier) ​​integrated within the sensor, ultimately outputting a low-impedance analog voltage signal (e.g., 0-5V or 4-20mA) precisely proportional to the welding current, which is then sent to the data acquisition card.

[0042] For example, the arc voltage can be sensed using a Hall effect voltage sensor. Internally, it is also based on the Hall effect, but the input is a voltage signal. Its installation method can be, for example, by connecting the positive and negative terminals of its input (high-voltage side) to the contact tip of the welding torch and the grounding wire of the workbench, respectively, thereby directly measuring the potential difference between these two points, i.e., the arc voltage. Signal generation can be achieved, for example, by converting the high voltage into a small current signal through a precision voltage divider network within the sensor. This current flows through the built-in magnetic core coil to generate a magnetic field, which is then detected by the Hall element and converted into an analog voltage signal output proportional to the arc voltage.

[0043] This application's embodiment achieves electrical isolation by employing a Hall effect sensor system. The Hall sensor achieves complete electrical isolation between the primary side (high-voltage, high-current welding circuit) and the secondary side (low-voltage, low-voltage signal acquisition circuit). This greatly improves system safety, effectively preventing high-voltage intrusion from harming data acquisition equipment and operators, while also fundamentally suppressing common-mode interference. The Hall sensor has a wide bandwidth response, accurately capturing rapid and dynamic changes in current and voltage during welding, providing a reliable data foundation for subsequent calculations of the actual instantaneous heat input. It eliminates the need to disconnect the original welding circuit for series or parallel connection, achieving truly "non-invasive" measurement. This not only simplifies the installation process and avoids impacting the performance of the original welding system, but also ensures that the measurement process itself does not introduce additional losses or risks. Combined with the overall shielding, partial shielding (permalloy shielding cover), and single-point grounding hardware anti-interference design of this application, the Hall effect sensor system can stably output analog electrical signals with a high signal-to-noise ratio in the strong electromagnetic interference environment of the welding site.

[0044] Step 130: Using a multi-channel data acquisition device with a unified hardware clock as the time reference, the pulse trigger signal and the analog electrical signal are acquired synchronously, and the analog electrical signal is converted into a digital signal with a timestamp.

[0045] In some implementations, the step of synchronously acquiring the pulse trigger signal and the analog electrical signal using a multi-channel data acquisition device with a unified hardware clock as the time reference includes: synchronously acquiring the pulse trigger signal and the analog electrical signal using a multi-channel high-speed data acquisition card and a data adapter terminal, with the hardware clock of the multi-channel high-speed data acquisition card as the absolute time reference.

[0046] For example, the sampling frequency is 0 to 250 kHz.

[0047] This application embodiment achieves high-precision synchronous acquisition and digitization of pulse trigger signals and analog electrical signals through a multi-channel data acquisition device. This feature is the core to ensure the accuracy of welding heat input calculation.

[0048] This application employs a multi-channel high-speed data acquisition card (DAQ) as the core acquisition unit. Its hardware design supports synchronous sampling of digital input channels (receiving photoelectric switch pulse signals) and analog input channels (receiving Hall sensor current and voltage signals), physically eliminating time differences between signals. The entire system uses the high-stability hardware clock built into the data acquisition card as the absolute time reference. This clock is driven by a high-precision crystal oscillator, providing a unified, microsecond (μs) accurate time reference for every sampling of all input channels, laying the foundation for the consistency of data timing throughout the system. Key acquisition tasks are directly executed by the hardware timing engine on the DAQ board, making it completely independent of the computer operating system. This avoids timing uncertainties and delays caused by software scheduling and system load fluctuations, ensuring the real-time and deterministic nature of the acquisition. Under the control of the hardware clock, the analog input channels perform analog-to-digital conversion (ADC) on the current and voltage signals at a constant sampling rate (e.g., 10kHz); simultaneously, the digital input channels accurately capture the instant of each pulse edge transition. Each conversion and capture event is automatically marked with a precise timestamp under the same time base. Before entering the acquisition card, the signal has already undergone preliminary processing by the conditioning circuit inside the sensor and the front-end RC filter circuit on the PCB board, effectively suppressing high-frequency noise and interference introduced during transmission, providing a clean signal source for subsequent high-precision digital conversion. The acquisition card driver (such as NI-DAQmx) packages the converted digital signal values ​​(current, voltage values, pulse level states) with the corresponding hardware timestamps to form a time-stamped data stream, which is uploaded in real time to the data processing unit (industrial control computer) via a high-speed bus (such as PCIe) for subsequent filtering, calculation, and matching. In summary, this multi-channel synchronous acquisition device, through careful hardware design, achieves precise synchronization and digitization of multi-source heterogeneous signals, providing a crucial data foundation for ultimately achieving perfect matching of welding current, voltage, and speed in the time dimension and accurately calculating instantaneous thermal input values.

[0049] Step 140: Based on the time sequence of the pulse trigger signal and the preset spatial geometric relationship of multiple diffuse reflection photoelectric switches, calculate the real-time welding speed data.

[0050] In some embodiments, the calculation of real-time welding speed data includes: obtaining the angle difference between the first and last diffuse reflection photoelectric switches using an inclination meter, and calculating the arc length based on the pipe diameter; determining the time difference between the welding torch or the welding wire triggering adjacent photoelectric switches according to the time sequence of the pulse trigger signal; and calculating real-time welding speed data based on the arc length and the time difference.

[0051] For example, when the welding torch sequentially triggers each equally spaced photoelectric switch, a series of precise pulse signals are generated. The system records the precise moment of each pulse occurrence via a data acquisition card, forming a time series. , , , ···, ).

[0052] For example, calculate the time difference between the welding torch passing two adjacent photoelectric switches. This time difference has a direct correspondence with the fixed arc length distance (L) between the two switches.

[0053] For example, the real-time welding speed according to the formula This allows the instantaneous linear velocity of the welding torch at that location to be calculated. This calculation process is automatically completed by the system software and continuously updated, thus outputting a real-time velocity data stream synchronized with the movement of the welding torch.

[0054] This system achieves real-time calculation of welding speed by analyzing the time series of pulse trigger signals and combining them with the preset spatial geometric relationship of photoelectric switches.

[0055] Step 150: Perform multi-stage filtering on the current data and the voltage data to eliminate noise interference.

[0056] In some embodiments, the multi-stage filtering process for the current data and the voltage data includes: processing the current data and the voltage data sequentially through a Butterworth low-pass filter, a Chebyshev band-stop filter, and a Savitzky-Golay smoothing filter.

[0057] For example, the Butterworth low-pass filter is used for low-pass filtering, and the cutoff frequency of the Butterworth low-pass filter is adaptively adjusted based on background noise spectrum analysis.

[0058] For example, the Chebyshev band-stop filter is used for band-stop filtering, and the stopband center frequency of the Chebyshev band-stop filter automatically tracks the power frequency and its harmonics.

[0059] For example, the Savitsky-Gore smoothing filter is used for smoothing filtering.

[0060] For example, referring to the parameters in Table 1 below, a multi-stage filtering strategy can be used to process current signals in the following order: In the preprocessing stage, a Butterworth low-pass filter is used to remove high-frequency noise and retain the main trend of the current signal. Due to its flat passband response, it is suitable as a preprocessing filter and can avoid introducing phase distortion. Before feature extraction, a Chebyshev band-stop filter is used to further filter out interference in specific frequency bands and improve signal quality. It has a steep roll-off characteristic and can effectively suppress interference in specific frequency bands, so it is suitable for filtering out power frequency interference or arc noise. In the time-series smoothing stage, a Savitsky-Gore smoothing filter is used to smooth the signal, which facilitates the extraction of time-series features (such as kurtosis, waveform factors, etc.) while retaining the local features of the signal. With its conformal smoothing characteristics, it can retain peak values ​​and trends while smoothing the signal, so it is suitable for the input preprocessing of time-series prediction models.

[0061] Table 1: Current Signal Processing Table

[0062] The embodiments of this application are based on the sequential execution of the above-mentioned third-order filters, which work together to form a complete processing chain from broadband suppression to specific frequency rejection and waveform shape preservation smoothing. The final output is smooth, stable and truly reflects the essence of the welding process, laying a solid foundation for subsequent accurate calculations.

[0063] Step 160: Determine a unified timestamp, and perform time-series matching of the filtered current data, voltage data, and welding speed data based on the unified timestamp.

[0064] In order to achieve strict synchronization and fusion of current signals, velocity data and spatial location information, this application can construct a unified timestamp system based on hardware triggering to ensure that all data are marked under the same high-precision time base, with synchronization accuracy reaching the microsecond (μs) level.

[0065] In some implementations, determining the unified timestamp includes: using the hardware clock of the data acquisition board as the absolute time reference for the entire system; configuring a hardware timing engine for analog input sampling, digital input counting, and digital output triggering; starting the current acquisition task at a fixed sampling rate, with each batch of data carrying a hardware-generated sampling timestamp; reading the value of the encoder counter at each sampling interval and binding it to the absolute timestamp of the time being read; and when the triggering condition is met, executing a digital output operation by the hardware timing engine to accurately record the pulse generation time.

[0066] For example, the core clock source uses the backplane clock of NI's high-performance embedded controllers (such as CompactRIO) or the PXIe chassis as the absolute time reference for the entire system. This clock is provided by a high-stability crystal oscillator with extremely low jitter and drift. The data acquisition card (DAQ) has an independent hardware timing engine. All critical timing and triggering tasks (such as analog sampling and digital input counting) are executed directly by these hardware engines, completely independent of the computer's operating system clock, thus avoiding timing uncertainties and delays caused by software scheduling, system load, and other factors.

[0067] Different synchronization and stamping strategies are adopted for data sources with different characteristics, as shown in Table 2 below: Table 2: Data Source Synchronization and Stamping Strategies

[0068] For example, determining a unified timestamp includes initialization synchronization and data stream stamping. Initialization synchronization, for instance, can be achieved after system power-on by configuring the data acquisition card's clock source to synchronize with the backplane clock of the central controller (such as CRIO) via the NI-DAQmx driver API, thus establishing a unified time base for the entire system. Data streams and stamping include current acquisition, position acquisition, and photoelectric triggering. For example, current acquisition can be performed at a fixed sampling rate. (e.g., 10kHz) Initiate a hardware-timed analog input task. Each sampling batch of data includes hardware-generated data relative to... Precise timestamps; location acquisition, for example, can be configured with a hardware-timed digital input task for encoder 4x frequency counting, with the counter value at each sampling interval. Read once and compared with the absolute timestamp at that moment. Binding; photoelectric triggering, for example, allows the software to immediately command the DAQ board to generate a TTL pulse from its digital output line when the decision algorithm meets the conditions. This operation is performed by a hardware timing engine, with a deterministic and extremely small response delay (typically <1μs). The generation time of this pulse... It was recorded precisely.

[0069] For example, time-series matching of the filtered current data, voltage data, and welding speed data based on the unified timestamp includes: associating all information (the current data, voltage data, and welding speed data) using the unified timestamp. For example, for a single image... The time-triggered signal can precisely find the timestamp closest to the nearest one. The current sampling point is used to obtain the current value at the moment of triggering. Then extract the time window. All current data within the data are used for subsequent data processing to ultimately obtain... The spatial coordinates of the welding torch calculated by the encoder at any time This enables precise positioning.

[0070] This embodiment establishes a high-precision unified time reference through the hardware clock of the data acquisition card. This clock, driven by a stable crystal oscillator, provides an absolute timestamp accurate to the microsecond level for every sampling event of all input channels (including analog and digital inputs). After analog-to-digital conversion, each sampling point of the current and voltage analog electrical signals is automatically marked with this hardware timestamp. Simultaneously, the edge transition moment of the pulse signal generated by the photoelectric switch triggered by the welding torch is also captured by the hardware timing engine and marked with the same reference timestamp. The filtered current and voltage data and the calculated welding speed data both carry this unified timestamp, and the system performs timing matching based on these timestamps. The matching algorithm compares the timestamps of data points and associates current, voltage, and speed values ​​with the same timestamp or timestamp differences within a preset allowable error range (e.g., ±1ms). This precise timing matching ensures that the welding electrical parameters (current, voltage) and motion parameters (speed) are completely synchronized in the time dimension, providing a strictly consistent data foundation for subsequent calculations of instantaneous welding heat input and avoiding calculation errors introduced by signal transmission or processing delays.

[0071] Step 170: Calculate the welding heat input value based on the matched current data, voltage data, and welding speed data.

[0072] In some embodiments, the method further includes: using armored shielded cables to transmit sensor signals and adding an RC low-pass filter circuit at the hardware level to achieve anti-interference processing.

[0073] In some implementations, after calculating the welding heat input value, the process includes outputting and storing the acquired data.

[0074] For example, the collected data includes the timestamp, the current data, the voltage data, the welding speed data, and the heat input value, and is stored in a structured CSV file format.

[0075] For example, after each successful trigger, a standardized data record is generated. All records are eventually aggregated to generate a CSV file named with a timestamp (compatible with Excel, but the CSV format is more conducive to automated processing). Its data structure is clear, and the core fields that can be included, as shown in Table 3, are as follows: Table 3: Core Fields of CSV File

[0076] This application also includes post-processing and analysis, which refers to importing the generated CSV file into an independent data analysis platform for processing and calculation, ultimately forming welding heat input data.

[0077] In some implementations, this application also incorporates comprehensive anti-interference design in terms of hardware design and software algorithms, as well as optical anti-interference measures and robust outlier handling mechanisms to ensure the accuracy of the system's collected data and operational stability.

[0078] For example, hardware-level anti-interference measures include overall shielding, partial shielding, signal filtering, and grounding. Overall shielding includes using an all-metal chassis (such as aluminum alloy) and designing it as an electromagnetically shielded enclosure; all interfaces (power, communication, and sensors) use shielded armored cables to ensure 360° shielding. Partial shielding includes adding a permalloy shield to the Hall current sensor and its preamplifier circuit. Signal filtering includes designing an RC low-pass filter circuit on the PCB board as a hardware front-end filter before the sensor signal enters the data acquisition card (DAQ), serving as the first line of defense for software filtering. Grounding includes using a single-point grounding scheme. All shielding layers and chassis grounds converge at a single point and are connected to the field grounding stake via a thick conductor to avoid forming grounding loops that introduce common-mode interference.

[0079] For example, optical anti-interference measures include anti-splash protection, which includes installing a replaceable transparent quartz glass protective sheet in front of the photoelectric switch to prevent welding spatter from contaminating and damaging the device.

[0080] For example, the software algorithm layer's anti-interference enhancement includes configuring an adaptive filter chain (Butterworth-Chebyshev-Savitzky-Golay) and dynamic tracking of the power frequency notch filter. The adaptive filter chain parameters can be fine-tuned online according to the on-site interference conditions. The Butterworth cutoff frequency is adaptive: during system initialization, a short period of background noise (welding machine unloaded) is collected, and its spectrum is analyzed using FFT. The cutoff frequency of the Butterworth low-pass filter is automatically set to 1.5 times the dominant noise frequency to achieve optimal high-frequency noise suppression. The dynamic tracking of the power frequency notch filter includes setting the stopband center frequency of the Chebyshev band-stop filter to automatically track the 50 / 60Hz power frequency and its main harmonics (100 / 120Hz, 150 / 180Hz), effectively suppressing even slight fluctuations in the on-site power grid frequency.

[0081] For example, the robust outlier handling mechanism includes transient pulse rejection, which uses median filtering preprocessing within the sliding window of current feature extraction to effectively reject transient spike pulses caused by the start-up and shutdown of high-power equipment; and signal loss diagnosis, for example, the software continuously monitors the amplitude and variance of the current signal. If the signal amplitude is zero for a long time or the variance is extremely low, it is determined that the sensor cable is detached or the power is cut off, and the system alarm is triggered instead of continuing to make incorrect predictions.

[0082] This application embodiment calculates welding heat input values ​​in real time based on precisely matched multi-source data. The calculation process strictly follows the basic principles of welding technology. The system uses a high-precision time-series matching algorithm to ensure that the current, voltage, and speed values ​​at each calculation moment completely correspond to the same welding instant, thus guaranteeing the spatiotemporal consistency of the three key parameters in the calculation formula. The calculation process is dynamically refreshed at a frequency exceeding 500Hz, accurately capturing instantaneous changes in heat input caused by operational fluctuations during manual welding. To adapt to different welding process requirements, the system has a built-in configurable thermal efficiency coefficient η database, allowing users to select appropriate parameters based on the welding materials, shielding gas, and bevel type used. The final calculated welding heat input value, along with its corresponding original data, timestamp, and spatial location information, is recorded and output, providing a complete data foundation for welding quality traceability, process optimization, and academic research.

[0083] This application's embodiments effectively solve the problem of inaccurate welding speed measurement due to human operation during manual welding by employing non-contact optical sensing technology. Its photoelectric switch array, based on the principle of diffuse reflection, reliably captures the welding torch displacement and generates pulse signals with high timing accuracy, providing a stable data source for speed calculation. Hall effect sensing technology enables safe and accurate acquisition of welding current and voltage, and its electrical isolation characteristics completely avoid the risk of high-voltage intrusion, ensuring the safety of the system and operators. The hardware synchronous acquisition architecture ensures the timing consistency of multi-source heterogeneous signal acquisition, and a unified hardware clock reference provides accurate timestamps for all data, laying a solid foundation for subsequent multi-parameter fusion calculations. The speed calculation method based on pulse time series and spatial geometric relationships cleverly solves the displacement measurement problem on curved circumferential welds, outputting a true reflection of the instantaneous speed value of the welding torch movement through real-time calculation of arc length and time difference. By innovatively employing a multi-level digital filter chain to process the electrical signal, high-frequency noise and power frequency interference are suppressed sequentially, and the waveform is smoothed. While removing noise, the true signal characteristics are preserved to the maximum extent, ensuring the quality of the original data used for thermal input calculations. By using a timestamp matching algorithm to precisely align data from different physical sources in the time dimension, it ensures that each set of current, voltage, and speed data strictly corresponds to the same welding moment, eliminating calculation errors caused by transmission delays. Finally, based on the strictly matched multi-source data, the welding heat input value is calculated in real time, realizing precise quantitative monitoring of the manual welding process. This provides reliable data support for welding quality assessment and process optimization, effectively solving the long-standing problem of real-time heat input monitoring in this field.

[0084] This application also discloses a real-time acquisition system for welding heat input. (See attached document.) Figure 3 As shown, the system includes: a displacement sensing and signal triggering module 310, an analog electrical signal generation module 320, a digital signal acquisition module 330, a welding speed determination module 340, a multi-level filtering processing module 350, a timing matching module 360, and an input value determination module 370.

[0085] For example, the displacement sensing and signal triggering module 310 is used to sense the displacement movement of the welding torch or welding wire through a diffuse reflection photoelectric switch array installed on the surface of the pipe, and generate a corresponding pulse trigger signal.

[0086] For example, the analog electrical signal generation module 320 is used to sense the current and voltage data of the welding circuit through the Hall sensor system and generate the corresponding analog electrical signal.

[0087] For example, the digital signal acquisition module 330 is used to synchronously acquire the pulse trigger signal and the analog electrical signal through a multi-channel data acquisition device with a unified hardware clock as the time reference, and convert the analog electrical signal into a digital signal with a timestamp.

[0088] For example, the welding speed determination module 340 is used to calculate real-time welding speed data based on the time sequence of the pulse trigger signal and the preset spatial geometric relationship of multiple diffuse reflection photoelectric switches.

[0089] For example, the multi-stage filtering module 350 is used to perform multi-stage filtering on the current data and the voltage data to eliminate noise interference.

[0090] For example, the timing matching module 360 ​​is used to determine a unified timestamp and perform timing matching between the filtered current data, the voltage data and the welding speed data based on the unified timestamp.

[0091] For example, the input value determination module 370 is used to calculate the welding heat input value based on the matched current data, voltage data and welding speed data.

[0092] In some embodiments, this application uses a test X80 D1219×18.4mm pipe and Kobelco LB-70L welding rods. The test results show that the welding speed sampling frequency is 20Hz / channel; the welding speed range is 7.23cm / min-9.4cm / min, the speed error is ±1.5cm / min, and the response time is ≤10ms; the welding current range is 94.2A-112.5A, the linearity is ≤0.2%FS, and the sampling frequency is 500Hz; the arc voltage range is 19.7-26.6V, the linearity is ≤0.1%FS, and the sampling frequency is 500Hz.

[0093] In some implementations, the system also includes an object capture mechanism and a data acquisition and analysis system.

[0094] The object capture mechanism's main component is a laser diffuse reflection photoelectric switch. The switch is mounted on a double-chain, with equal spacing (s) between the chain segments. The chains can rotate relative to each other via buckles, allowing the entire mechanism to bend according to the weld seam. A clamp secures the chains to the pipe surface. The photoelectric switch is mounted at one end of the welding torch, ensuring the photosensitive area faces the object. The horizontal and vertical distance between the photoelectric switch and the weld seam centerline is adjusted according to the switch's effective detection distance. Photoelectric switches of the same specification have identical laser aperture and receiver depth, resulting in the same welding speed reading for objects at different distances.

[0095] For example, welding speed acquisition utilizes diffuse reflection photoelectric switches. When a detected object passes by, a sufficient amount of light emitted by the photoelectric switch transmitter is reflected to the receiver, generating on / off pulse signals. These pulse signals are used to detect changes in the welding torch's position, thereby measuring the welding speed. When multiple laser switches are equidistantly arranged around a weld seam, the welding torch (wire) sequentially generates on / off pulse signals through the switches. These pulse signals are acquired using a multi-channel data acquisition card, and the time difference between the on / off pulse signals is obtained using LabVIEW software. An inclinometer is used to obtain the arc length of the laser switch's start and end positions, thus calculating the welding speed. In the welding process of pipe butt ring welds, the detection surface is curved, and the spacing between photoelectric switches is calculated by measuring the arc. Furthermore, a spatial angle electronic measuring instrument is used to measure the angles of the first and last photoelectric switches, calculating their arc difference. Based on the pipe diameter and chain hole height, the relative positions of the photoelectric switches are obtained, thus calculating the photoelectric switch spacing. The inclinometer's angular resolution is 0.01°.

[0096] In some implementations, the data acquisition and analysis system may include, for example, a pulse generated by the object-catching mechanism when an object moves to the corresponding photoelectric switch. This application uses an NPN normally open type photoelectric switch, which continuously emits a positive voltage signal in normal operation and outputs 0V when an object is detected. By designing all the object-catching mechanisms in parallel in the circuit, the pulses from each object-catching mechanism can be connected in series to form a series of pulse signals. These pulse signals are then acquired and analyzed using a data acquisition card and LabVIEW custom software. The acquisition of welding electrical parameters such as welding current and welding voltage is achieved using an integrated system of Hall current and Hall voltage sensors. The current sensor is fitted onto the ground cable to measure the welding current, and the welding torch nozzle and the grounding wire of the workbench are used as measurement points connected to the input terminal of the voltage sensor to measure the arc voltage. A multi-channel high-speed data acquisition card (acquisition frequency 0~250kHz) and data adapter terminals are used to achieve real-time synchronous acquisition of welding current, voltage, and other values. The current and voltage waveforms and the switch signal are filtered using Chebyshev filters, Butterworth filters, and Savitzky-Golay filters, respectively. The filtered electrical signal data removes the AC component from the pulsating voltage, resulting in a smooth signal and avoiding harmonic interference in the circuit. The high-speed signal is displayed and saved as current and voltage waveforms in specific software via a computer and some software, automatically generating an Excel file containing all electrical parameter waveform information for subsequent offline data processing and analysis. Logical calculation software was developed based on LabVIEW. For the manual welding process of pipe butt joint circumferential welds, the position of the welding rod and welding torch is accurately determined, forming spatial coordinates. Based on this, the system's synchronous triggering algorithm is optimized to achieve a one-to-one matching of welding electrical parameters and welding torch spatial position using an absolute time reference, obtaining real-time and accurate welding heat input calculation values ​​for manual welding of pipe butt joint circumferential welds.

[0097] This application embodiment, based on a combination of precision optical sensing and logical computation, utilizes an object capture mechanism to obtain real-time welding speed for manual welding of pipe butt joint circumferential welds. It employs an integrated system of Hall current and Hall voltage sensors to achieve real-time acquisition of welding electrical parameters such as welding current and voltage. Based on a synchronous triggering algorithm using logical computation and optimization systems, it achieves a one-to-one matching between welding electrical parameters and the spatial position of the welding torch using an absolute time reference, obtaining real-time and accurate calculated values ​​of welding heat input. This application embodiment enables real-time acquisition of welding heat input for manual welding of pipe butt joint circumferential welds.

[0098] It is understood that the above embodiments are merely exemplary implementations used to illustrate the principles of this application, and this application is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of this application, and these modifications and improvements are also considered to be within the scope of protection of this application.

Claims

1. A method for real-time acquisition of welding heat input, characterized in that, include: The displacement of the welding torch or welding wire is sensed by a diffuse reflection photoelectric switch array installed on the surface of the pipe, and a corresponding pulse trigger signal is generated. The current and voltage data of the welding circuit are sensed by the Hall effect sensor system, and corresponding analog electrical signals are generated. Using a multi-channel data acquisition device with a unified hardware clock as the time reference, the pulse trigger signal and the analog electrical signal are acquired synchronously, and the analog electrical signal is converted into a digital signal with a timestamp. Based on the time sequence of the pulse trigger signal and the preset spatial geometric relationship of multiple diffuse reflection photoelectric switches, the real-time welding speed data is calculated. The current data and the voltage data are subjected to multi-stage filtering to eliminate noise interference; A unified timestamp is determined, and the filtered current data, voltage data, and welding speed data are time-matched based on the unified timestamp; The welding heat input value is calculated based on the matched current data, voltage data, and welding speed data.

2. The method according to claim 1, characterized in that, The diffuse reflection photoelectric switch array is mounted on the surface of the pipe via a chain-type fixing base, and the photosensitive part of each diffuse reflection photoelectric switch is facing the movement path of the welding gun or the welding wire.

3. The method according to claim 2, characterized in that, Prior to sensing the displacement of the welding torch or welding wire via a diffuse reflection photoelectric switch array mounted on the pipe surface, the following steps are included: The position of each diffuse reflection photoelectric switch is adjusted by multiple three-dimensional fine-tuning gimbals to make the laser spot coincide with the center line of the weld. The distance between the front end of each diffuse reflection photoelectric switch and the surface of the pipe is measured using a gauge block, and the Z-axis of the multiple three-dimensional fine-tuning gimbals is adjusted so that the distance reaches the preset working range of the sensor. The angle between the central axis of each diffuse reflection photoelectric switch and the normal to the surface of the pipe is measured using a digital angle ruler, and the Y-axis and X-axis of the multiple three-dimensional fine-tuning gimbals are finely adjusted to ensure that the angle is 90°. Locking device for locking the three-dimensional fine-tuning gimbal.

4. The method according to claim 1, characterized in that, The step of synchronously acquiring the pulse trigger signal and the analog electrical signal using a multi-channel data acquisition device with a unified hardware clock as the time reference includes: Using a multi-channel high-speed data acquisition card and data transfer terminals, the pulse trigger signal and the analog electrical signal are synchronously acquired with the hardware clock of the multi-channel high-speed data acquisition card as the absolute time reference. The sampling frequency ranges from 0 to 250 kHz.

5. The method according to claim 1, characterized in that, The calculated real-time welding speed data includes: The angle difference between the first and last diffuse reflection photoelectric switches is obtained by using an inclination meter, and the arc length is calculated based on the pipe diameter. Based on the time sequence of the pulse trigger signal, determine the time difference between the triggering of adjacent photoelectric switches by the welding torch or the welding wire; Based on the arc length and the time difference, calculate the real-time welding speed data.

6. The method according to claim 1, characterized in that, The multi-level filtering process for the current data and the voltage data includes: The current data and the voltage data are processed sequentially by a Butterworth low-pass filter, a Chebyshev band-stop filter, and a Savitsky-Gore smoothing filter, respectively. The Butterworth low-pass filter is used for low-pass filtering, and its cutoff frequency is adaptively adjusted based on background noise spectrum analysis; the Chebyshev band-stop filter is used for band-stop filtering, and its stopband center frequency automatically tracks the power frequency and its harmonics; the Savitsky-Gore smoothing filter is used for smoothing filtering.

7. The method according to claim 1, characterized in that, The determination of the unified timestamp includes: The hardware clock of the data acquisition board is used as the absolute time reference for the entire system. Configure a hardware timing engine for analog input sampling, digital input counting, and digital output triggering; The current acquisition task starts at a fixed sampling rate, and the data of each sampling batch includes a sampling timestamp generated by hardware. The value of the encoder counter is read at each sampling interval and bound to the absolute timestamp of the time it is read; When the triggering conditions are met, the hardware timing engine performs a digital output operation to accurately record the pulse generation time.

8. The method according to claim 1, characterized in that, The method further includes: Armored shielded cables are used to transmit sensor signals, and an RC low-pass filter circuit is added at the hardware level to achieve anti-interference processing.

9. The method according to claim 1, characterized in that, After calculating the welding heat input value, the following is included: Output and store the collected data; The collected data includes the timestamp, current data, voltage data, welding speed data, and heat input value, and is stored in a structured CSV file format.

10. A real-time acquisition system for welding heat input, characterized in that, include: The displacement sensing and signal triggering module is used to sense the displacement of the welding torch or welding wire through a diffuse reflection photoelectric switch array installed on the surface of the pipe, and generate a corresponding pulse trigger signal. The analog electrical signal generation module is used to sense the current and voltage data of the welding circuit through the Hall sensor system and generate corresponding analog electrical signals. The digital signal acquisition module is used to synchronously acquire the pulse trigger signal and the analog electrical signal through a multi-channel data acquisition device, using a unified hardware clock as the time reference, and convert the analog electrical signal into a digital signal with a timestamp. The welding speed determination module is used to calculate real-time welding speed data based on the time sequence of the pulse trigger signal and the preset spatial geometric relationship of multiple diffuse reflection photoelectric switches. A multi-stage filtering module is used to perform multi-stage filtering on the current data and the voltage data to eliminate noise interference; The timing matching module is used to determine a unified timestamp and perform timing matching between the filtered current data, the voltage data and the welding speed data based on the unified timestamp. The input value determination module is used to calculate the welding heat input value based on the matched current data, voltage data, and welding speed data.

Images