A method and system for fault interlock protection of a carbon dioxide laser based on standing wave ratio monitoring

Abstract

A kind of carbon dioxide laser fault interlock protection method and system based on standing wave ratio monitoring, it is related to the field of device using stimulated emission, this method includes: collecting output pulse width value and reflected power cumulative value, respectively calculating pulse width variation gradient value and reflected power variation gradient value;Read stored gradient collaborative characterization data, and when the absolute value of pulse width variation gradient value is greater than or equal to the preset minimum gradient threshold, correct gradient collaborative characterization data;Real-time decision threshold is generated based on gradient collaborative characterization data;Calculate the standing wave ratio monitoring value of current detection period;Combining standing wave ratio monitoring value and real-time decision threshold, determine standing wave ratio state determination result and output corresponding interlock protection level signal. Implement the present application, can optimize laser protection mechanism, improve laser processing efficiency.

Description

Technical Field

[0001] This application relates to the field of devices utilizing stimulated emission, and more particularly to a method and system for interlocking protection of carbon dioxide laser faults based on standing wave ratio monitoring. Background Technology

[0002] Carbon dioxide lasers are widely used in industrial processing. However, if the laser energy encountered impedance mismatch or optical component contamination during transmission, energy reflection would occur, leading to an increased standing wave ratio (SWR). In five-axis linkage three-dimensional spatial machining, as the robotic arm drives the laser head to move along complex spatial curved surfaces, stable laser energy output and real-time monitoring of the SWR are fundamental to ensuring processing quality and preventing reflected energy from burning out optical components.

[0003] In related technologies, to protect the laser during processing, the monitoring system establishes a synchronous detection window that matches the output pulse width and performs gated cumulative sampling of the reflected power detection signal within this synchronous detection window. After obtaining the cumulative result, the cumulative reflected power value is compared with a certain ratio of the output pulse width value. When the cumulative reflected power value is greater than the ratio threshold, a standing wave ratio fault signal is generated, and the laser output is cut off according to the number of shielding cycles mapped to the duty cycle interval of the current pulse.

[0004] However, when processing three-dimensional curved surfaces, it is often necessary to adjust the laser head attitude and output parameters at high frequency. At this time, the dynamic deflection of the laser incident angle will be accompanied by a surge in normal transient geometric reflection. Related technologies perform laser protection based on a fixed proportional threshold, which will frequently trigger false alarms. It is necessary for staff to change the proportional threshold in time, but stopping the machine to change the settings will affect work efficiency. Summary of the Invention

[0005] This application provides a method and system for interlocking protection of carbon dioxide laser faults based on standing wave ratio monitoring, which is used to optimize the laser protection mechanism and improve laser processing efficiency.

[0006] In a first aspect, this application provides a fault interlocking protection method for a carbon dioxide laser based on standing wave ratio (SWR) monitoring, applied to a laser protection system. The method includes: acquiring the output pulse width value and the accumulated reflected power value of the current detection cycle; calculating the difference between the output pulse width value, the accumulated reflected power value, and the corresponding values ​​from the previous detection cycle, respectively, to generate pulse width change gradient values ​​and reflected power change gradient values; reading stored gradient co-characterization data, and when the absolute value of the pulse width change gradient value is greater than or equal to a preset minimum gradient threshold, calculating the ratio of the reflected power change gradient value to the pulse width change gradient value to correct the gradient co-characterization data; performing a compensation operation on a preset initial SWR judgment threshold based on the gradient co-characterization data to generate a real-time judgment threshold; calculating the ratio of the accumulated reflected power value to the output pulse width value to generate the SWR monitoring value for the current detection cycle; and combining the SWR monitoring value and the real-time judgment threshold to determine the SWR state judgment result and output the corresponding interlocking protection level signal.

[0007] In the above embodiments, the laser protection system corrects the gradient collaborative characterization data based on the gradient ratio and compensates the initial standing wave ratio judgment threshold accordingly, so that the real-time judgment threshold follows the changes in operating conditions, and can distinguish between normal reflection fluctuations and abnormal reflection increases, thereby reducing the probability of false triggering while maintaining fault detection capability.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, after acquiring the output pulse width value and the cumulative reflected power value of the current detection period, the method further includes: acquiring the time-domain envelope signal of the reflected power in the current detection period; calculating the time-domain energy centroid of the time-domain envelope signal and determining the hysteresis offset of the time-domain energy centroid relative to the reference output pulse centroid; and correcting the cumulative reflected power value based on the hysteresis offset.

[0009] In the above embodiments, the laser protection system performs time-domain correction on the cumulative value of reflected power by calculating the hysteresis offset of the energy centroid of the time-domain envelope of the reflected power relative to the centroid of the reference output pulse. This quantifies and compensates for the time-domain offset caused by optical path transmission delay or multiple reflections, ensuring the time-domain alignment accuracy of the VSWR calculation.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, before performing compensation calculations on a preset initial standing wave ratio (VSWR) determination threshold based on gradient collaborative characterization data to generate a real-time determination threshold, the method further includes: acquiring a pre-reading interpolation command sent by the CNC system, and extracting the curvature change rate of the target machining trajectory from the pre-reading interpolation command; when the curvature change rate is greater than a preset attitude change threshold, generating a feedforward compensation coefficient positively correlated with the curvature change rate; and correcting the gradient collaborative characterization data based on the feedforward compensation coefficient.

[0011] In the above embodiments, the laser protection system obtains the rate of curvature change by pre-reading interpolation instructions and generates feedforward compensation coefficients to pre-correct gradient collaborative characterization data, so that the threshold compensation calculation incorporates the expected transient reflection increment before the attitude deflection occurs, avoiding misjudgment of the normal geometric reflection surge during the rapid reversal of the robotic arm.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, the step of generating a feedforward compensation coefficient positively correlated with the rate of curvature change when the rate of curvature change is greater than a preset attitude change threshold specifically includes: extracting the machining attitude deflection angular velocity corresponding to the rate of curvature change in the pre-read interpolation command when the rate of curvature change is greater than the preset attitude change threshold; inputting the machining attitude deflection angular velocity into a preset transient reflection mapping model to calculate the expected reflection power increment; calculating the ratio of the expected reflection power increment to the cumulative reflection power value to generate a target weight factor; and determining the feedforward compensation coefficient based on the preset base compensation base and the target weight factor.

[0013] In the above embodiments, the laser protection system maps the deflection angular velocity to the expected reflection power increment through a transient reflection mapping model, and determines the feedforward compensation coefficient by the ratio of the increment to the cumulative reflection power value, so that the compensation amplitude is proportionally related to the actual reflection energy level, and matching compensation can be obtained in both high power and low power ranges.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, the step of performing compensation calculation on a preset initial VSWR determination threshold based on gradient collaborative characterization data to generate a real-time determination threshold specifically includes: constructing a data sequence containing gradient collaborative characterization data from the current detection period and N consecutive historical detection periods; where N is an integer greater than 1; calculating the average rate of change of the data increasing interval in the data sequence as the rising edge average rate of change, and the average rate of change of the data decreasing interval as the falling edge average rate of change; calculating the ratio of the rising edge average rate of change to the falling edge average rate of change to determine the asymmetric distortion coefficient; and combining the asymmetric distortion coefficient and the gradient collaborative characterization data to perform compensation on the preset initial VSWR determination threshold to generate a real-time determination threshold.

[0015] In the above embodiments, the laser protection system obtains the asymmetric distortion coefficient by calculating the ratio of the average rate of change of the rising edge to the falling edge of the multi-cycle gradient collaborative characterization data sequence, thereby quantifying the degree of time-domain waveform skew, so that the threshold compensation operation can identify the progressive fault trend and adjust the compensation amplitude accordingly.

[0016] In some embodiments of the first aspect, the step of compensating for a preset initial VSWR determination threshold by combining asymmetric distortion coefficients and gradient collaborative characterization data to generate a real-time determination threshold specifically includes: when the asymmetric distortion coefficients are within a preset elastic recovery range, determining a target tolerance value based on gradient collaborative characterization data and a preset gain coefficient; when the asymmetric distortion coefficients are not within the elastic recovery range, calculating an attenuation penalty factor based on the asymmetric distortion coefficients and determining an attenuation compensation value based on gradient collaborative characterization data; and correcting the initial VSWR determination threshold based on the target tolerance value or the attenuation compensation value to determine the real-time determination threshold.

[0017] In the above embodiments, the laser protection system selects different compensation paths according to the range of the asymmetric distortion coefficient. Within the elastic recovery range, a target tolerance value is generated using the gain coefficient, and outside the range, an attenuation penalty factor is introduced to generate an attenuation compensation value. This ensures that the real-time judgment threshold maintains a margin when the operating condition is recoverable and tightens the sensitivity when the distortion exceeds the limit.

[0018] In some embodiments of the first aspect, after determining the standing wave ratio (SWR) state determination result and outputting the corresponding interlocking protection level signal by combining the SWR monitoring value and the real-time determination threshold, the method further includes: when the SWR state determination result is in a normal state and the duration for which the SWR monitoring value is greater than the initial SWR determination threshold exceeds a preset duration, calculating the difference between the SWR monitoring value and the initial SWR determination threshold as the energy overflow amount; performing a time integral operation combining the heat dissipation attenuation factor on the energy overflow amount to generate a thermal lensing effect cumulative characteristic value; and overwriting the SWR state determination result as a warning state when the thermal lensing effect cumulative characteristic value is greater than a preset thermal damage threshold.

[0019] In the above embodiments, the laser protection system accumulates the characteristic value of thermal lensing effect by performing a time integral operation with a heat dissipation attenuation factor on the subthreshold energy overflow. When the characteristic value exceeds the thermal damage threshold, it is overwritten as a warning state, so that the protection mechanism can capture the risk of thermal distortion of optical components caused by long-term subthreshold reflection accumulation.

[0020] In a second aspect, embodiments of this application provide a laser protection system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors and is used to store computer program code, which includes computer instructions, and the one or more processors call the computer instructions to cause the laser protection system to perform the method described in the first aspect and any possible implementation thereof.

[0021] Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a laser protection system, cause the laser protection system to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a laser protection system, cause the laser protection system to perform the method described in the first aspect and any possible implementation thereof.

[0023] Understandably, the laser protection system provided in the second aspect, the computer storage medium provided in the third aspect, and the computer program product provided in the fourth aspect are all used to execute the methods provided in the embodiments of this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0025] 1. By employing a technique that corrects the gradient collaborative characterization data using the ratio of the gradient values ​​of pulse width changes and reflection power changes in adjacent detection cycles, and then performing compensation calculations on the initial VSWR determination threshold to generate a real-time determination threshold, the VSWR determination threshold can be synchronously adapted to the dynamic adjustment of the laser output parameters. This effectively solves the technical defect of frequent false triggering of interlock protection under high-frequency laser head attitude adjustment conditions based on a fixed ratio threshold in the existing technology. It achieves a reduction in the false trigger rate of interlock protection and a reduction in processing efficiency loss caused by downtime for threshold adjustment while maintaining the optical path fault detection capability.

[0026] 2. By employing the technique of extracting the rate of curvature change from the pre-read interpolation command of the CNC system and generating feedforward compensation coefficients to pre-correct the gradient collaborative characterization data, the laser protection system can incorporate the expected reflection increment into the threshold compensation calculation before the actual deviation of the robotic arm's posture occurs. This effectively solves the technical defect in the existing technology where threshold compensation is performed only based on the already occurred reflection power data, resulting in a delayed response and false alarms in the trajectory curvature change section. It realizes the feedforward adjustment of the interlock protection threshold and the suppression of false alarms under the rapid reversal condition of the robotic arm.

[0027] 3. By employing a technique that combines time integration with heat dissipation attenuation factor to generate a cumulative characteristic value of thermal lensing effect on the subthreshold energy overflow, and overwriting the judgment result as an early warning state when the characteristic value exceeds the thermal damage threshold, the laser protection system can continuously track and quantify the thermal effect of long-term subthreshold reflection energy accumulation. This effectively solves the technical defect in the existing technology that cannot identify the cumulative thermal distortion damage of optical components based on a single comparison of instantaneous standing wave ratio and fixed threshold, and realizes early warning protection against progressive damage of optical components caused by thermal lensing effect. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating a method for interlocking protection of carbon dioxide laser faults based on standing wave ratio monitoring in an embodiment of this application.

[0029] Figure 2 This is another flowchart illustrating the carbon dioxide laser fault interlocking protection method based on VSWR monitoring in the embodiments of this application;

[0030] Figure 3 This is a schematic diagram of the physical device structure of a laser protection system in the embodiments of this application. Detailed Implementation

[0031] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0032] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0033] The laser protection system involved in this application embodiment is deployed between the radio frequency excitation source and the optical transmission link of the carbon dioxide laser, and is communicatively connected to the motion control module of the CNC system.

[0034] In a five-axis linkage three-dimensional spatial machining scenario, a carbon dioxide laser generates an infrared laser with a wavelength of 10.6 micrometers through radio frequency pulse excitation. This laser is transmitted via the articulated arm optical path or flexible optical fiber to a laser processing head mounted at the end of the robotic arm, performing processes such as cutting, welding, or surface treatment on metal or non-metal workpieces. Driven by the five-axis interpolation of the CNC system, the laser processing head moves along the three-dimensional curved surface of the workpiece, its attitude angle continuously changing along the surface normal to maintain a set incident angle between the laser beam and the workpiece surface. During this movement, the laser beam undergoes Fresnel reflection through the interfaces of the workpiece surface, the focusing lens, and the various mirrors in the transmission optical path. The reflected energy propagates backward along the optical path towards the laser resonant cavity. The standing wave ratio (VSWR) characterizes the ratio of forward transmission power to backward reflection power; an increase in the VSWR indicates an abnormal state in the optical path system, such as impedance mismatch, component surface contamination, or beam pointing deviation.

[0035] The laser protection system acquires the reflected power signal through a directional coupler or power detector coupled in the optical path during the detection cycle of each laser output pulse, and performs a standing wave ratio (VSWR) assessment in conjunction with the pulse width information output by the RF excitation module. This detection cycle is synchronized with the laser pulse repetition frequency, which typically ranges from 1 kHz to 100 kHz in industrial applications. The output pulse width value corresponds to the effective duration of a single laser pulse, and the cumulative reflected power value is the energy metric obtained by time integration of the reflected power envelope signal within the pulse duration window. Gradient co-characterization data describes the degree of co-correlation between changes in output pulse width and changes in reflected power, reflecting whether the reflected power response maintains a linear following relationship with pulse width changes during dynamic adjustment of laser output parameters. When this co-correlation deviates from a linear following relationship, it indicates that changes in reflected power are contributed by factors other than pulse width adjustment, suggesting a possible optical path anomaly. The interlock protection level signal is a hardware level output by the laser protection system to the RF excitation source enable terminal or an external safety relay. A state reversal of this level signal directly cuts off the RF excitation or closes the optical shutter to prevent further laser energy output.

[0036] In 3D surface machining, when a robotic arm performs high-speed attitude deflection in a trajectory segment with a small radius of curvature, the transient shift in the laser beam incident angle leads to an increase in the geometric reflection energy of the workpiece surface. This reflection increment is a transient response under normal machining conditions rather than an optical path fault signal. The method provided in this application introduces a dynamic correction mechanism for gradient collaborative characterization data and a compensation calculation mechanism for real-time judgment thresholds. This enables the laser protection system to distinguish between the aforementioned normal transient reflection fluctuations and the abnormal increase in the true standing wave ratio caused by optical component contamination, optical path misalignment, or transmission medium damage. This suppresses the occurrence of false triggering of interlocking protection without weakening the fault detection sensitivity.

[0037] The method provided in this embodiment is described in detail below. Please refer to [link / reference]. Figure 1 This is a flowchart illustrating a method for interlocking protection of carbon dioxide laser faults based on standing wave ratio monitoring in an embodiment of this application.

[0038] S101. Collect the output pulse width value and cumulative reflection power value of the current detection cycle.

[0039] The output pulse width value refers to the numerical value of the effective duration of the current laser pulse extracted by the laser protection system from the pulse width modulation signal of the RF excitation module, measured in microseconds (μs). This value corresponds to the high-level duration of the RF excitation pulse. The cumulative reflected power value refers to the cumulative energy metric obtained by the laser protection system through discrete-time integration of the analog reflected power signal output from the directional coupler after analog-to-digital conversion within the current pulse duration window, measured in microjoules (μJ). The current detection period refers to the signal acquisition and processing time interval synchronized with a single laser output pulse, its start and end times defined by the rising and falling edges of the RF excitation pulse.

[0040] Specifically, when the rising edge of the RF excitation of each laser output pulse arrives, the synchronous trigger module of the laser protection system generates a start-up signal, initiating the analog-to-digital converter (ADC) to continuously sample the power detection signal output from the directional coupler's reflection port. When the falling edge of the RF excitation arrives, the synchronous trigger module generates a stop-up signal, and the ADC stops sampling. The laser protection system counts the time interval between the start and stop-up signals to obtain the output pulse width. Simultaneously, the laser protection system performs an accumulation operation on the reflection power values ​​of all sampled points within this time interval and multiplies it by the sampling interval to obtain the cumulative reflection power value. Under operating conditions of a pulse repetition frequency of 20kHz and a sampling rate of 10MHz, the maximum pulse width of a single detection cycle is 50μs, corresponding to 500 sampling points participating in the accumulation operation.

[0041] In some embodiments, the acquisition of the output pulse width value can be achieved by capturing the edge of the RF excitation signal using a high-speed counter module inside the FPGA, or by using an external time-to-digital converter chip to obtain higher time resolution, or by other methods, which are not limited here. The acquisition of the accumulated reflected power value depends on the accurate establishment of the synchronization window. If there is a fixed deviation between the start time of the synchronization window and the start time of the actual pulse, it can be corrected by a pre-calibrated delay compensation amount.

[0042] S102. Calculate the difference between the output pulse width value, the cumulative reflection power value and the corresponding value in the previous detection cycle, respectively, and generate the pulse width change gradient value and the reflection power change gradient value.

[0043] The pulse width gradient value refers to the rate of change of the output pulse width between two adjacent detection cycles. The reflected power gradient value refers to the rate of change of the accumulated reflected power between two adjacent detection cycles. The corresponding values ​​in the previous detection cycle refer to the output pulse width value and the accumulated reflected power value collected and stored in the register by the laser protection system during step S101 of the previous detection cycle.

[0044] Specifically, after completing the acquisition in step S101, the laser protection system reads the output pulse width value stored in the previous detection cycle from the data register. and cumulative reflected power Where k represents the sequence number of the current detection cycle. The laser protection system acquires the output pulse width value for the current detection cycle. minus Obtain the pulse width variation gradient value ΔPW= - Meanwhile, the laser protection system collects the cumulative value of reflected power during the current detection cycle. minus Obtain the gradient value of the reflected power change ΔPR= - During the accelerated cutting phase of the CNC system, the output pulse width increases periodically, and ΔPW is a positive value; during the deceleration phase, ΔPW is a negative value; and during the constant speed phase, ΔPW approaches zero. The sign and amplitude of ΔPR are simultaneously affected by both the pulse width change and the optical path state change.

[0045] In some embodiments, there may be a situation where historical data from the previous cycle is unavailable for difference calculation during the first detection cycle after the laser protection system is powered on and initialized. To address this, the laser protection system pre-fills the output pulse width and cumulative reflected power values ​​from the previous cycle in the data register with zero values ​​during the power-on initialization phase. This ensures that the pulse width gradient value in the first detection cycle equals the currently acquired output pulse width value, and the reflected power gradient value equals the currently acquired cumulative reflected power value. In subsequent steps, a minimum gradient threshold is used to filter out any abnormal gradients that may occur in this initial cycle and to correct the resulting data.

[0046] S103. Read the stored gradient co-representation data, and when the absolute value of the pulse width change gradient value is greater than or equal to the preset minimum gradient threshold, calculate the ratio of the reflection power change gradient value to the pulse width change gradient value to correct the gradient co-representation data.

[0047] The gradient coordination characteristic data refers to the numerical parameters used to quantify the degree of linear following relationship between changes in reflected power and changes in output pulse width. These parameters are stored in the data register of the laser protection system, and their initial value is set by a pre-calibrated reference coordination ratio during system startup. The reference coordination ratio is determined as follows: During the factory calibration phase of the laser protection system, the laser is set to standard output power and the optical path is in standard alignment. The CNC system drives the laser processing head to move along a standard straight trajectory at a preset calibration speed, while simultaneously adjusting the output pulse width stepwise using a preset step function. The laser protection system collects the corresponding gradient values ​​of reflected power change and pulse width change at each step change in pulse width. The arithmetic mean of the ratios of these two values ​​at all effective step points is obtained as the reference coordination ratio. This reference coordination ratio characterizes the standard linear following coefficient of reflected power to pulse width change under conditions of no optical path abnormalities and stable processing posture. Its typical value range varies depending on the optical path structure and laser power level; in a 1kW-class carbon dioxide laser, a typical value is 0.02 to 0.08 μJ / μs.

[0048] In addition, the minimum gradient threshold is a preset threshold value used to determine whether the gradient value of the pulse width change has effective discrimination. When the absolute value of the gradient value of the pulse width change is lower than this threshold value, it indicates that the output pulse width is in the quasi-steady state range. At this time, the denominator of the gradient ratio calculation is too small, which will lead to numerical instability, so no correction is performed.

[0049] Specifically, in step S103, the laser protection system can read the currently stored gradient co-characterization data from the register. Subsequently, the laser protection system compares the absolute value of the pulse width variation gradient ΔPW, |ΔPW|, with a preset minimum gradient threshold. Compare. When |ΔPW|≥ When the condition is met, the laser protection system calculates the ratio of ΔPR / ΔPW and assigns this ratio as the gradient co-characterization data for the current period. =ΔPR / ΔPW. When |ΔPW| < Under these conditions, the laser protection system maintains the gradient cooperative characterization data unchanged, that is... = Taking the machining process where the CNC system command pulse width increases cycle by cycle from 30μs to 35μs as an example, if the minimum gradient threshold is set to 0.5μs, then ΔPW in each cycle is greater than this threshold, and the gradient co-representation data is updated in each cycle. If the pulse width remains stable at 35μs, |ΔPW| is zero, and the gradient co-representation data retains the value from the last update.

[0050] In some embodiments, the correction of gradient co-representation data can be achieved in several ways: optionally, the laser protection system directly overwrites the instantaneous ratio of ΔPR / ΔPW. This method offers the fastest response speed but is sensitive to single-cycle noise, making it suitable for rapid power adjustment sections with large pulse width variations. Optionally, the laser protection system updates the gradient collaborative characterization data using a weighted filtering method. =α*(ΔPR / ΔPW)+(1-α)* Where α is a preset filtering coefficient and 0 < α < 1, it suppresses the interference of single-cycle outliers while maintaining the tracking performance, and is suitable for uniform processing sections with gradually changing pulse widths. It is understood that other filtering strategies can also be used to correct gradient collaborative representation data, which is not limited here.

[0051] In some embodiments, when the pulse width variation gradient value is exactly near the minimum gradient threshold boundary, quantization noise may cause the correction condition to switch repeatedly between adjacent cycles. To address this, the laser protection system introduces a hysteresis band in the minimum gradient threshold determination logic, requiring |ΔPW| ≥ 1 when the correction condition switches from unsatisfied to satisfied. +Δh, when the correction condition changes from satisfied to unsatisfied, requires |ΔPW|< -Δh, where Δh is a preset hysteresis value, typically 5% to 10% of the minimum gradient threshold, and satisfies >Δh to ensure the lower boundary determination condition −Δh is a positive value to suppress the impact of boundary oscillations on the stability of gradient collaborative characterization data.

[0052] S104. Based on the gradient collaborative characterization data, perform compensation calculation on the preset initial standing wave ratio (SWR) determination threshold to generate a real-time determination threshold.

[0053] The initial standing wave ratio (SWR) determination threshold refers to a fixed threshold parameter written into the non-volatile memory during the factory calibration or power-on initialization phase of the laser protection system. This parameter corresponds to the upper limit of the maximum SWR monitoring value allowed by the optical path system under standard conditions.

[0054] Specifically, after completing the gradient collaborative characterization data update in step S103, the laser protection system reads the initial standing wave ratio (VSWR) determination threshold stored in the non-volatile memory. and gradient co-representation data for the current period Laser protection system based on Calculate the compensation offset ΔTH and add it to the initial threshold to generate the real-time judgment threshold. The calculation method for the compensation offset ΔTH depends on the direction and magnitude of the deviation of the gradient co-representation data relative to the baseline co-representation ratio: when When the ratio is greater than the baseline coordination ratio, it indicates that the rate of increase in reflected power is higher than the normal following level corresponding to the pulse width increase rate. Under the premise that the change in reflected power and the change in pulse width maintain a coordinated following relationship, the compensation offset is positive to appropriately relax the threshold; when... When the ratio is less than the baseline coordination ratio, it indicates that the increase in reflected power lags behind the pulse width change, and the compensation offset is negative to tighten the threshold. The magnitude of the compensation offset is constrained by a preset maximum compensation limit to prevent the threshold from deviating from the safe range in real time under extreme operating conditions.

[0055] In some embodiments, the compensation calculation can be implemented in several ways: optionally, the laser protection system performs the calculation using a linear compensation function, i.e., ΔTH=K*( - ), where K is the preset compensation gain coefficient, Using the benchmark coordination ratio as the benchmark, this method requires minimal computation and has low hardware implementation complexity; optionally, the laser protection system performs the calculation using a piecewise linear function. - Different gain coefficients are set for different amplitude ranges. A larger gain is used in the small deviation range to improve sensitivity, while a smaller gain is used in the large deviation range to limit the compensation amplitude. This method can achieve differentiated compensation characteristics in different operating conditions. It is understood that other function forms can also be used to implement the compensation calculation; this is not limited here.

[0056] S105. Calculate the ratio of the cumulative reflected power value to the output pulse width value to generate the VSWR monitoring value for the current detection cycle.

[0057] Among them, the VSWR monitoring value refers to the ratio of the cumulative reflected power value to the output pulse width value within the current detection cycle. This ratio characterizes the average reflected power level per unit pulse time and is used to measure the impedance matching status of the optical path system. The larger the ratio, the higher the proportion of reflected energy in the optical path and the more severe the impedance mismatch of the optical path.

[0058] Specifically, after generating the real-time judgment threshold in step S104, the laser protection system reads the accumulated reflection power value collected in step S101. and output pulse width value Perform division operation This operation obtains the VSWR monitoring value for the current detection cycle. The physical meaning of this division operation is to normalize the total accumulated reflected energy within the pulse duration window to a unit time, eliminating the dimensional differences in the accumulated value caused by different pulse widths in different detection cycles. For example, under the conditions of a pulse width of 40 μs and a cumulative reflected power of 120 μJ, the VSWR monitoring value is 3.0 W.

[0059] In some embodiments, the division operation of the VSWR monitoring value can be implemented by a fixed-point divider inside the FPGA, or by a floating-point division instruction of an embedded DSP module, or by other methods such as lookup table approximation or the CORDIC algorithm; no limitation is made here. In some embodiments, the output pulse width value may approach zero due to the CNC system issuing extremely short pulse instructions, leading to overflow during the division operation. To address this, the laser protection system will determine the appropriate value before performing the division operation. Is it less than the preset minimum effective pulse width value? ,like < If the current period's VSWR monitoring value is set to zero and subsequent comparison and judgment steps are skipped, the interlock protection level signal that maintains the current state will be directly output to avoid misjudgment caused by division by zero overflow.

[0060] S106. Combining the VSWR monitoring value and the real-time judgment threshold, determine the VSWR status judgment result and output the corresponding interlocking protection level signal.

[0061] Among them, the interlock protection level signal refers to the digital level driven by the hardware output port of the laser protection system. In the normal state, it corresponds to a high-level enable signal to maintain the normal operation of the RF excitation, and in the fault state, it corresponds to a low-level disable signal to cut off the RF excitation output.

[0062] Specifically, after completing the VSWR monitoring value calculation in step S105, the laser protection system will... Compared with the real-time judgment threshold generated in step S104 Perform a numerical comparison. If < The laser protection system sets the VSWR state determination result to normal and maintains a high-level enable signal at the hardware output port, allowing the RF excitation source to continue operating. If ≥ The laser protection system sets the VSWR (Standing Wave Ratio) status judgment result to a fault state and flips the level of the hardware output port to a low-level inhibit signal. Upon receiving the low level, the RF excitation source stops outputting RF power, and the laser pulse output terminates. The response delay of the level flipping action is determined by the propagation delay of the hardware output drive circuit, typically on the order of nanoseconds. The flipping of the interlock protection level signal simultaneously triggers an emergency stop interrupt in the CNC system. Upon receiving this interrupt, the CNC system executes machining pause and robotic arm braking operations.

[0063] In some embodiments, the interlocking protection level signal can be directly transmitted to the enable terminal of the RF excitation source through the optocoupler isolation circuit driven by the FPGA's general-purpose input / output port, or it can be physically disconnected by controlling an external safety relay via the SPI bus. Other signal output methods can also be used, which are not limited here. In some embodiments, electromagnetic interference may cause spike pulses in the VSWR monitoring value within a single detection cycle, which may lead to... Instantly passed In this situation, the laser protection system will set a continuous judgment counter. ≥ The counter is incremented when the condition is first met, and only incremented when the counter value reaches a preset threshold for the number of consecutive triggers. Level toggling and state setting are only performed when the counter has not reached the specified time. a previous cycle Falling back to Then, the counter is reset to zero to filter out interference spikes in a single cycle or a few cycles.

[0064] The above steps S101 to S106 achieve dynamic compensation of the VSWR determination threshold based on data acquisition within a fixed detection cycle and a single-cycle gradient difference correction mechanism. In actual five-axis linkage three-dimensional space machining, the robot arm's attitude deflection angular velocity changes drastically when the CNC system executes high-curvature trajectory sections. The transient surge in reflected power may trigger a misjudgment before it is captured in the gradient difference calculation. Based on this, the method provided in this embodiment will be described in more detail below. Please refer to... Figure 2 This is another flowchart illustrating the carbon dioxide laser fault interlocking protection method based on VSWR monitoring in this application embodiment.

[0065] S201. Collect the output pulse width value and cumulative reflection power value of the current detection cycle.

[0066] Refer to step S101, which will not be repeated here.

[0067] In some embodiments, the laser protection system performs a time-domain position correction on the accumulated reflected power value after acquisition to improve its alignment accuracy with the current pulse output window. Specifically, the laser protection system acquires the time-domain envelope signal of the reflected power within the current detection period; calculates the time-domain energy centroid of the time-domain envelope signal and determines the hysteresis offset of the time-domain energy centroid relative to the reference output pulse centroid; and corrects the accumulated reflected power value based on the hysteresis offset.

[0068] The time-domain envelope signal refers to the discrete-time sequence formed by the analog-to-digital converter sampling the reflected power detection signal point by point within the acquisition window of the current detection period. The horizontal axis of this sequence represents the sampling time, and the vertical axis represents the instantaneous value of the reflected power at the corresponding time. Its overall outline constitutes the time-domain envelope of the reflected power within the pulse duration. The time-domain energy centroid refers to the weighted average time calculated using the power values ​​of each sampling point in the time-domain envelope signal as weights and the time values ​​of each sampling point as positions. Its physical meaning is the equivalent concentration location of the reflected energy in the time domain. The reference output pulse centroid refers to the midpoint of the output pulse width, that is, the start time of the acquisition window plus half of the output pulse width value.

[0069] Specifically, during the acquisition process in step S201, the laser protection system will collect all sampling point data from the analog-to-digital converter within the acquisition window. , ,..., Store the data in a buffer, where M is the total number of sampling points. The laser protection system calculates the time-domain energy centroid. Where i is the sampling point index and Δt is the sampling interval. Reference output pulse centroid = / 2. Hysteresis offset When τ is positive, it indicates that the main distribution of reflected energy lags behind the midpoint of the pulse output time. This phenomenon is usually caused by the superposition of multiple reflections in the optical path or long optical path propagation delay. The laser protection system corrects the accumulated reflected power value based on the hysteresis offset τ: The physical meaning of this correction is to normalize the reflected energy to a time window equivalent to the output pulse's duration, eliminating the artificial increase in cumulative value caused by time-domain spread or delay of the reflected energy. The corrected... Alternative Used for calculations in subsequent steps S202 and S208.

[0070] For this corrected formula, under ideal conditions where the time-domain envelope of the reflected power is perfectly aligned with the time-domain of the output pulse, the cumulative value of the reflected power is... Equal to average reflected power and pulse width The product of; when the reflected energy extends backward in the time domain due to optical path transmission delay or multiple reflections, the actual reflected energy captured by the gating window is distributed in Over the equivalent time span of +τ, where τ is the hysteresis offset, the data acquired within the gated window at this time... It includes some delayed reflection components that should belong to outside the window, therefore it needs to be calculated according to the effective pulse time. With actual energy distribution time The proportion of +τ Scaling correction is performed to obtain the equivalent cumulative reflected power value corresponding only to the current pulse output window. Taking a specific value as an example, if the output pulse width is 40μs and the hysteresis offset is 2μs, then the equivalent time domain span of the reflected energy is 42μs, and the correction factor is 40 / 42≈0.952. That is, the original cumulative reflected power value needs to be reduced by about 4.8% to eliminate the artificial contribution of the delayed reflection component.

[0071] In some embodiments, the calculation of the time-domain energy centroid can be completed in real time by a multiply-accumulate unit in a pipelined structure of an FPGA, or by batch processing using vector operations by an embedded DSP processor, or other methods may be used, which are not limited here. In some embodiments, the time-domain envelope signal of the reflected power may exhibit a bimodal or multimodal distribution due to the presence of multiple reflection interfaces at different distances in the optical path. In this case, a single energy centroid value cannot accurately characterize the time-domain distribution characteristics of the reflected energy. To address this, the laser protection system performs peak detection on the time-domain envelope signal. When more than one energy peak is detected, the local energy centroid and local cumulative power value of each peak are calculated separately. The local centroids are then weighted by the cumulative power value of each peak to obtain an equivalent composite centroid, which is used to replace the single centroid in the calculation of the hysteresis offset.

[0072] S202. Calculate the difference between the output pulse width value, the cumulative reflection power value and the corresponding value in the previous detection cycle, respectively, and generate the pulse width change gradient value and the reflection power change gradient value.

[0073] Refer to step S102, which will not be repeated here.

[0074] S203. Read the stored gradient co-representation data, and when the absolute value of the pulse width change gradient value is greater than or equal to the preset minimum gradient threshold, calculate the ratio of the reflection power change gradient value to the pulse width change gradient value to correct the gradient co-representation data.

[0075] Refer to step S103, which will not be repeated here.

[0076] In five-axis linkage three-dimensional machining, the CNC system typically performs pre-reading and analysis of several subsequent interpolation trajectory segments when executing the machining program, so that the motion control module can plan acceleration / deceleration curves and attitude transition paths in advance. The laser protection system is connected to the CNC system via a high-speed communication bus, and can acquire the subsequent trajectory information that the CNC system has completed pre-reading and analysis during the execution of the current machining segment. Using the curvature change rate data contained in this pre-reading trajectory information, the laser protection system can perform feedforward correction on the gradient cooperative characterization data before the actual attitude deflection of the robotic arm occurs, so that the subsequent threshold compensation calculation has reserved the corresponding decision margin before the transient surge of reflected power arrives.

[0077] S204. Obtain the pre-read interpolation command sent by the CNC system, and extract the curvature change rate of the target machining trajectory from the pre-read interpolation command.

[0078] The pre-read interpolation instruction refers to the instruction data packet sent by the interpolation calculation module of the CNC system to the laser protection system in data frame format via a high-speed communication bus after kinematic decomposition of the subsequent trajectory segments to be executed in the machining program. This data packet contains the spatial coordinate sequence of the target trajectory segment, feed rate, and kinematic features such as curvature parameters. The rate of change of curvature refers to the first derivative of the curvature value of the target machining trajectory along the arc length direction. This parameter characterizes the spatial rate of change of the curvature degree of the trajectory. The larger the rate of change of curvature, the greater the amplitude of attitude deflection the robotic arm needs to complete within a shorter trajectory segment.

[0079] Specifically, after completing the gradient collaborative characterization data processing in step S203, the laser protection system reads the pre-read interpolation command data frame sent by the CNC system through the high-speed communication bus interface module. The laser protection system performs protocol parsing on the received data frame, extracting the spatial coordinate sequence of the target machining trajectory segment and the corresponding arc length parameter from the frame payload field. Based on the spatial positions of three consecutive coordinate points, the laser protection system calculates the curvature value κ at each trajectory sampling point using the discrete curvature formula. Subsequently, the laser protection system divides the curvature difference between adjacent trajectory sampling points by the arc length increment to obtain the curvature change rate dκ / ds, where s is the arc length parameter. If the pre-read interpolation command directly contains a numerical field of the curvature change rate, the laser protection system directly reads this field value without recalculation. The pre-read depth of the CNC system is typically 2 to 5 trajectory segments after the current execution segment, and the curvature change rate extracted by the laser protection system corresponds to the entry position of the next trajectory segment to be executed.

[0080] In some embodiments, the rate of curvature change can be read directly from the CNC system via the EtherCAT real-time bus, or it can be obtained locally after reading the trajectory coordinate sequence via a shared dual-port RAM interface. Other communication interfaces or computing architectures can also be used, which are not limited here.

[0081] S205. When the rate of curvature change is greater than the preset attitude change threshold, generate a feedforward compensation coefficient that is positively correlated with the rate of curvature change.

[0082] Among them, the attitude change threshold refers to the curvature change rate threshold used to determine whether the target trajectory will cause the robot arm's attitude to deflect rapidly. This threshold is determined during the system calibration stage based on the maximum angular acceleration of the robot arm joints and the optical path geometry parameters.

[0083] Specifically, after completing the curvature change rate extraction in step S204, the laser protection system compares the curvature change rate dκ / ds with a preset attitude change threshold. Perform a numerical comparison. If dκ / ds ≤ This indicates that the curvature change of the trajectory segment to be executed is gradual, the robotic arm's attitude deflection rate is low, and no feedforward compensation is required. The laser protection system sets the feedforward compensation coefficient to 1.0 to avoid altering the gradient cooperative characterization data. If dκ / ds > This indicates that the trajectory segment to be executed has a large curvature abrupt change, and the robotic arm will perform a rapid attitude deflection. The laser protection system uses dκ / ds as input and calculates the feedforward compensation coefficient through a preset mapping function. The mapping function is a monotonically increasing function, with a lower limit of 1.0 for its output value and an upper limit constrained by a preset maximum feedforward limit.

[0084] In some embodiments, the feedforward compensation coefficients can be generated in several ways: optionally, the laser protection system uses a linear mapping function. =1.0+ *(dκ / ds- ) Calculate the feedforward compensation coefficient, where The preset feedforward gain coefficient is used, which has low computational complexity and intuitive parameter calibration. Alternatively, the laser protection system can map different ranges of curvature change rate to corresponding feedforward compensation coefficient values ​​using a lookup table. The entries in the lookup table are established during the factory calibration phase by measuring the reflection power of different curvature trajectories. This method can accurately match the nonlinear reflection characteristics under a specific optical path geometry. It is understood that other mapping methods can also be used to generate the feedforward compensation coefficient, which is not limited here.

[0085] In some embodiments, the update cycle of the CNC system's pre-reading instructions and the detection cycle of the laser protection system may not be strictly synchronized, resulting in the curvature change rate data arriving late or early. To address this, the laser protection system extracts the target execution timestamp from the pre-reading interpolation instruction and compares it with the timestamp of the current detection cycle. If the time difference exceeds one detection cycle, the laser protection system performs time alignment correction on the curvature change rate using linear interpolation before executing subsequent calculations.

[0086] S206, Gradient collaborative characterization data corrected based on feedforward compensation coefficient.

[0087] Specifically, after generating the feedforward compensation coefficients in step S205, the laser protection system will... Gradient co-representation data with the current period Perform multiplication to obtain the corrected gradient co-representation data. .when When the value is 1.0, that is, the rate of change of curvature does not exceed the attitude change threshold, = The correction operation does not change the original data. When When >1.0, > The corrected gradient co-representation data is larger, and in subsequent step S207, this increased value will provide a larger positive compensation offset for the real-time decision threshold, reserving a decision margin for transient reflection increments during attitude deflection. Corrected gradient co-representation data Write to register, replace For use in subsequent steps.

[0088] In some embodiments, the feedforward correction multiplication operation can be implemented using a fixed-point multiplier within the FPGA, a floating-point multiplication instruction of an embedded processor, or other methods such as shift-accumulation approximation; no limitation is made here. In some embodiments, a feedforward compensation coefficient exists. This is due to the presence of high-frequency oscillation components in the curvature change rate data, causing significant fluctuations between adjacent detection cycles. To address this, the laser protection system performs low-pass filtering on the feedforward compensation coefficients for multiple consecutive detection cycles, replacing the original values ​​with the smoothed, filtered values. When participating in the correction calculation, the filter cutoff frequency must be set lower than the servo bandwidth of the robotic arm joint in order to retain effective attitude change information while suppressing non-physical high-frequency disturbances.

[0089] In some embodiments, to determine the feedforward compensation coefficient, the laser protection system extracts the machining attitude deflection angular velocity corresponding to the curvature change rate in the pre-read interpolation command when the curvature change rate is greater than a preset attitude change threshold; inputs the machining attitude deflection angular velocity into a preset transient reflection mapping model to calculate the expected reflection power increment; calculates the ratio of the expected reflection power increment to the cumulative reflection power value to generate a target weight factor; and determines the feedforward compensation coefficient based on the preset basic compensation base and the target weight factor.

[0090] The target weighting factor refers to the ratio of the expected increase in reflected power to the current cumulative value of reflected power, representing the proportion of the expected increase in reflected power relative to the current reflection level. The base compensation factor refers to the benchmark multiplication factor of the feedforward compensation coefficient determined during the system calibration phase, with a typical value of 1.0. In the embodiments of this application, the base compensation factor... The value can be 1.0 to ensure that the rate of curvature change does not exceed the attitude change threshold, thus serving as the feedforward compensation coefficient. = ×(1+0)=1.0, which is the same as when feedforward compensation is not performed in step S205. The setting of =1.0 remains consistent.

[0091] In addition, the construction of the transient reflection mapping model includes: during the system calibration phase, the laser is set to a fixed output power and a fixed pulse width. The CNC system drives the robotic arm to perform attitude deflection motion at a series of discrete deflection angular velocity values. The motion is maintained for several detection cycles at each deflection angular velocity value. The laser protection system collects the increment of the cumulative reflection power value under each deflection angular velocity condition relative to the cumulative reflection power value in the static state as the corresponding measured reflection power increment. The data set consisting of the deflection angular velocity value and the measured reflection power increment is arranged in ascending order of deflection angular velocity and stored as a lookup table. Linear interpolation is used between adjacent entries to obtain the expected reflection power increment output corresponding to any deflection angular velocity input. In scenarios requiring higher precision, quadratic or cubic polynomial least squares fitting can also be performed on the above data set. The polynomial coefficients obtained by fitting are stored in non-volatile memory. During runtime, the expected reflection power increment is directly calculated by substituting the deflection angular velocity into the polynomial.

[0092] Specifically, after confirming that the rate of curvature change is greater than the attitude change threshold, the laser protection system extracts the feed rate of the corresponding trajectory segment from the pre-read interpolation command. Given curvature κ, calculate the machining attitude deflection angular velocity ω= *κ. The laser protection system uses ω as an index to look up the lookup table of the transient reflection mapping model or substitutes the result into the polynomial to obtain the expected reflection power increment. Subsequently, the laser protection system calculates the target weighting factor W= / ,in This is the cumulative reflected power value collected in step S201 for the current detection period. Finally, the laser protection system uses the base compensation factor... The feedforward compensation coefficient is determined by the target weight factor W. When the W value is small near The feedforward compensation is limited; when the W value is large... The corresponding increase allows for a larger feedforward correction magnitude in the gradient collaborative characterization data. This determination method correlates the magnitude of the feedforward compensation coefficient with the current actual reflection power level, avoiding overcompensation under low reflection power conditions and providing sufficient compensation under high reflection power conditions.

[0093] In some embodiments, the transient reflection mapping model can be quickly retrieved via address indexing from a lookup table stored in the ROM of the FPGA, or obtained via floating-point polynomial evaluation using an embedded processor, with runtime coefficient updates supported. Other computational methods can also be used, which are not limited here. In some embodiments, the transient reflection mapping model may become mismatched with the actual reflection characteristics after replacing the focusing lens or adjusting the optical path geometry. To address this, the laser protection system provides an online model calibration interface. After optical path maintenance is completed, a calibration command is sent via the CNC system to trigger the laser protection system to execute a set of preset angular velocity scanning sequences. Under each angular velocity condition, the actual reflection power increment is collected and compared with the model output. The model parameters (lookup table entries or polynomial coefficients) are updated using the least squares method, so that the model output is realigned with the actual reflection characteristics after component replacement.

[0094] S207. Based on the gradient collaborative characterization data, perform compensation calculation on the preset initial standing wave ratio (SWR) determination threshold to generate a real-time determination threshold.

[0095] The gradient collaborative representation data used in this step is the data after feedforward correction in step S206. The meaning of the initial VSWR determination threshold and compensation calculation is explained in step S104.

[0096] Specifically, after completing the feedforward correction in step S206, the laser protection system reads the corrected gradient collaborative characterization data. And the initial standing wave ratio determination threshold Perform compensation calculations to generate real-time decision thresholds. The difference from step S104 is that the input data in this step... The feedforward compensation coefficient is already included, so the threshold is determined in real time when the robotic arm is about to perform a high curvature attitude deflection. The positive compensation offset is compared to that in step S104 based only on The calculated result is larger, providing ample margin for the impending surge in transient reflected power. The specific function form and amplitude constraint of the compensation operation are consistent with those described in step S104.

[0097] In some embodiments, the compensation operation including feedforward correction can... Directly substitute into the linear compensation function ΔTH=K*( - This can be achieved by setting independent gain coefficients for the feedforward incremental part and the original part, respectively, to achieve separate adjustment, i.e., ΔTH = *( - )+ *( - Other compensation structures can also be used, which are not limited here. In some embodiments, there may be a situation where feedforward compensation causes the real-time judgment threshold to rise sharply in a short period of time, resulting in the actual attitude deflection of the robotic arm being less than the value indicated by the pre-read command. In this case, an excessive judgment margin may cause the real fault signal to be blocked. To address this, the laser protection system will monitor whether the actual reflection power change gradient value matches the feedforward expectation within several detection cycles after the feedforward compensation takes effect. If the actual gradient value continues to be lower than the preset proportion of the feedforward expectation value, the laser protection system will gradually reduce the feedforward compensation coefficient to 1.0, so that the real-time judgment threshold returns to the normal compensation level when the feedforward expectation is not met.

[0098] In some embodiments, the laser protection system performs more refined multi-cycle data analysis on the threshold compensation operation, specifically including: constructing a data sequence containing gradient co-characterization data from the current detection cycle and N consecutive historical detection cycles; where N is an integer greater than 1; calculating the average rate of change of the data increasing interval in the data sequence as the rising edge average rate of change, and the average rate of change of the data decreasing interval as the falling edge average rate of change; calculating the ratio of the rising edge average rate of change to the falling edge average rate of change to determine the asymmetric distortion coefficient; and combining the asymmetric distortion coefficient and the gradient co-characterization data to perform compensation on a preset initial standing wave ratio (VSWR) determination threshold to generate a real-time determination threshold.

[0099] The data sequence refers to the gradient co-characterization data of the current detection cycle that the laser protection system reads from the circular buffer. and gradient co-characterization data for the first N detection cycles { , ,..., The ordered set of values ​​is arranged chronologically. N is a preset historical depth parameter, typically ranging from 8 to 64. The asymmetric distortion coefficient is the ratio of the average rate of change along the rising edge to the average rate of change along the falling edge. A ratio close to 1.0 indicates good symmetry, while a ratio deviating from 1.0 indicates a trend of continuous unilateral increase or decrease.

[0100] Specifically, after updating the gradient co-characterization data in each detection cycle, the laser protection system will... Write to the current position pointer in the circular buffer. The laser protection system reads N+1 data points from the circular buffer in chronological order to form a data sequence S={ , ,..., The difference between two adjacent points in a data sequence. Perform calculations one by one, Points greater than 0 are assigned to the increasing interval set. Points less than 0 are assigned to the decreasing interval set. The laser protection system applies to all points in the increasing interval set. Calculate the arithmetic mean to obtain the average rate of change along the rising edge. For all | in the decreasing interval set | Calculate the arithmetic mean to obtain the average rate of change along the falling edge. .exist When the asymmetric distortion coefficient A is not zero, the asymmetric distortion coefficient A = / .exist When the value is zero (i.e., there is no decreasing interval in the sequence), the laser protection system sets A to a preset upper limit saturation value. Laser protection systems are based on A and right Perform compensation calculations to generate .

[0101] In some embodiments, the calculation of asymmetric distortion coefficients can be implemented using an FPGA in a pipelined architecture, or using an embedded processor in a loop-based manner, or other data processing architectures can be employed, which are not limited here. In some embodiments, when the data sequence length N is short, insufficient sampling points may lead to insufficient statistical reliability of the average rate of change of the rising or falling edges. To address this, the laser protection system sets a minimum effective sample number threshold. The number of data points in the increasing or decreasing interval is less than When the corresponding average rate of change is invalid, the asymmetric distortion coefficient remains unchanged from the previous period to avoid abnormal distortion coefficient output due to insufficient statistical samples.

[0102] In some embodiments, the laser protection system performs differentiated compensation calculations for different value ranges of the asymmetric distortion coefficient. Specifically, when the asymmetric distortion coefficient is within a preset elastic recovery range, the laser protection system determines a target tolerance value based on gradient collaborative characterization data and a preset gain coefficient. When the asymmetric distortion coefficient is not within the elastic recovery range, the system calculates an attenuation penalty factor based on the asymmetric distortion coefficient and determines an attenuation compensation value in conjunction with the gradient collaborative characterization data. The system then corrects the initial VSWR judgment threshold based on the target tolerance value or the attenuation compensation value to determine the real-time judgment threshold.

[0103] The elastic recovery range refers to a preset numerical range of the asymmetric distortion coefficient. , The upper and lower bounds of this range are determined during the system calibration phase based on the typical fluctuation characteristics of gradient collaborative characterization data under normal processing conditions, with typical values ​​ranging from [0.7, 1.5]. When the asymmetric distortion coefficient is within this range, it indicates that the rise and fall behavior of the gradient collaborative characterization data is basically symmetrical, meaning that the fluctuation of reflected power belongs to recoverable normal operating condition changes. The target tolerance value refers to the threshold positive compensation offset calculated under the elastic recovery range conditions. The attenuation penalty factor is an attenuation coefficient less than 1.0 calculated when the asymmetric distortion coefficient deviates from the elastic recovery range, used to compress the compensation amplitude to tighten the protection sensitivity.

[0104] Specifically, after obtaining the asymmetric distortion coefficient A, the laser protection system determines whether A is within the elastic recovery range. , ]Inside. like ≤A≤ The laser protection system uses the gain coefficient G and gradient to jointly characterize the data. Calculate the target tolerance value =G*| - |, among which As the benchmark synergy ratio, and with = + As a real-time judgment threshold. If A < Or A> The laser protection system calculates the attenuation penalty factor D = 1 / (1 + |A - 1.0|). This formula means that the further A deviates from 1.0, the smaller D becomes, and the more strongly the compensation amplitude is attenuated. The laser protection system uses D and... Calculate the attenuation compensation value =D*G*| - | and with = + This serves as the real-time threshold. Since D < 1.0, < When the threshold is determined in real time and the asymmetric distortion coefficient deviates from the elastic recovery range, a smaller positive compensation offset is obtained, and the protection sensitivity is improved accordingly.

[0105] In some embodiments, the differential compensation operation can be completed within a deterministic clock cycle using the comparator and arithmetic unit of an FPGA, or implemented using an embedded processor with a conditional branch structure. Other compensation strategies can also be employed, which are not limited here. In some embodiments, there may be a situation where the asymmetric distortion coefficient repeatedly crosses the boundary near the elastic recovery interval, causing frequent switching between the target tolerance value and the attenuation compensation value, resulting in a jump in the real-time judgment threshold. To address this, the laser protection system sets a transition band at the interval boundary. When the value A is within the transition band, the distance of A from the center of the interval is used as the weight to adjust the value. and Linear interpolation is performed to ensure that the real-time threshold transitions smoothly near the interval boundary rather than switching abruptly.

[0106] S208. Calculate the ratio of the cumulative reflected power value to the output pulse width value to generate the VSWR monitoring value for the current detection cycle.

[0107] Refer to step S105, which will not be repeated here.

[0108] S209. Combining the VSWR monitoring value and the real-time judgment threshold, determine the VSWR status judgment result and output the corresponding interlocking protection level signal.

[0109] Refer to step S106, which will not be repeated here.

[0110] After step S209 completes the VSWR state determination and outputs the interlock protection level signal, the following working condition requires further processing: the VSWR monitoring value is lower than the real-time determination threshold (the determination result is normal), but it remains higher than the initial VSWR determination threshold. This state indicates that there is a continuous overflow of reflected energy in the optical path, the amplitude of which does not reach the interlock trigger condition. After this overflow energy is absorbed by the optical element, it will cause a continuous increase in the temperature of the lens or lens substrate.

[0111] When the uneven temperature distribution of optical components accumulates to a certain extent, the thermally induced change in the refractive index of the material will cause a thermal lensing effect in the lens, leading to beam focus drift and mode distortion. If this accumulation process is not tracked, the optical components may suffer permanent thermal damage if the VSWR monitoring value never triggers the interlocking protection. The following steps S210 to S212 quantify, track, and provide early warning of the cumulative damage risk caused by the thermal lensing effect due to the accumulation of subthreshold reflection energy.

[0112] S210. When the standing wave ratio (SWR) status determination result is normal, and the duration for which the SWR monitoring value is greater than the initial SWR determination threshold exceeds a preset duration, calculate the difference between the SWR monitoring value and the initial SWR determination threshold as the energy overflow amount.

[0113] Among them, the energy spillover refers to the difference between the current standing wave ratio (SWR) monitoring value and the initial SWR determination threshold. The unit is the same as the SWR monitoring value (W). This difference represents the level of additional reflected power density that the optical element experiences beyond the standard design conditions during the current period.

[0114] Specifically, after the laser protection system determines the VSWR status judgment result to be normal in step S209, it performs the following additional judgment: The VSWR monitoring value of the current detection cycle is... Compared with the initial standing wave ratio threshold Compare. If > The laser protection system will continuously increment the value of the exceeding counter by one; if ≤ The counter is reset to zero. The counter value is then multiplied by the detection period duration. The duration obtained exceeds the preset duration. At that time, the laser protection system calculates the energy overflow ΔE= - This energy spillover represents the additional thermal load on the optical element due to the absorption of excess reflected energy during the current cycle. For example, in =2.8w Under the condition of 2.5W, the energy overflow is 0.3W. The continuously exceeding counter value will continue to operate until the VSWR monitoring value drops back to [value missing]. The following values ​​are reset to zero to reflect the thermal recovery process of the optical element after the reflected energy is eliminated.

[0115] In some embodiments, duration monitoring can be implemented through a hardware counter module within the FPGA, a timer interrupt service routine of the embedded processor, or other timing architectures, which are not limited here. In some embodiments, the VSWR monitoring value may frequently cross the initial VSWR threshold, causing the continuous overshoot counter to repeatedly reset to zero. In this case, although the duration of each instance does not exceed the preset duration, the heat accumulated by the optical element during multiple short-term overshoots continues to accumulate because the intervals are insufficient to completely dissipate. To address this, the laser protection system will set an auxiliary accumulation counter, which will be used when the counter... When the count drops back to zero, the accumulated count value is multiplied by the preset residual heat coefficient and then added to the auxiliary accumulator counter. The value of the auxiliary accumulator counter decays exponentially with the preset decay time constant. When the equivalent duration of the auxiliary accumulator counter exceeds the preset duration, the calculation of energy overflow is also triggered.

[0116] S211. Perform a time integral operation on the energy overflow amount combined with the heat dissipation attenuation factor to generate the cumulative characteristic value of the thermal lensing effect.

[0117] The heat dissipation attenuation factor refers to the rate at which an optical element loses heat to the surrounding environment through thermal conduction and convection. The specific calibration method for this attenuation factor includes: during the system calibration phase, the optical element is continuously irradiated with a constant, known reflected power level until its temperature reaches thermal equilibrium. Then, the reflected power source is instantaneously removed, and a temperature sensor mounted on the optical element's base records the element's temperature over time as a natural cooling curve. An exponential decay fitting is performed on this cooling curve, and the reciprocal of the resulting exponential decay time constant is the heat dissipation attenuation factor γ. This process must be performed under the actual cooling conditions of the laser protection system (including cooling water flow rate and ambient temperature) to ensure that the calibration value reflects the real heat dissipation environment. Time integration refers to a discrete recursive calculation that accumulates and sums the energy spillover along the time axis while simultaneously taking into account the heat dissipation attenuation effect. The cumulative characteristic value of the thermal lensing effect is the accumulated amount obtained after time integration with heat dissipation attenuation. This value characterizes the degree of heat accumulation of the optical element at the current moment, and its physical meaning is positively correlated with the temperature gradient between the center and edge of the optical element's substrate.

[0118] Specifically, after completing the energy overflow calculation in step S210, the laser protection system performs the following discrete recursive calculation to update the cumulative characteristic value of the thermal lensing effect. The recursive formula is: ,in The cumulative characteristic value of the thermal lensing effect in the previous detection cycle is given by γ, where γ is the heat dissipation attenuation factor. For the duration of the detection cycle, This represents the energy overflow amount for the current cycle. This term characterizes the proportion of heat stored in an optical element due to heat dissipation over a detection cycle. This term represents the increase in heat storage within the current cycle. When... =0 (i.e., the VSWR monitoring value does not exceed the initial VSWR judgment threshold). It contains only an attenuation term, with the cumulative eigenvalue decaying exponentially to zero periodically, reflecting the natural cooling process of the optical element without additional thermal load. When If the value remains positive, and the input heat exceeds the dissipation, It will continue to increase.

[0119] In some embodiments, the time integration operation including heat dissipation attenuation can be implemented using an FPGA in a fixed-point format as the recursive formula. (Pre-calculated as a constant), or the heat dissipation attenuation factor can be dynamically adjusted at runtime via an embedded processor in floating-point format, or other discretization methods can be used, which are not limited here. In some embodiments, the actual heat dissipation capacity of the external cooling system for optical components may deviate from the calibrated value due to flow fluctuations. To address this, the laser protection system periodically reads the values ​​from the cooling water flow sensor and the optical component holder temperature sensor, and performs online correction of the heat dissipation attenuation factor γ based on the measured values ​​of flow and temperature: the γ value is reduced when the cooling flow is lower than the calibrated flow to reflect insufficient heat dissipation capacity, and the γ value is increased when the cooling flow is higher than the calibrated flow to reflect enhanced heat dissipation capacity. The corrected γ value is used for recursive calculations in subsequent cycles.

[0120] S212. When the cumulative characteristic value of the thermal lensing effect is greater than the preset thermal damage threshold, the standing wave ratio state determination result is overwritten as a warning state.

[0121] The thermal damage threshold refers to the upper limit of the cumulative characteristic value of thermal lensing effect when the refractive index non-uniformity shift of an optical element due to thermal accumulation exceeds the allowable range. This threshold is determined during the system calibration phase by calculating the thermo-optical coefficient, geometric dimensions, and allowable focal length drift of the optical element material. Specifically, it includes: the thermo-optical coefficient dn / dT (characterizing the rate of refractive index change caused by temperature change) of the optical material used in the optical element, the effective aperture D and thickness L of the element, and the maximum focal length drift allowed by the manufacturing process. Through the thermal lens focal length formula ;in The equivalent thermal lens focal length of the optical element due to thermal effects, when Shorten until the focus drift reaches When the center-edge temperature gradient is reached, the corresponding temperature gradient is: This allows us to deduce the maximum permissible temperature gradient between the center and edge of the optical element. Then the Multiplying the heat capacity of the optical element (i.e., the product of specific heat capacity and mass) yields the maximum allowable heat storage energy, which is the thermal damage threshold. The setting basis; in actual engineering, it is usually multiplied by a safety factor less than 1 (typically 0.6 to 0.8) to leave a safety margin. The warning state refers to a third VSWR state determination result, different from the normal state and fault state. In this state, the laser protection system does not cut off the RF excitation output, but sends a warning flag to the CNC system through a dedicated warning signal channel. The CNC system then performs power reduction operation or arranges shutdown for maintenance based on the warning flag. The overwrite operation refers to forcibly changing the normal state determination result determined in step S209 to the warning state.

[0122] Specifically, after completing the recursive calculation in step S211, the laser protection system updates the accumulated eigenvalue of the thermal lensing effect. With the preset thermal damage threshold Compare. If < The laser protection system maintains the normal state determination result determined in step S209 unchanged, and the interlock protection level signal continues to remain in a high-level enabled state. If ≥ The laser protection system overwrites the VSWR state determination result from the normal state to the warning state, and simultaneously generates a warning level or warning data frame at the warning signal output port. The interlock protection level signal does not flip to a low level in the warning state; the RF excitation source continues to output, but the CNC system, upon receiving the warning flag, can choose to reduce the laser output power or pause processing after the current processing segment is completed to allow the optical components to cool down, depending on the processing requirements. If in subsequent detection cycles... It dropped back to due to heat dissipation degradation. The laser protection system will then deactivate the warning and return to normal operation.

[0123] In some embodiments, the output of the warning status can be implemented in several ways: Optionally, the laser protection system outputs a warning level signal through a dedicated digital output port independent of the interlock protection level signal. This signal is connected to the input module of the CNC system to generate a software interrupt. The CNC system reads the warning level information and executes the corresponding power reduction strategy in the interrupt service routine; Optionally, the laser protection system sends a warning flag and a warning signal to the CNC system via a high-speed communication bus. The numerical status data frame, the CNC system according to Compared to Different levels of response measures will be selected based on the proportion exceeding the threshold. It is understood that other early warning signal transmission methods and response mechanisms may also be used, and no limitations are specified here.

[0124] In some embodiments, after prolonged processing, the absorptivity of optical elements may slowly increase due to the gradual accumulation of surface deposits, resulting in an actual temperature rise exceeding the expected temperature rise under calibrated conditions for the same energy spillover. To address this, the laser protection system reads the online transmittance monitoring data of the optical elements (acquired via a transmittance sampling detector in the optical path) before each processing task begins. The ratio of the measured transmittance to the factory-calibrated transmittance is used as an absorptivity correction coefficient. This correction coefficient is multiplied by the energy spillover amount and then substituted into the recursive calculation in step S211, ensuring that the calculation of the cumulative characteristic value of the thermal lensing effect reflects the enhanced thermal absorption characteristics of the optical elements due to surface degradation.

[0125] In this embodiment, the VSWR determination threshold of the laser protection system can be synchronously adapted and feedforward adjusted according to the dynamic changes in processing conditions and robotic arm posture. At the same time, it can quantitatively assess and warn of the risk of thermal lensing damage caused by subthreshold reflection accumulation, reduce the false trigger rate of interlocking protection while maintaining the sensitivity of optical path fault detection, and provide early protection against progressive thermal damage to optical components.

[0126] The laser protection system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference]. Figure 3 This is a schematic diagram of a physical device structure of a laser protection system in an embodiment of this application.

[0127] It should be noted that, Figure 3 The structure of the laser protection system shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0128] like Figure 3 As shown, the laser protection system includes a CPU 301, which can perform various appropriate actions and processes according to a program stored in ROM 302 or a program loaded from storage section 308 into RAM 303, such as performing the methods described in the above embodiments. RAM 303 also stores various programs and data required for system operation. CPU 301, ROM 302, and RAM 303 are interconnected via bus 304. I / O interface 305 is also connected to bus 304.

[0129] The following components are connected to I / O interface 305: input section 306 including audio input devices, push-button switches, etc.; output section 307 including liquid crystal display (LCD) and audio output devices, indicator lights, etc.; storage section 308 including hard disks, etc.; and communication section 309 including network interface cards such as LAN (Local Area Network) cards, modems, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0130] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by CPU 301, it performs the various functions defined in the present invention.

[0131] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0132] Specifically, the laser protection system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it implements the carbon dioxide laser fault interlocking protection method based on VSWR monitoring provided in the above embodiment.

[0133] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the laser protection system described in the above embodiments; or it may exist independently and not assembled into the laser protection system. The storage medium carries one or more computer programs that, when executed by a processor of the laser protection system, cause the laser protection system to implement the carbon dioxide laser fault interlocking protection method based on VSWR monitoring provided in the above embodiments.

[0134] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0135] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

Claims

1. A fault interlocking protection method for a carbon dioxide laser based on standing wave ratio monitoring, characterized in that, The method, applied to laser protection systems, includes: Collect the output pulse width and cumulative reflected power value for the current detection cycle; The differences between the output pulse width value, the cumulative reflected power value, and the corresponding value in the previous detection cycle are calculated respectively to generate the pulse width change gradient value and the reflected power change gradient value. Read the stored gradient co-representation data, and when the absolute value of the pulse width change gradient value is greater than or equal to the preset minimum gradient threshold, calculate the ratio of the reflection power change gradient value to the pulse width change gradient value to correct the gradient co-representation data. Based on the gradient collaborative characterization data, a compensation operation is performed on the preset initial standing wave ratio (SWR) determination threshold to generate a real-time determination threshold. Calculate the ratio of the cumulative reflected power value to the output pulse width value to generate the VSWR monitoring value for the current detection period; By combining the VSWR monitoring value and the real-time judgment threshold, the VSWR status judgment result is determined and the corresponding interlocking protection level signal is output.

2. The method according to claim 1, characterized in that, After the step of acquiring the output pulse width value and the cumulative reflection power value of the current detection cycle, the method further includes: Obtain the time-domain envelope signal of the reflected power within the current detection period; Calculate the time-domain energy centroid of the time-domain envelope signal and determine the hysteresis offset of the time-domain energy centroid relative to the reference output pulse centroid. The cumulative value of reflected power is corrected based on the hysteresis offset.

3. The method according to claim 1, characterized in that, Before the step of performing compensation calculation on the preset initial VSWR determination threshold based on the gradient collaborative characterization data to generate a real-time determination threshold, the method further includes: Obtain the pre-read interpolation command sent by the CNC system, and extract the rate of curvature change of the target machining trajectory from the pre-read interpolation command; When the rate of curvature change is greater than a preset attitude change threshold, a feedforward compensation coefficient that is positively correlated with the rate of curvature change is generated. The gradient collaborative characterization data is corrected based on the feedforward compensation coefficient.

4. The method according to claim 3, characterized in that, The step of generating a feedforward compensation coefficient positively correlated with the rate of curvature change when the rate of curvature change is greater than a preset attitude change threshold specifically includes: When the rate of curvature change is greater than a preset attitude change threshold, the machining attitude deflection angular velocity corresponding to the rate of curvature change in the pre-read interpolation command is extracted; The processing posture deflection angular velocity is input into a preset transient reflection mapping model to calculate the expected reflection power increment; Calculate the ratio of the expected increase in reflected power to the cumulative value of reflected power to generate the target weighting factor; The feedforward compensation coefficient is determined based on the preset basic compensation base and the target weight factor.

5. The method according to claim 1, characterized in that, The step of performing compensation calculations on a preset initial VSWR determination threshold based on the gradient collaborative characterization data to generate a real-time determination threshold specifically includes: Construct a data sequence containing gradient collaborative representation data from the current detection period and N consecutive historical detection periods; where N is an integer greater than 1. The average rate of change of the data increasing interval in the data sequence is calculated as the rising edge average rate of change, and the average rate of change of the data decreasing interval is calculated as the falling edge average rate of change. Calculate the ratio of the average rate of change of the rising edge to the average rate of change of the falling edge to determine the asymmetric distortion coefficient; By combining the asymmetric distortion coefficient and the gradient collaborative characterization data, compensation is performed on the preset initial standing wave ratio (SWR) determination threshold to generate the real-time determination threshold.

6. The method according to claim 5, characterized in that, The step of combining the asymmetric distortion coefficient and the gradient co-characterization data to compensate for the preset initial VSWR determination threshold and generate the real-time determination threshold specifically includes: When the asymmetric distortion coefficient is within a preset elastic recovery range, the target tolerance value is determined based on the gradient collaborative characterization data and the preset gain coefficient. When the asymmetric distortion coefficient is not within the elastic recovery range, an attenuation penalty factor is calculated based on the asymmetric distortion coefficient, and an attenuation compensation value is determined in combination with the gradient collaborative characterization data. The initial VSWR determination threshold is corrected based on the target tolerance value or the attenuation compensation value, and the real-time determination threshold is determined.

7. The method according to claim 1, characterized in that, After the step of combining the VSWR monitoring value and the real-time judgment threshold to determine the VSWR state judgment result and output the corresponding interlocking protection level signal, the method further includes: When the standing wave ratio (SWR) status determination result is normal, and the duration for which the SWR monitoring value is greater than the initial SWR determination threshold exceeds a preset duration, the difference between the SWR monitoring value and the initial SWR determination threshold is calculated as the energy overflow amount. Perform a time integral operation on the energy overflow amount combined with the heat dissipation attenuation factor to generate a cumulative characteristic value of thermal lensing effect; When the cumulative characteristic value of the thermal lensing effect exceeds the preset thermal damage threshold, the standing wave ratio state determination result is overwritten as a warning state.

8. A laser protection system, characterized in that, The laser protection system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the laser protection system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the laser protection system, it causes the laser protection system to perform the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the laser protection system, it causes the laser protection system to perform the method as described in any one of claims 1-7.

Images