Operation data monitoring method of intelligent laser grooving machine

Abstract

The invention discloses an operation data monitoring method of an intelligent laser grooving machine, particularly relates to the field of laser grooving, and aims to solve the problem of data acquisition misalignment caused by pollution of an optical sensor. The stability of high-precision feature recognition and parameter control can still be kept in a complex environment in the laser grooving process; by means of unified conditions established based on raw material characteristic parameters, on the basis of optical sensor collection, fine intervention can be achieved under pollution accumulation, efficiency requirements and multi-factor interference conditions through adaptive adjustment of cutting head motion parameters; according to the method, the accuracy of cutting quality judgment is effectively maintained in multi-source and multi-scale data changes, and continuous intensification of optical data distortion caused by pollution spreading is avoided, so that the reliability, precision and consistency of the whole cutting process are ensured.

Description

Technical Field

[0001] The present invention relates to the field of laser slotting, and more specifically, to an operation data monitoring method of an intelligent laser slotting machine. Background Art

[0002] In the field of high-precision laser processing, the laser cutting process is accompanied by a variety of unexpected interference factors. Especially in process links with high-precision requirements, optical sensors are often used to detect the quality of the cutting edge and the morphology of the molten pool in real time to assist in determining the position of the cutting head and its movement strategy. However, during the grooving process, there are often slag jumping, slag splashing, and steam condensate caused by the high temperature environment attached to the surface of the protective cover, resulting in hidden and continuous impact on the quality of optical sensor data acquisition. This pollution condition interferes with the normal propagation path of the transmitted and reflected light, so that the data obtained by the sensor deviates from the actual cutting state, which not only weakens the accuracy of the judgment of the cutting edge quality, but also interferes with the interpretation of the molten pool morphology and the positioning accuracy of the cutting head. When the pollution accumulation reaches a certain level, the traditional control method based on fixed process parameters is difficult to respond to this complex, multi-factor intertwined pollution change in time, which in turn causes the instability of the final processing accuracy and performance. In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an operation data monitoring method for an intelligent laser slotting machine, which dynamically adjusts the optical signal distortion caused by pollution, so that the laser cutting process can still maintain the stability of high-precision feature recognition and parameter control in a complex environment. With the help of unified conditions established based on the characteristic parameters of raw materials, on the basis of optical sensor acquisition, by adaptively adjusting the motion parameters of the cutting head, refined intervention can be achieved under conditions of pollution accumulation, efficiency requirements and multi-factor interference. This effectively maintains the accuracy of cutting quality judgment in multi-source and multi-scale data changes, avoids the continuous aggravation of optical data distortion caused by the spread of pollution, thereby ensuring the reliability, accuracy and consistency of the overall cutting process, so as to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A method for monitoring the operation data of an intelligent laser slotting machine comprises the following steps:

[0006] S1. Under the same raw material conditions, fix the raw material parameters and use all parameter data as the initial benchmark conditions for pollution judgment and efficiency requirement analysis;

[0007] S2. Under the initial reference conditions, the optical sensor is used to obtain the transmission and reflection data of the protective cover surface during the cutting process, the pollution level index is calculated, and the index is compared with the preset pollution threshold;

[0008] S3. If the pollution level index exceeds the pollution threshold, the initial approximation step is set according to the cutting efficiency requirements and the pollution index deviation, the cutting head movement speed is corrected, the corrected speed parameter is returned to step S2 and the pollution level index is calculated again to confirm the impact of the speed correction on the pollution status.

[0009] In a preferred embodiment, step S1 includes the following contents:

[0010] Under the same raw material conditions, based on known database records, the four parameters of raw material type, thickness, density and melting point are uniformly extracted, and the values ​​of the four parameters are converted into deterministic scalars through pre-calibrated quantitative tools to ensure that there is no miscellaneous information in the records.

[0011] In a preferred embodiment, step S2 includes the following contents:

[0012] The calculation of the transmission index aims to quantify the attenuation and scattering effects of pollutants on the laser beam when it passes through the protective cover, thereby reflecting the degree of decline in the light transmittance of the protective cover; the calculation of the reflection index focuses on evaluating the influence of pollutants on the morphology and intensity of the reflected light signal; the transmission index and the reflection index are coupled through a nonlinear iterative process to construct a two-variable equation group, iteratively solve the steady-state point, and calculate the final pollution level index through a nonlinear mapping formula.

[0013] In a preferred embodiment, the calculation process of the transmission index TI is:

[0014] A1. Under the condition of fixing the characteristic parameters of the raw materials, the original transmission signal is obtained from the transmission optical sensor; in order to eliminate environmental interference and weak light noise, the segmented adaptive filtering method is used to initially purify the original transmission signal to form a purified transmission signal;

[0015] A2. Input the purified transmission signal into the improved empirical mode decomposition process, and decompose the purified transmission signal into several intrinsic mode function IMF sets by using the nonlinear kernel decomposition of the adaptive kernel function. Each eigenmode function corresponds to a specific time-frequency local feature of the signal;

[0016] A3. For each intrinsic mode function IMF T,i (t) Complexity measurement is performed to capture the non-stationary perturbation characteristics caused by pollution; the complexity measurement uses fractional-order operators and saddle point mapping method:

[0017] First, the fractional integral transform of each element of the eigenmode function set in a given interval is calculated get After transformation Using the multi-scale saddle point optimization algorithm, the local extreme point distribution of the signal at multiple scales is statistically analyzed, and the saddle point density function φ is recorded. T,i (λ), where λ represents the multi-scale parameter;

[0018] A4. Use the cross-dependency operator to integrate the complexity information of each eigenmode function: Where M represents the total number of intrinsic mode functions decomposed by the nonlinear intrinsic mode decomposition method; i and j are indexes used to traverse the intrinsic mode function pairs, where i ranges from 1 to M-1 and j ranges from i+1 to M; where β>1 is the nonlinear amplification parameter;

[0019] A5. Finally, Γ T Normalized to the interval 0,1, the transmission index TI is obtained.

[0020] In a preferred embodiment, the calculation process of the reflection index RI is:

[0021] B1. Obtain the reflection signal from the reflection optical sensor, first suppress the interference through adaptive nonlinear filtering, and then map the purified reflection signal to the time-frequency-morphology joint space; this mapping uses the curve manifold expansion technology:

[0022] The time-frequency distribution P is obtained by performing time-frequency analysis on the purified reflection signal. R (t,f); project the time-frequency distribution into the two-dimensional embedding space, use the Rips complex network construction method, take the signal energy density as the weight, and form a weighted graph with a topological structure from the point set in the two-dimensional embedding space;

[0023] B2. Compute topological invariants and complex features for weighted graphs, including:

[0024] High-order Betti number array: Calculate the scale-varying curves of the 1st and 2nd order Betti numbers to measure the changes in the number of holes and cavities in the signal at different scales;

[0025] Asymmetric adjacency spectrum energy: Spectral decomposition is performed on the adjacency matrix of the weighted graph, and a specific singular value distribution is selected to perform integral measurement on the asymmetric structure, thereby obtaining the non-uniform characteristics of the reflected signal morphology in the high-dimensional embedding space;

[0026] B3. Using multi-scale box counting, the topological features are combined with fractal complexity. Based on the box counting, the box size is corrected by the non-uniform distortion mapping function Ψ, and the equivalent fractal dimension is calculated in a nonlinear way: The non-uniform distortion mapping function dynamically adjusts the box weights according to the adjacency spectrum characteristics and the changes in the Betti number to obtain heterogeneous fractal metrics;

[0027] B4. Combine the high-order Betti number distribution with the fractal dimension to calculate the reflection index: Among them, γ1, γ2, μ are all superlinear parameters greater than 1, (D R,ref ) is the fractal dimension value under the reference state.

[0028] In a preferred embodiment, the calculation process of the pollution level index is:

[0029] C1. Take the transmittance index and reflectance index as initial conditions, construct a nonlinear iterative equation group, and obtain the pollution level index by iteratively solving the steady-state solution;

[0030] Assume that two coupled equations are defined:

[0031] X n+1 =f(X n ,Y n )=X n +α1(TI-X n RI);

[0032] Y n+1 =g(X n ,Y n )=Y n +ζ(RI-Y n TI);

[0033] Where X n With Y n is the iterative variable, α1,ζ are the adjustment parameters, and the initial value is X0=TI, Y0=RI; through this dual variable coupling iteration, X and Y are mutually balanced, and the iteration is sought to reach a stable point (X * ,Y * ) state, if the stable point satisfies the convergence condition |X n+1 -X n |<∈ and |Y n+1 -Y n |<∈, then this steady-state point is processed to calculate the final pollution level index;

[0034] C2. After obtaining the steady-state point, perform nonlinear mapping on it to obtain the pollution level index: Where θ>1 is a nonlinear amplification parameter used to enhance the * With Y * Sensitivity when significant differences exist;

[0035] C3. Finally, the calculated pollution level index is compared with the preset pollution threshold:

[0036] If the pollution level index is greater than the pollution threshold, it means that after comprehensively considering the multi-scale complexity of the transmission and reflection signals, the pollution level has exceeded the acceptable range; if the pollution level index is less than or equal to the pollution threshold, it means that under the current parameter conditions, the pollution is still within a reasonable range and no cutting parameter adjustment is required.

[0037] In a preferred embodiment, step S3 includes the following contents:

[0038] After confirming that the pollution level index has exceeded the preset pollution threshold, it is necessary to make an initial approximation correction to the cutting head movement speed, so as to achieve a balance between the dynamic regulation of the pollution level and the efficiency requirements; on this basis, by using the comparison result between the pollution level index and the pollution threshold as the input condition, this step develops the following specific processing logic:

[0039] Firstly, according to the difference between the pollution level index and its preset pollution threshold and the cutting efficiency requirement, a setting strategy for the initial approximation step size is constructed. The correlation between the pollution degree deviation and the efficiency deviation is characterized by nonlinear mapping and multidimensional characteristic functions. Specifically, the magnitude of the pollution index exceeding the pollution threshold is recorded as ΔP, and the deviation between the cutting efficiency requirement and the current efficiency inference value is recorded as ΔE. Perform coupling transformation on ΔP and ΔE to obtain the initial approximation step length δv:

[0040] In a preferred embodiment, step S3 further includes the following contents:

[0041] After obtaining the initial approximation step length, set the current cutting head moving speed to v current Update the new speed setting value v according to the following formula new :v new =v current ±δv; the direction of ± depends on the speed change trend required to reduce or stabilize the pollution level index;

[0042] After completing the speed parameter correction, the new speed setting value is fed back to step S2 and step S3, and the optical sensor is used to collect and analyze the transmission and reflection data of the protective cover surface again under the same raw material conditions as required by step S2, and the pollution level index is calculated repeatedly; once the recalculated pollution level index is input into this step, a closed-loop judgment and correction process is formed.

[0043] The technical effects and advantages of the operation data monitoring method of an intelligent laser slotting machine of the present invention are as follows:

[0044] By selecting uniform and clearly defined raw material characteristic parameters as the basis, and on this basis, implementing adaptive adjustments with nonlinear coupling characteristics to the optical signal distortion caused by pollution, laser cutting can still achieve dynamic balance at a subtle level under multi-dimensional disturbances. By using complex analysis of transmission and reflection data, potential pollution factors are mapped into measurable pollution level indicators, and through approximation strategies, a balance point that does not rely on traditional linear relationships is found between pollution diffusion and efficiency requirements. As a result, the cutting process can still quickly stabilize through precise parameter iteration when facing different types and degrees of pollution intrusion, thereby maintaining high precision and consistency in cutting edge identification, molten pool morphology interpretation and cutting head positioning. Under this technical logic, laser cutting equipment is no longer dragged down by passive tolerance of pollution accumulation, but can actively perceive, analyze and instantly match the preferred parameters required for complex environments, so that the entire processing process remains controllable and excellent in the iterative changes of potential influencing factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 The present invention is a flowchart of an operation data monitoring method of an intelligent laser slotting machine. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Embodiment 1: Figure 1 The present invention provides an operation data monitoring method of an intelligent laser slotting machine, comprising:

[0048] S1. Under the same raw material conditions, fix the raw material parameters and use all parameter data as the initial benchmark conditions for pollution judgment and efficiency requirement analysis;

[0049] S2. Under the initial reference conditions, the optical sensor is used to obtain the transmission and reflection data of the protective cover surface during the cutting process, the pollution level index is calculated, and the index is compared with the preset pollution threshold;

[0050] S3. If the pollution level index exceeds the pollution threshold, the initial approximation step is set according to the cutting efficiency requirements and the pollution index deviation, the cutting head movement speed is corrected, the corrected speed parameter is returned to step S2 and the pollution level index is calculated again to confirm the impact of the speed correction on the pollution status.

[0051] Step S1 includes the following contents:

[0052] Under the same raw material conditions, based on the known database records, the four parameters of raw material type, thickness, density, and melting point are uniformly extracted, and the values ​​of the four parameters are converted into deterministic scalars through pre-calibrated quantitative tools to ensure that there is no miscellaneous information in the records. After the parameter extraction is completed, the four scalar parameter sets are input into the subsequent judgment process as the initial benchmark conditions. This benchmark condition provides a strictly defined physical boundary for the comparison link between the pollution level index and the efficiency requirements, so that the subsequent optical data analysis, pollution threshold comparison and cutting head movement speed adjustment steps are all carried out within the fixed raw material characteristics framework, thereby ensuring that the subsequent judgment process is not affected by changes in material characteristics.

[0053] Under the condition of fixed raw materials, this step uses a high-precision optical sensor system to simultaneously collect the transmitted light and reflected light data of the protective cover surface during the cutting process. Through complex signal processing and multi-level nonlinear analysis methods, the transmission index and reflection index are calculated respectively, and the two are combined through a nonlinear coupling algorithm to generate a pollution level index. Finally, the calculated PLI is strictly compared with the preset pollution threshold to determine whether the pollution level during the cutting process exceeds the acceptable range, thereby providing a scientific basis for the subsequent adjustment of cutting parameters.

[0054] Step S2 includes the following contents:

[0055] During the laser cutting process, the contamination of the protective cover has a significant impact on the cutting quality and equipment performance. In order to fully evaluate the impact of contamination on the cutting process, the changes in the transmitted light and the reflected light must be analyzed simultaneously. The calculation of the transmission index aims to quantify the attenuation and scattering effects of contaminants on the laser beam when it passes through the protective cover, thereby reflecting the degree of decline in the light transmittance of the protective cover. By analyzing the frequency components and energy distribution in the transmitted light signal in detail, the transmission index can reveal the changes in the optical signal caused by tiny contaminants and provide a sensitivity index for the degree of contamination. At the same time, the calculation of the reflection index focuses on evaluating the impact of contaminants on the morphology and intensity of the reflected light signal. Through complex time-frequency-morphology joint analysis, the reflection index can capture the nonlinear and heterogeneous changes in the reflected light signal caused by surface contamination, reflecting the damage and contamination degree of the surface structure of the protective cover. Combining the comprehensive analysis of the transmission index and the reflection index, a comprehensive contamination level index can be generated, which not only reflects the comprehensive changes in optical transmission and reflection characteristics, but also ensures the accuracy and reliability of contamination assessment. Therefore, calculating the transmission index and reflection index is a key step to achieve accurate monitoring of protective cover contamination and optimization of the cutting process.

[0056] On the premise of unifying the characteristic parameters of the raw materials, the laser cutting operation is started, and the transmission and reflection optical sensor systems are started at the same time. First, the transmission optical sensor captures the laser transmission signal after passing through the protective cover, and the reflection optical sensor captures the laser signal reflected back to the protective cover by the cutting head. For the transmission signal, adaptive segmented filtering and nonlinear intrinsic mode decomposition methods are used to decompose it into multiple intrinsic mode functions. For each intrinsic mode function, fractional-order integral transform and multi-scale saddle point mapping method are applied to extract complexity metrics, and the complexity information of all intrinsic mode functions is integrated through nonlinear coupling operators, and finally the transmission index is normalized.

[0057] Calculation process of transmission index TI:

[0058] A1. Under the condition of fixed raw material characteristic parameters, the original transmission signal S is obtained from the transmission optical sensor T (t). In order to eliminate environmental interference and weak light noise, the segmented adaptive filtering method is used to initially purify the original transmission signal to form a purified transmission signal S T,clean (t).

[0059] A2. Input the purified transmission signal into the improved empirical mode decomposition (EMD) process, but the empirical mode decomposition here is not the traditional form, but the nonlinear kernel EMD decomposition using the adaptive kernel function to decompose the purified transmission signal into a set of intrinsic mode functions (IMFs). Each eigenmode function corresponds to a specific time-frequency local feature of the signal.

[0060] A3. For each intrinsic mode function IMF T,i (t) Complexity measurement is performed to capture the non-stationary perturbation characteristics caused by pollution. The complexity measurement uses fractional order operators and saddle point mapping method:

[0061] First, the fractional integral transform of each element of the eigenmode function set in a given interval is calculated (α≠1, non-integer order), we get Fractional integrals are defined using Riemann–Liouville or Caputo to achieve increased sensitivity to local fine structures.

[0062] After transformation Using the multi-scale saddle point optimization algorithm, the local extreme point distribution of the signal at multiple scales is statistically analyzed, and the saddle point density function φ is recorded. T,i (λ), where λ represents a multi-scale parameter. This saddle point density function is used to measure the internal microstructure fluctuation of the intrinsic mode function.

[0063] A4. Use the cross-dependency operator to integrate the complexity information of each eigenmode function: Where M represents the total number of intrinsic mode functions decomposed by the nonlinear intrinsic mode decomposition method; i and j are indexes used to traverse the intrinsic mode function pairs, where i ranges from 1 to M-1, and j ranges from i+1 to M.

[0064] Here, β>1 is a nonlinear amplification parameter. By performing double integration and absolute difference power processing on the saddle point density distribution between different intrinsic mode functions, the differences between the complexity of each intrinsic mode function are nonlinearly coupled to capture the complex interweaving effect caused by contamination in the overall transmission signal.

[0065] A5. Finally, Γ T Normalized to the interval 0, 1, the transmission index TI is obtained. A higher transmission index value indicates that the saddle point distribution of the transmission signal at multiple scales has more significant and complex structural changes, that is, the surface contamination is more serious.

[0066] For the reflection signal, adaptive nonlinear filtering is first performed to purify the signal, and then the curve manifold expansion technique is used to map it to the time-frequency-morphology joint space. In this space, multiple topological features are extracted through high-order Betti number arrays and asymmetric adjacent spectrum energy, and the fractal dimension is calculated by combining the multi-scale box counting method and the non-uniform distortion mapping function. Subsequently, the nonlinear mapping function is used to couple the high-order Betti number distribution with the fractal dimension to obtain the reflection index.

[0067] Calculation process of reflection index RI:

[0068] B1. Obtaining reflection signal S from the reflection optical sensor R (t), first suppress the interference through adaptive nonlinear filtering, and then purify the reflected signal S R,clean (t) is mapped to the joint time-frequency-morphology space. This mapping uses the curve manifold expansion technique:

[0069] Perform time-frequency analysis (such as Chirplet transform or S transform) on the purified reflection signal to obtain the time-frequency distribution P R (t,f).

[0070] The time-frequency distribution is projected into a two-dimensional embedding space, and the Rips complex network construction method is used to form a weighted graph with a topological structure using the signal energy density as the weight.

[0071] B2. Compute topological invariants and complex features for weighted graphs, including:

[0072] High-order Betti number array: Calculate the scale-varying curves of the 1st and 2nd order Betti numbers to measure the changes in the number of holes and cavity structures in the signal at different scales.

[0073] Asymmetric adjacency spectrum energy: The adjacency matrix of the weighted graph is spectrally decomposed, and a specific singular value distribution is selected to perform integral measurement on the asymmetric structure, thereby obtaining the non-uniform characteristics of the reflected signal morphology in the high-dimensional embedding space.

[0074] B3. Using multi-scale box counting, the topological features are combined with fractal complexity. Based on the box counting, the box size is corrected by the non-uniform distortion mapping function Ψ, and the equivalent fractal dimension is calculated in a nonlinear way: The non-uniform distortion mapping function dynamically adjusts the box weights according to the adjacency spectrum characteristics and the changes in the Betti number to obtain a heterogeneous fractal metric.

[0075] B4. Combine the high-order Betti number distribution with the fractal dimension to calculate the reflection index: Among them, γ1, γ2, μ are all superlinear parameters greater than 1, (D R,ref ) is the fractal dimension value under the reference state. A high reflection index value indicates that the reflection signal has a more complex and heterogeneous topological structure and fractal properties in the high-dimensional morphological space, corresponding to a higher degree of pollution impact.

[0076] The transmission index and reflection index are coupled through a nonlinear iterative process to construct a two-variable equation set, iteratively solve the steady-state point, and calculate the final pollution level index through a nonlinear mapping formula. This index comprehensively reflects the multi-scale complex impact of surface pollution on the transmission and reflection light signals of the protective cover. Finally, the calculated pollution level index is compared with the preset pollution threshold to determine whether the pollution level reaches the trigger condition that needs to be adjusted, and the comparison results are recorded for use in subsequent steps.

[0077] Calculation process of pollution level index:

[0078] C1. Taking the transmittance index and reflectance index as initial conditions, a nonlinear iterative equation group is constructed, and the pollution level index is obtained by iteratively solving the steady-state solution.

[0079] Assume that two coupled equations are defined:

[0080] X n+1 =f(X n ,Y n )=X n +α1(TI-X n RI);

[0081] Y n+1 =g(X n ,Y n )=Y n +ζ(RI-Y n TI);

[0082] Where X n With Y n is the iterative variable, α1,ζ are the adjustment parameters, and the initial value is X0=TI, Y0=RI; through this dual variable coupling iteration, X and Y are mutually balanced, and the iteration is sought to reach a stable point (X * ,Y * ) state, if the stable point satisfies the convergence condition |X n+1 -X n |<∈ and |Y n+1 -Y n |<∈, then this steady-state point will be further processed to calculate the final pollution level index.

[0083] C2. After obtaining the steady-state point, perform nonlinear mapping on it to obtain the pollution level index: Where θ>1 is a nonlinear amplification parameter used to enhance the * With Y * Sensitivity when there is a significant difference. This formula renormalizes the relationship of the iterative steady-state point and transforms the complex coupling relationship between the transmission index and the reflection index into an independent indicator after nonlinear balance.

[0084] C3. Finally, the calculated pollution level index is compared with the preset pollution threshold:

[0085] If the pollution level index is greater than the pollution threshold, it means that after comprehensively considering the multi-scale complexity of the transmission and reflection signals, the pollution level has exceeded the acceptable range, and this result will be passed to step S3.

[0086] If the pollution level index is less than or equal to the pollution threshold, it means that the pollution is still within a reasonable range under the current parameter conditions and there is no need to adjust the cutting parameters.

[0087] This step uses advanced optical sensing technology and complex signal processing algorithms to achieve high-precision monitoring and evaluation of the surface contamination of the protective cover during laser cutting. The transmission index and reflection index use multi-level nonlinear analysis methods to extract pollution characteristics from the frequency domain and time-frequency-morphology joint space perspectives, respectively, to ensure the high sensitivity and discrimination of the pollution level index. The pollution level index obtained by comprehensive calculation provides a scientific and reliable basis for pollution assessment, enabling the system to accurately judge the pollution status and lay a solid foundation for the subsequent optimization and adjustment of cutting parameters, ensuring the efficiency and cleanliness of the cutting process.

[0088] Based on the calculation of fixed raw material parameters and complex pollution level indicators, the current step attempts to implement precise parameter fine-tuning for the detected pollution over-limit state by determining the initial approximation step size and evaluating the cutting efficiency deviation. The previous step has provided a detailed coupling relationship between pollution and efficiency. Based on this premise, the cutting head movement speed will be corrected in an orderly manner to achieve a delicate balance between pollution suppression and efficiency maintenance. This move aims to establish a closed-loop process from detection to regulation, so that parameter adjustment is no longer a blind attempt, but rather condenses pollution and efficiency information through nonlinear mapping functions, thereby gradually approaching a more optimized operating state.

[0089] Step S3 includes the following contents:

[0090] After confirming that the pollution level index has exceeded the preset pollution threshold, it is necessary to make a refined initial approximation correction to the cutting head movement speed, so as to achieve a balance between the dynamic regulation of the pollution level and the efficiency requirements. On this basis, using the comparison result between the pollution level index and the pollution threshold as the input condition, this step develops the following specific processing logic:

[0091] First, according to the difference between the pollution level index and its preset pollution threshold and the cutting efficiency requirement, a setting strategy for the initial approximation step size is constructed. The correlation between the pollution level deviation and the efficiency deviation is characterized by nonlinear mapping and multidimensional characteristic functions. Specifically, the extent to which the pollution index exceeds the pollution threshold is recorded as ΔP, and the deviation between the cutting efficiency requirement (such as the minimum allowable efficiency parameter based on the pre-calibration) and the current efficiency inference value is recorded as ΔE. Through the nonlinear function Perform coupling transformation on ΔP and ΔE to obtain the initial approximation step length δv:

[0092] Nonlinear functions The design focuses on nonlinear characteristics and fractional-order perturbation responses. For example, a composite form based on fractional-order integration and piecewise monotonic mapping is adopted, so that when ΔP is large, δv tends to adjust slightly larger to quickly lower the pollution level index; but when ΔE shows that the cutting efficiency is close to the critical requirement, δv will be constrained to a smaller range, thereby slowly releasing the pollution index without significantly damaging the efficiency.

[0093] The following is an example of a nonlinear function F. The example formula is for reference only. The parameters and structure can be adjusted according to actual needs.

[0094] Suppose that some parameters α2, β2, γ2, ρ2, μ2, η2 are all positive real numbers greater than 1, C and D are positive constants, and ∈0 is a very small positive real number to avoid the denominator being zero. Introduce fractional-order operations and nonlinear operators, let:

[0095]

[0096] The meaning and processing process of each part are as follows:

[0097] Nonlinear fractional operations Use fractional integration (such as Riemann–Liouville or Caputo definition) to fractionally integrate |ΔPΔE| μ2>1 makes the process extremely sensitive to small disturbances and captures the subtle correlation characteristics of ΔP and ΔE. The existence of fractional integrals ensures that microstructural changes at different scales can be amplified and extracted.

[0098] Coupling term ratio adjustment: Partially emphasizes the impact on the initial approximation step size when ΔP is large, while |ΔE| β2 The existence of inhibits over-adjustment in the case of high efficiency deviation, that is, if the efficiency is close to the bottom line, the step size should not be too large.

[0099] The analysis of the composite deviation structure is further enhanced. When |ΔPΔE| is amplified by fractional-order integration, it presents non-uniform characteristics, and its power increase γ2 ensures that it has an additional impact on the step size when complex coupling anomalies occur; while the 1+|ΔP| in the denominator ρ2 This prevents uncontrolled large step sizes from occurring under extremely high contamination deviations.

[0100] The selection of α2, β2, γ2, ρ2, μ2, and η2 needs to be customized according to the actual application scenario, and reasonable values ​​can be determined through experiments or theoretical analysis.

[0101] C and D are calibration coefficients, which are used to constrain the calculation results within a suitable step size range.

[0102] After obtaining the initial approximation step length, set the current cutting head moving speed to v current Update the new speed setting value v according to the following formula new :v new =v current ±δv;

[0103] The direction of ± depends on the speed change trend required to lower or stabilize the pollution level index: if increasing the speed helps to reduce pollution deposition (such as reducing the adhesion time by passing through the cutting area faster), then + is taken; if reducing the speed is beneficial to optimizing the optical characteristic signal (such as slowing down the surface adhesion growth rate), then - is taken.

[0104] After the speed parameter correction is completed, the new speed setting value is fed back to step S2 and step S3, and the optical sensor collects and analyzes the transmission and reflection data of the protective cover surface again under the same raw material conditions as required by step S2, and the pollution level index is calculated repeatedly. At this time, the new pollution level index not only reflects the impact of the adjusted speed on the pollution situation, but also provides a basis for further optimization. Once the recalculated pollution level index is input into this step, a closed-loop judgment and correction process is formed.

[0105] In short, this step establishes a nonlinear approximation relationship with pollution deviation and efficiency deviation as input, selects an appropriate initial approximation step to make a slight correction to the cutting head speed, and then returns to the data acquisition and calculation process to analyze the change trend of the pollution level index again to determine the effectiveness and necessity of this step adjustment strategy. In this way, the processing logic of step S3 realizes the dynamic, precise and progressive adjustment of the complex relationship between pollution and efficiency on the basis of ensuring the consistency of the physical scene and the orderliness of data transmission, laying a high-precision foundation for the continuous optimization of subsequent iterative steps.

[0106] After obtaining the result of the pollution level index exceeding the limit, this step uses the nonlinear coupling method to convert the pollution deviation and efficiency deviation into the initial approximation step size, so as to make targeted corrections to the cutting head movement speed. By returning the corrected speed parameters to the data acquisition and calculation link and analyzing the change trend of the pollution level index again, the effectiveness of the selected approximation strategy can be confirmed. In this carefully designed iterative process, the speed parameters can not only meet the dual constraints of pollution and efficiency, but also gradually tend to more reasonable reference values ​​after multiple trials and adjustments.

[0107] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0108] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0109] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0110] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for monitoring the operation data of an intelligent laser slotting machine, characterized in that: Includes steps: S1. Under the same raw material conditions, fix the raw material parameters and use all parameter data as the initial benchmark conditions for pollution judgment and efficiency requirement analysis; S2. Under the initial reference conditions, the optical sensor is used to obtain the transmission and reflection data of the protective cover surface during the cutting process, the pollution level index is calculated, and the index is compared with the preset pollution threshold; S3. If the pollution level index exceeds the pollution threshold, the initial approximation step is set according to the cutting efficiency requirements and the pollution index deviation, the cutting head movement speed is corrected, the corrected speed parameter is returned to step S2 and the pollution level index is calculated again to confirm the impact of the speed correction on the pollution status.

2. The method for monitoring the operation data of an intelligent laser slotting machine according to claim 1, characterized in that: Step S1 includes the following contents: Under the same raw material conditions, various parameters of raw materials are uniformly extracted according to known database records, and the values ​​of various parameters are converted into deterministic scalars through pre-calibrated quantitative tools to ensure that there is no miscellaneous information in the records.

3. The method for monitoring the operation data of an intelligent laser slotting machine according to claim 2, characterized in that: Step S2 includes the following contents: The calculation of the transmission index aims to quantify the attenuation and scattering effects of pollutants on the laser beam when it passes through the protective cover, thereby reflecting the degree of decline in the light transmittance of the protective cover; the calculation of the reflection index focuses on evaluating the influence of pollutants on the morphology and intensity of the reflected light signal; the transmission index and the reflection index are coupled through a nonlinear iterative process to construct a two-variable equation group, iteratively solve the steady-state point, and calculate the final pollution level index through a nonlinear mapping formula.

4. The method for monitoring the operation data of an intelligent laser slotting machine according to claim 3, characterized in that: Calculation process of transmission index TI: A1. Under the condition of fixing the characteristic parameters of the raw materials, the original transmission signal is obtained from the transmission optical sensor; in order to eliminate environmental interference and weak light noise, the segmented adaptive filtering method is used to initially purify the original transmission signal to form a purified transmission signal; A2. Input the purified transmission signal into the improved empirical mode decomposition process, and decompose the purified transmission signal into several intrinsic mode function IMF sets by using the nonlinear kernel decomposition of the adaptive kernel function. Each eigenmode function corresponds to a specific time-frequency local feature of the signal; A3. For each intrinsic mode function IMF T,i (t) Complexity measurement is performed to capture the non-stationary perturbation characteristics caused by pollution; the complexity measurement uses fractional-order operators and saddle point mapping method: First, the fractional integral transform of each element of the eigenmode function set in a given interval is calculated get After transformation Using the multi-scale saddle point optimization algorithm, the local extreme point distribution of the signal at multiple scales is statistically analyzed, and the saddle point density function φ is recorded. T,i (λ), where λ represents the multi-scale parameter; A4. Use the cross-dependency operator to integrate the complexity information of each eigenmode function: Where M represents the total number of intrinsic mode functions decomposed by the nonlinear intrinsic mode decomposition method; i and j are indexes used to traverse the intrinsic mode function pairs, where i ranges from 1 to M-1 and j ranges from i+1 to M; where β>1 is the nonlinear amplification parameter; A5. Finally, Γ T Normalized to the interval 0,1, the transmission index TI is obtained.

5. The method for monitoring the operation data of an intelligent laser slotting machine according to claim 3, characterized in that: Calculation process of reflectance index RI: B1. Obtain the reflection signal from the reflection optical sensor, first suppress the interference through adaptive nonlinear filtering, and then map the purified reflection signal to the time-frequency-morphology joint space; this mapping uses the curve manifold expansion technology: The time-frequency distribution P is obtained by performing time-frequency analysis on the purified reflection signal. R (t,f); project the time-frequency distribution into the two-dimensional embedding space, use the Rips complex network construction method, take the signal energy density as the weight, and form a weighted graph with a topological structure from the point set in the two-dimensional embedding space; B2. Compute topological invariants and complex features for weighted graphs, including: High-order Betti number array: Calculate the scale-varying curves of the 1st and 2nd order Betti numbers to measure the changes in the number of holes and cavities in the signal at different scales; Asymmetric adjacency spectrum energy: Spectral decomposition is performed on the adjacency matrix of the weighted graph, and a specific singular value distribution is selected to perform integral measurement on the asymmetric structure, thereby obtaining the non-uniform characteristics of the reflected signal morphology in the high-dimensional embedding space; B3. Using multi-scale box counting, the topological features are combined with fractal complexity. Based on the box counting, the box size is corrected by the non-uniform distortion mapping function Ψ, and the equivalent fractal dimension is calculated in a nonlinear way: The non-uniform distortion mapping function dynamically adjusts the box weights according to the adjacency spectrum characteristics and the changes in the Betti number to obtain heterogeneous fractal metrics; B4. Combine the high-order Betti number distribution with the fractal dimension to calculate the reflection index: Among them, γ1, γ2, μ are all superlinear parameters greater than 1, (D R,ref ) is the fractal dimension value under the reference state.

6. The method for monitoring the operation data of an intelligent laser slotting machine according to claim 3, characterized in that: Calculation process of pollution level index: C1. Take the transmittance index and reflectance index as initial conditions, construct a nonlinear iterative equation group, and obtain the pollution level index by iteratively solving the steady-state solution; Assume that two coupled equations are defined: X n+1 =f(X n ,Y n )=X n +α1(TI-X n RI); Y n+1 =g(X n ,Y n )=Y n +ζ(RI-Y n TI); Where X n With Y n is the iterative variable, α1,ζ are the adjustment parameters, and the initial value is X0=TI, Y0=RI; through this dual variable coupling iteration, X and Y are mutually balanced, and the iteration is sought to reach a stable point (X * ,Y * ) state, if the stable point satisfies the convergence condition |X n+1 -X n |<∈ and |Y n+1 -Y n |<∈, then this steady-state point is processed to calculate the final pollution level index; C2. After obtaining the steady-state point, perform nonlinear mapping on it to obtain the pollution level index: Where θ>1 is a nonlinear amplification parameter used to enhance the * With Y * Sensitivity when significant differences exist; C3. Finally, the calculated pollution level index is compared with the preset pollution threshold: If the pollution level index is greater than the pollution threshold, it means that after comprehensively considering the multi-scale complexity of the transmission and reflection signals, the pollution level has exceeded the acceptable range; if the pollution level index is less than or equal to the pollution threshold, it means that under the current parameter conditions, the pollution is still within a reasonable range and no cutting parameter adjustment is required.

7. The method for monitoring the operation data of an intelligent laser slotting machine according to claim 3, characterized in that: Step S3 includes the following contents: After confirming that the pollution level index has exceeded the preset pollution threshold, it is necessary to make an initial approximation correction to the cutting head movement speed, so as to achieve a balance between the dynamic regulation of the pollution level and the efficiency requirements; on this basis, by using the comparison result between the pollution level index and the pollution threshold as the input condition, this step develops the following specific processing logic: Firstly, according to the difference between the pollution level index and its preset pollution threshold and the cutting efficiency requirement, a setting strategy for the initial approximation step size is constructed. The correlation between the pollution degree deviation and the efficiency deviation is characterized by nonlinear mapping and multidimensional characteristic functions. Specifically, the magnitude of the pollution index exceeding the pollution threshold is recorded as ΔP, and the deviation between the cutting efficiency requirement and the current efficiency inference value is recorded as ΔE. Perform coupling transformation on ΔP and ΔE to obtain the initial approximation step length δv:

8. The method for monitoring the operation data of an intelligent laser slotting machine according to claim 7, characterized in that: Step S3 also Includes the following: After obtaining the initial approximation step length, set the current cutting head moving speed to v current Update the new speed setting value v according to the following formula new :v new =v current ±δv; The direction of ± depends on the speed change trend required to lower or stabilize the pollution level index; After completing the speed parameter correction, the new speed setting value is fed back to step S2 and step S3, and the optical sensor is used to collect and analyze the transmission and reflection data of the protective cover surface again under the same raw material conditions as required by step S2, and the pollution level index is calculated repeatedly; once the recalculated pollution level index is input into this step, a closed-loop judgment and correction process is formed.

Images