A controller telemetry-based parameter security heuristics system and method

Abstract

The application discloses a parameter security trial system and method based on controller telemetry, relates to the technical field of numerical control machining process monitoring and numerical control parameter debugging, and comprises the following steps: extracting a machining fault flow, and acquiring controller telemetry configuration parameters; calling a workpiece machining record of a numerical control machine tool, and extracting an effective cutting period in a machining process; calling a machining fault record, performing time sequence matching and machining process feature comparison, and determining a target window; mapping an exception to a machining program context, and generating a history correlation score of different configuration parameters; calculating a reasonable degree of a current adjustment parameter; determining whether to trigger an abnormal early warning prompt or to execute a parameter machining trial step, and performing cyclic iteration verification and parameter fine adjustment process until machining quality reaches a standard or a termination condition is triggered. The application can judge whether the selected adjustment parameter is available, can solve the problems of blindly adjusting irrelevant control parameters and lengthening a trial cutting debugging cycle, and effectively improves numerical control machining process debugging efficiency and batch production stability.

Description

Technical Field

[0001] This invention relates to the field of CNC machining process monitoring and CNC parameter debugging technology, specifically a parameter safety testing system and method based on controller telemetry. Background Technology

[0002] When CNC machining produces quality defects such as vibration marks, corner burn-out, or overcutting / undercutting, the existing troubleshooting methods rely on the experience of machine operators to diagnose the faults and modify motion control parameters. However, these methods lack quantitative analysis of historical fault data, leading to subjective judgments and a tendency to blindly adjust irrelevant control parameters. This prolongs the trial cutting and debugging cycle, hinders the rapid elimination of existing quality defects, and can even trigger new machining problems. Therefore, a verification mechanism is urgently needed to quantitatively verify the reliability and rationality of the proposed parameter adjustments. Based on machine tool constraints and machining conditions, the mechanism should determine whether the parameters meet the trial conditions. Under the premise of ensuring machining safety, controlled parameter trial cutting should be carried out to achieve accurate attribution of quality defects and iterative optimization of parameters, effectively improving the efficiency of CNC machining process debugging and the stability of mass production. Summary of the Invention

[0003] The purpose of this invention is to provide a parameter safety testing system and method based on controller telemetry to solve the problems raised in the prior art.

[0004] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0005] A parameter safety probing method based on controller telemetry includes the following steps:

[0006] Extract the current processing fault process and obtain the corresponding controller telemetry configuration parameters;

[0007] In this scheme, timestamp refers to the various time-series markers generated according to a fixed sampling time interval. The controller telemetry relies on the CNC+PLC controller of the CNC machine tool system as the core data source. It can remotely read the original operation measurement data inside the machine tool in real time through industrial protocols and transmit it to the upper system for parsing and analysis.

[0008] Retrieve historical workpiece machining records from CNC machine tools, collect real-time positions of cutting tools and workpiece reference positioning during machining, filter and extract effective cutting time periods during machining, and archive workpiece machining records as qualified machining data that ensure the product quality of the workpiece meets the process specifications.

[0009] Identify the current fault type, extract multiple historical processing fault records with the same fault type, adopt a unified window segmentation rule, divide the effective cutting period and the fault operation process in the processing fault record into multiple time windows, and perform time sequence matching and processing process feature comparison on the corresponding time windows to determine the target window in the fault operation process.

[0010] Based on the target window, the exception is mapped to the processing context, the configuration parameters used in the exception processing are determined, and historical correlation scores for different configuration parameters are generated.

[0011] Obtain the currently selected adjustment parameters, calculate the rationality of the currently selected adjustment parameters based on historical correlation scores, preset a lightweight safety gate, and make a judgment based on the rationality and the lightweight safety gate, choose to trigger an abnormal warning prompt or execute parameter processing trial steps according to the adjustment parameters, set processing trial termination conditions, and iteratively verify and fine-tune the parameter process until the processing quality meets the standard or the termination condition is triggered.

[0012] Furthermore, the effective cutting time periods during the processing are screened and extracted, including:

[0013] A certain time period P is extracted from the workpiece processing record. The time period P is split into several timestamps according to a fixed sampling interval. An industrial camera is deployed at the processing station to collect on-site processing images of each timestamp in real time. The ROI region and target region are divided in each frame of the processing image. The ROI region is the cutting area of ​​the workpiece, and the target region is the cutting tool area that is in direct contact with the cutting area.

[0014] Extract several points on the ROI region at a certain timestamp T0, and calculate the average point value P. 00 Extract several points in the target area at timestamp T0, and calculate the average point value P. 01 Point P 00 Point P 01 The vector is used as the point vector V0 of the timestamp T0;

[0015] Obtain the point vectors V0 and V1 for two adjacent timestamps T0 and T1 respectively. Combine the timestamps T0 and T1 as a time combination and calculate the vector difference modulus of this time combination: |V0-V1|.

[0016] The average value of the vector difference modulus of several time combinations is calculated. If the average value is greater than a preset threshold, the time period P is determined to be an effective cutting time period.

[0017] The vector differential value here represents the relative positional offset of the ROI region relative to the target region within two adjacent timestamps during machining. Since the machine tool needs to make a significant displacement and attitude deflection in the relative position of the two regions during machining to indicate that an effective machining action of actual cutting feed is being performed, the time period with a larger offset value is taken as the effective cutting time period in the scheme.

[0018] Furthermore, the target window in the fault operation process is determined, including:

[0019] The critical moments in the fault operation process and the effective cutting period are extracted respectively. The critical moment is the machining stage switching time corresponding to the cutting tool. The time period between two adjacent critical moments is taken as a time window to obtain several time windows.

[0020] Based on the same processing stage, the working conditions are matched. A time window H1 is extracted from the fault operation process, and a time window H2 matching the working condition H1 is extracted from the effective cutting period. The actual machining trajectory curves S1 and S2 corresponding to the cutting tool in the time windows H1 and H2 are extracted respectively.

[0021] Matching analysis is performed on the points on the processing trajectory curves S1 and S2 to determine whether the time window H1 is the target window.

[0022] Furthermore, determining whether time window H1 is the target window includes:

[0023] Collect feature points on the workpiece surface. These feature points are the points on the workpiece that were not involved in the cutting process after the workpiece was clamped.

[0024] Time windows H1 and H2 are time-aligned. Several synchronous detection times corresponding to the two time windows are selected. The points of trajectory curve S1 at each detection time are extracted, and the spatial distance from each point to the feature point is calculated. Based on the frequency distribution of all spatial distance values, a first distance histogram is constructed. Similarly, the points of trajectory curve S2 at each detection time are extracted, and a second distance histogram is constructed. The cosine similarity between the first distance histogram and the second distance histogram is calculated as the processing similarity between time windows H1 and H2. If the processing similarity is less than a preset threshold, time window H1 is determined as the target window.

[0025] The cosine similarity between the first and second distance histograms represents the trajectory similarity of the two feature curves during the processing. The smaller the value, the less similar the trajectories are, and the more likely the time window when the faulty operation process fails is within this window. Therefore, this time window is used as the target window. Subsequently, statistical analysis is carried out based on the configuration parameters that have changed before and after the target window, and the historical correlation score of each parameter is calculated. This can accurately pinpoint the fault-related parameters and improve the effectiveness of fault root cause localization and parameter optimization.

[0026] Furthermore, generate historical correlation scores for different configuration parameters, including:

[0027] Set the initial impact value of each configuration parameter to 0;

[0028] Extract the target window K0 from the processing fault record R, and the previous time window K1 of the target window K0. If the time window K1 is not the target window, and only the configuration parameter S is changed when the time window K1 turns to the target window K0, then increment the influence value of the configuration parameter S by 1.

[0029] Multiple target windows are extracted from multiple processing fault records. The influence values ​​of configuration parameters are accumulated and statistically analyzed for each record according to the aforementioned rules to obtain the final influence value of each configuration parameter. The final influence value of each configuration parameter is then normalized to obtain the historical correlation scores of different configuration parameters.

[0030] The normalization formula maps the final impact value to 0 to 1, meaning the final historical correlation score ranges from 0 to 1. The larger the final impact value, the closer the historical correlation score is to 1.

[0031] Furthermore, the reasonableness of the currently selected adjustment parameters is calculated, including:

[0032] First, set the target value for each adjustment parameter to 1;

[0033] According to the rule that the larger the target value, the smaller the adjustment weight, the adjustment weight of each adjustment parameter is set, where the sum of the adjustment weights is 1.

[0034] Obtain the historical correlation score and adjustment weight of each currently selected adjustment parameter, multiply the historical correlation score and adjustment weight of each adjustment parameter, sum them up, and calculate the reasonableness of the currently selected adjustment parameter.

[0035] Furthermore, the process involves iterative verification and parameter fine-tuning until the processing quality meets the standards or the termination condition is triggered, including:

[0036] If the reasonableness level is less than the preset threshold, it is determined that there are irrelevant control parameters in the adjustment parameters, triggering an abnormality warning and prompting personnel to readjust the parameters;

[0037] If the reasonableness is not less than the preset threshold and the selected configuration parameters meet the lightweight safety gate verification conditions, the maximum number of iterations is set as the processing test termination condition. If the processing test termination condition is not met, the parameter processing test steps are executed according to the selected configuration parameters. If the processing quality meets the standard after the processing test, the set of parameters is determined to be reasonable and usable. The processing quality meeting the standard means that all key dimensions of the workpiece fall within the preset reasonable tolerance range after processing. During the test, the controller alarm status, shaft load and following error are continuously monitored. When there is a controller alarm or any indicator of shaft load and following error exceeds the preset safety threshold, the test is terminated and an abnormal warning is triggered.

[0038] If the processing quality still fails to meet the standard after the processing trial, the target values ​​of all adjustment parameters are recalculated based on the processing trial results. The adjustment weights of the adjustment parameters and the rationality of the currently selected adjustment parameters are also recalculated. The judgment result is then re-output. The process of iteratively verifying and fine-tuning parameters is repeated until the processing quality meets the standard or the termination condition is triggered.

[0039] Furthermore, recalculate the target values ​​for all adjusted parameters, including:

[0040] If the machining quality still fails to meet the standard after the machining trial, the machining trial process is divided into several time windows. The actual machining trajectory curves of the cutting tool in the time window of the machining trial process and in the time window of the effective cutting period are extracted respectively, and the machining similarity between the corresponding time windows is calculated.

[0041] If, in two adjacent time windows during the processing trial, only the value of a certain adjustment parameter B changes, the target value of the adjustment parameter B is increased by |G1-G2| based on the processing similarity G1 corresponding to the previous time window and the processing similarity G2 corresponding to the subsequent time window; the final target value of all adjustment parameters is then calculated.

[0042] The larger the target value, the greater the fluctuation in trajectory matching caused by the change in adjustment parameter B, indicating that the adjustment parameter B has a greater impact on the processing. Therefore, the adjustment weight of the adjustment parameter B will be smaller. This is consistent with the rule mentioned above that the larger the target value, the smaller the adjustment weight. In this scheme, a smaller weight is assigned to the high-sensitivity parameter, which can enhance the ability of the rationality to identify low-sensitivity parameters and potentially irrelevant parameters.

[0043] Furthermore, the verification conditions for lightweight safety doors include: the adjustment parameter belongs to the preset set of testable parameters, the adjustment parameter value is within the preset allowable adjustment range, the single adjustment range of the parameter does not exceed the preset range limit, and the controller is in the parameter modification allowed state.

[0044] A parameter security probing system based on controller telemetry includes a memory and a processor, as well as a computer program stored in the memory and running on the processor. The processor is coupled to the memory, and the processor implements the aforementioned parameter security probing method based on controller telemetry when executing the computer program.

[0045] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention provides a parameter safety testing system and method based on controller telemetry, including: extracting the machining fault process and obtaining controller telemetry configuration parameters; retrieving CNC machine tool workpiece machining records and extracting the effective cutting time periods during the machining process; retrieving machining fault records, performing time-series matching and machining process feature comparison, and determining the target window; mapping the anomaly to the machining program context and generating historical correlation scores for different configuration parameters; calculating the rationality of the currently adjusted parameters; deciding whether to trigger an anomaly warning prompt or execute parameter machining testing steps, iteratively verifying and fine-tuning the parameter process until the machining quality meets the standards or the termination condition is triggered. This invention determines whether the selected adjustment parameters are usable, solving the problem of blindly adjusting irrelevant control parameters and lengthening the trial cutting and debugging cycle, effectively improving the debugging efficiency of CNC machining processes and the stability of batch production. Attached Figure Description

[0046] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a flowchart illustrating a parameter safety probing method based on controller telemetry according to the present invention.

[0048] Figure 2 This is a flowchart illustrating the implementation process of parameter safety testing for the present invention. Detailed Implementation

[0049] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0050] Example: Figure 1 As shown, this invention provides a technical solution for a parameter safety probing method based on controller telemetry, comprising the following steps:

[0051] Extract the current processing fault process and obtain the corresponding controller telemetry configuration parameters;

[0052] Retrieve historical workpiece machining records from CNC machine tools, collect the real-time position of the cutting tool and the workpiece reference positioning position during the machining process, filter and extract the effective cutting time periods during the machining process, and archive the workpiece machining records as qualified machining data that ensure the product quality of the machined workpiece meets the process specifications.

[0053] Screening and extracting effective cutting time periods during the machining process, including:

[0054] A certain time period P is extracted from the workpiece processing record. The time period P is split into several timestamps according to a fixed sampling interval. An industrial camera is deployed at the processing station to collect on-site processing images of each timestamp in real time. The ROI region and target region are divided in each frame of the processing image. The ROI region is the cutting area of ​​the workpiece, and the target region is the cutting tool area that is in direct contact with the cutting area.

[0055] Extract several points on the ROI region at a certain timestamp T0, and calculate the average point value P. 00 Extract several points in the target area at timestamp T0, and calculate the average point value P. 01 Point P 00 Point P 01 The vector is used as the point vector V0 of the timestamp T0;

[0056] Obtain the point vectors V0 and V1 for two adjacent timestamps T0 and T1 respectively. Combine the timestamps T0 and T1 as a time combination and calculate the vector difference modulus of this time combination: |V0-V1|.

[0057] The average value of the vector difference modulus of several time combinations is calculated. If the average value is greater than a preset threshold, the time period P is determined to be an effective cutting time period.

[0058] Identify the current fault type, extract multiple historical processing fault records with the same fault type, and use a unified window segmentation rule to divide the effective cutting period and the fault operation process in the processing fault record into multiple time windows. Perform time sequence matching and processing process feature comparison on the corresponding time windows to determine the target window in the fault operation process.

[0059] In this embodiment, the current fault types include various types, such as high-speed small line segment vibration, height tracking instability, and five-axis RTCP. Different fault types are usually affected by different configuration parameters. Therefore, it is necessary to extract multiple historical machining fault records with the same fault type to more accurately realize the parameter probing process of this solution.

[0060] Identify the target window in the fault handling process, including:

[0061] The critical moments in the fault operation process and the effective cutting period are extracted respectively. The critical moment is the machining stage switching time corresponding to the cutting tool. The time period between two adjacent critical moments is taken as a time window to obtain several time windows.

[0062] Based on the same processing stage, the working conditions are matched. A time window H1 is extracted from the fault operation process, and a time window H2 matching the working condition H1 is extracted during the effective cutting period. The actual machining trajectory curves S1 and S2 corresponding to the cutting tool in the time windows H1 and H2 are extracted respectively.

[0063] Matching analysis is performed on the points on the processing trajectory curves S1 and S2 to determine whether the time window H1 is the target window. Specifically:

[0064] Collect feature points on the workpiece surface. These feature points are the points on the workpiece that have not been involved in the cutting and removal process throughout the entire clamping process.

[0065] Time windows H1 and H2 are time-aligned. Several synchronous detection times corresponding to the two time windows are selected. The points of trajectory curve S1 at each detection time are extracted, and the spatial distance from each point to the feature point is calculated. Based on the frequency distribution of all spatial distance values, a first distance histogram is constructed. Similarly, the points of trajectory curve S2 at each detection time are extracted, and the spatial distance from each point to the feature point is calculated. Based on the frequency distribution of all spatial distance values, a second distance histogram is constructed. The cosine similarity between the first distance histogram and the second distance histogram is calculated as the processing similarity between time windows H1 and H2. If the processing similarity is less than a preset threshold, time window H1 is determined as the target window.

[0066] In this embodiment, the rules for establishing the first distance histogram are as follows:

[0067] First, set several distance intervals, such as 0-1, 1-2, 2-3, etc., in cm. Then, based on the spatial distance between the feature point and each point on the trajectory curve S1, obtain the number of points that fall into each distance interval. Then, the probability of each spatial distance falling into each distance interval can be obtained. Based on the probability of each distance interval, establish a first distance histogram that shows the probability changing with the distance interval. Similarly, the second distance histogram can also be established in the same way, which will not be elaborated here.

[0068] Based on the target window, the exception is mapped to the processing context, the configuration parameters used in the exception processing are determined, and historical correlation scores for different configuration parameters are generated.

[0069] Set the initial impact value of each configuration parameter to 0;

[0070] Extract the target window K0 from the processing fault record R, and the previous time window K1 of the target window K0. If the time window K1 is not the target window, and only the configuration parameter S is changed when the time window K1 turns to the target window K0, then increment the influence value of the configuration parameter S by 1.

[0071] Multiple target windows are extracted from multiple processing fault records. The influence values ​​of configuration parameters are accumulated and statistically analyzed for each record according to the aforementioned rules to obtain the final influence value of each configuration parameter. The final influence value of each configuration parameter is then normalized to obtain the historical correlation scores of different configuration parameters.

[0072] In this embodiment, the normalization formula is y = 1 - e -x Where y is the historical correlation score, e is the natural constant, and x is the final impact value. The formula is a function of y ranging from 0 to 1 when x is greater than 0, and y increases as x increases.

[0073] Obtain the currently selected adjustment parameters and calculate the reasonableness of the currently selected adjustment parameters based on historical correlation scores.

[0074] Calculate the reasonableness of the currently selected adjustment parameters, including:

[0075] First, set the target value for each adjustment parameter to 1;

[0076] According to the rule that the larger the target value, the smaller the adjustment weight, the adjustment weight of each adjustment parameter is set, where the sum of the adjustment weights is 1.

[0077] In this embodiment, an adjustment weight calculation method is provided: first, calculate the reciprocal of each target value, then add the reciprocals together to get the total value, and then divide each reciprocal by the total value to obtain the adjustment weight of each adjustment parameter. This method also ensures that the sum of the adjustment weights of all adjustment parameters is 1, and that the value of all adjustment weights is between 0 and 1. For example, if there are 4 adjustment parameters, when calculating the adjustment weight of each adjustment parameter for the first time, the target value of each adjustment parameter is 1, and the total target value is 4, then the adjustment weight of each adjustment parameter is 0.25.

[0078] It should be noted that a larger historical correlation score for an adjustment parameter indicates a stronger correlation between that parameter and the current fault, thus increasing the likelihood that the parameter is a cause of the fault. Therefore, a larger historical correlation score generally indicates a more reasonable adjustment parameter. Conversely, a smaller result in the following calculation of reasonableness suggests a greater likelihood of irrelevant control parameters being included in the selected adjustment weights. In the formula for calculating reasonableness, the adjustment weights are used to weight and scale the historical correlation scores. This is because a larger target value indicates a higher proportion of the adjustment parameter's influence on the fault outcome, thus requiring a weighted reduction in the historical correlation score. Conversely, a smaller target value indicates a higher probability that the adjustment parameter is an irrelevant control parameter, requiring a weighted amplification in the historical correlation score. This ensures a more reliable reasonableness calculation. Therefore, in this scheme, the adjustment weights for each parameter should be set according to the rule that a larger target value corresponds to a smaller adjustment weight.

[0079] Obtain the historical correlation score and adjustment weight of each currently selected adjustment parameter, multiply the historical correlation score and adjustment weight of each adjustment parameter, sum them up, and calculate the reasonableness of the currently selected adjustment parameter.

[0080] It should be noted that this solution assesses the rationality of the selected adjustment parameters, but does not analyze the specific values ​​of the adjustment parameters. The formula for calculating the rationality is as follows: Where M is the total number of adjustment parameters, Q m W is the historical correlation score of the m-th adjusted parameter. m It is the adjustment weight of the m-th adjustment parameter. The purpose of calculating the reasonableness here is to determine whether there are any irrelevant control parameters among the adjustment parameters.

[0081] It should be noted that since the historical association score is a normalized value, the historical association score ranges from 0 to 1, and the adjustment weight also ranges from 0 to 1. Furthermore, the sum of the adjustment weights of all adjustment parameters is 1, so the reasonableness Y also ranges from 0 to 1.

[0082] To facilitate understanding, a comparison is provided here between manually selected adjustment parameters and the actual parameters that affect the processing, representing examples of cases with and without irrelevant control parameters, as shown below:

[0083] Assume that the number of adjustment parameters currently selected manually is M=4, and the parameters are a, b, c, and d;

[0084] When the parameters actually affected by the processing are a, b, and c, this falls under the case where there is an irrelevant control parameter d. Assume the historical correlation scores for each adjusted parameter are Q... a=0.8, Q b =0.8, Q c =0.95, Q d =0.15, W in the initial calculation a =0.25, W b =0.25, W c =0.25, W d =0.25, and the calculated reasonableness Y=0.675. Since only the target values ​​of parameters a, b, and c will increase after multiple iterations, the final adjustment weight of parameter d will be greater than the adjustment weights of a, b, and c, for example, W. a =0.15, W b =0.12, W c =0.2, W d =0.53, and the final calculated reasonableness Y=0.4855.

[0085] When the parameters actually affected by the processing are a, b, c, d, e, and f, this falls under the case where there are no irrelevant control parameters. Assume the historical correlation scores for each adjusted parameter are Q... a =0.8, Q b =0.8, Q c =0.95, Q d =0.95, W in the initial calculation a =0.25, W b =0.25, W c =0.25, W d =0.25, and the calculated reasonableness Y=0.875. Since the target value increases after multiple iterations, the target values ​​of the adjustment parameters a, b, c, and d will be relatively similar. Therefore, the final adjustment weights will not be significantly affected by the reasonableness calculation result. Even if the final selected adjustment parameters are a, c, e, and f, since the historical correlation scores of each adjustment parameter are relatively large, even if there is a certain difference between the adjustment weights, it will not have a significant impact on the reasonableness calculation result. The reasonableness will still be at a relatively large value. However, as long as irrelevant control parameters are mixed into the adjustment parameters, the reasonableness will decrease. Therefore, this scheme can ensure that the parameter reasonableness judgment result is robust.

[0086] A lightweight safety gate is preset. Based on the reasonableness of the situation and the lightweight safety gate, a judgment is made to choose whether to trigger an abnormal warning prompt or to execute a parameter processing trial step according to the adjusted parameters. A processing trial termination condition is set, and the process of iteratively verifying and fine-tuning parameters is repeated until the processing quality meets the standard or the termination condition is triggered. The specific steps are as follows:

[0087] If the reasonableness level is less than the preset threshold, it is determined that there are irrelevant control parameters in the adjustment parameters, triggering an abnormality warning and prompting personnel to readjust the parameters; if the reasonableness level is low, it means that the adjustment parameters currently selected by the staff are not reasonable and contain irrelevant control parameters.

[0088] If the reasonableness is not less than the preset threshold and the selected configuration parameters meet the lightweight safety gate verification conditions, the maximum number of iterations is set as the processing test termination condition. If the processing test termination condition is not met, the parameter processing test steps are executed according to the selected configuration parameters. If the processing quality meets the standard after the processing test, the set of parameters is determined to be reasonable and usable. The processing quality meeting the standard means that all key dimensions of the workpiece fall within the preset reasonable tolerance range after processing. During the test, the controller alarm status, shaft load and following error are continuously monitored. When there is a controller alarm or any indicator of shaft load and following error exceeds the preset safety threshold, the test is terminated and an abnormal warning is triggered.

[0089] Controller alarms include various faults and anomalies generated by the controller itself, such as hardware fault alarms, communication anomaly alarms, parameter limit exceeding alarms, and drive fault alarms. Axis load refers to the real-time load index of the servo motion axis, including the servo motor output load rate and motor torque percentage. If the load is consistently too high, it means mechanical jamming or excessive resistance, which can easily cause damage to mechanical parts. Operation should be stopped immediately. Follow-up error refers to the real-time deviation between the servo system command position and the actual feedback position of the motor. If the follow-up error is too large, it means that the actual movement of the axis cannot keep up with the movement command. When an anomaly is present, operation should be stopped immediately and an anomaly warning should be triggered to ensure the safety of the machining process.

[0090] The verification conditions for lightweight safety doors include: the adjustment parameter belongs to the preset set of testable parameters, the adjustment parameter value is within the preset allowable adjustment range, the single adjustment range of the parameter does not exceed the preset range limit, and the controller is in the parameter modification allowed state.

[0091] If the machining quality still fails to meet the standard after the machining trial, the machining trial process is divided into several time windows. The actual machining trajectory curves of the cutting tool in the time window of the machining trial process and in the time window of the effective cutting period are extracted respectively, and the machining similarity between the corresponding time windows is calculated.

[0092] If, in two adjacent time windows during the processing trial, only the value of a certain adjustment parameter B changes, the target value of the adjustment parameter B is increased by |G1-G2| based on the processing similarity G1 corresponding to the previous time window and the processing similarity G2 corresponding to the subsequent time window; the final target value of all adjustment parameters is then calculated.

[0093] It should be noted that the adjustment parameter B for the change in the target value here refers to the parameter that affects the current fault. For example, if the specific value of adjustment parameter B in the previous time window is a reasonable value, but the specific value in the subsequent time window is an unreasonable value, and at this time the processing similarity G1 is greater than the processing similarity G2, then because the processing similarity has changed, it proves that the change in adjustment parameter B has affected the processing process. Therefore, the target value of adjustment parameter B should be increased. Conversely, if the specific value of adjustment parameter B in the previous time window is an unreasonable value, but the specific value in the subsequent time window is a reasonable value, and at this time the processing similarity G1 is less than the processing similarity G2, the same principle applies. This indicates that the change in adjustment parameter B has affected the processing process. However, regardless of the situation, it shows that the adjustment parameter B does indeed affect the overall processing process. Therefore, its target value should be increased to ensure the rationality of the parameter calculation.

[0094] The larger the target value, the greater the fluctuation in trajectory matching caused by the change in adjustment parameter B, indicating that the adjustment parameter B has a greater impact on the processing. Therefore, the adjustment weight of adjustment parameter B will also be greater. This is consistent with the rule mentioned above that the larger the target value, the greater the adjustment weight, and the adjustment weight of each adjustment parameter is set accordingly.

[0095] Recalculate the target values ​​of all adjustment parameters, and recalculate the adjustment weights of the adjustment parameters and the rationality of the currently selected adjustment parameters. Then, re-output the judgment results and iterate through the process of verification and parameter fine-tuning until the processing quality meets the standards or the termination condition is triggered.

[0096] This embodiment also provides a parameter security probing system based on controller telemetry, including a memory and a processor, and a computer program stored in the memory and running on the processor. The processor is coupled to the memory, and the processor implements the aforementioned parameter security probing method based on controller telemetry when executing the computer program. Since this parameter security probing method based on controller telemetry has been described in detail above, it will not be repeated here.

[0097] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0098] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0099] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A parameter safety probing method based on controller telemetry, characterized in that, Includes the following steps: Extract the current processing fault process and obtain the corresponding controller telemetry configuration parameters; Retrieve historical workpiece machining records from CNC machine tools, collect real-time positions of cutting tools and workpiece reference positioning positions during machining, and filter and extract effective cutting time periods during machining. The workpiece machining records are qualified machining archive data that ensure the product quality of the workpiece meets the process specifications. Identify the current fault type, extract multiple historical processing fault records with the same fault type, adopt a unified window segmentation rule, divide the effective cutting period and the fault operation process in the processing fault record into multiple time windows, and perform time sequence matching and processing process feature comparison on the corresponding time windows to determine the target window in the fault operation process. Based on the target window, the exception is mapped to the processing context, the configuration parameters used in the exception processing are determined, and historical correlation scores for different configuration parameters are generated. Obtain the currently selected adjustment parameters, and calculate the reasonableness of the currently selected adjustment parameters based on historical correlation scores; A lightweight safety gate is preset. Based on the reasonableness of the situation and the lightweight safety gate, a judgment is made to select whether to trigger an abnormal warning prompt or to execute a parameter processing test step according to the adjusted parameters. The processing test termination condition is set, and the process of iterative verification and parameter fine-tuning is repeated until the processing quality meets the standard or the termination condition is triggered.

2. The parameter safety probing method based on controller telemetry according to claim 1, characterized in that, Screening and extracting effective cutting time periods during the machining process, including: A certain time period P is extracted from the workpiece processing record. The time period P is split into several timestamps according to a fixed sampling interval. An industrial camera is deployed at the processing station to collect on-site processing images of each timestamp in real time. The ROI region and target region are divided in each frame of the processing image. The ROI region is the cutting area of ​​the workpiece, and the target region is the cutting tool area that is in direct contact with the cutting area. Extract several points on the ROI region at a certain timestamp T0, and calculate the average point value P. 00 Extract several points in the target area at timestamp T0, and calculate the average point value P. 01 Point P 00 Point P 01 The vector is used as the point vector V0 of the timestamp T0; Obtain the point vectors V0 and V1 for two adjacent timestamps T0 and T1 respectively. Combine the timestamps T0 and T1 as a time combination and calculate the vector difference modulus of this time combination: |V0-V1|. The average value of the vector difference modulus of several time combinations is calculated. If the average value is greater than a preset threshold, the time period P is determined to be an effective cutting time period.

3. The parameter safety probing method based on controller telemetry according to claim 2, characterized in that, Identify the target window in the fault handling process, including: The key moments in the fault operation process and the effective cutting period are extracted respectively. The key moments are the machining stage switching moments corresponding to the cutting tool. The time period between two adjacent key moments is taken as a time window to obtain several time windows. Based on the same processing stage, the working conditions are matched. A time window H1 is extracted from the fault operation process, and a time window H2 matching the working condition H1 is extracted from the effective cutting period. The actual machining trajectory curves S1 and S2 corresponding to the cutting tool in the time windows H1 and H2 are extracted respectively. Matching analysis is performed on the points on the processing trajectory curves S1 and S2 to determine whether the time window H1 is the target window.

4. The parameter safety probing method based on controller telemetry according to claim 3, characterized in that, Determining whether time window H1 is the target window includes: Collect feature points on the surface of the workpiece. These feature points are those points that have not been involved in the cutting and removal process throughout the entire process after the workpiece is clamped. Time windows H1 and H2 are time-aligned. Several synchronous detection times corresponding to the two time windows are selected. The points of trajectory curve S1 at each detection time are extracted, and the spatial distance from each point to the feature point is calculated. Based on the frequency distribution of all spatial distance values, a first distance histogram is constructed. Similarly, the points of trajectory curve S2 at each detection time are extracted, and a second distance histogram is constructed. The cosine similarity between the first distance histogram and the second distance histogram is calculated as the processing similarity between time windows H1 and H2. If the processing similarity is less than a preset threshold, time window H1 is determined as the target window.

5. The parameter safety probing method based on controller telemetry according to claim 1, characterized in that, Generate historical correlation scores for different configuration parameters, including: Set the initial impact value of each configuration parameter to 0; Extract the target window K0 from the processing fault record R, and the previous time window K1 of the target window K0. If the time window K1 is not the target window, and only the configuration parameter S is changed when the time window K1 turns to the target window K0, then increment the influence value of the configuration parameter S by 1. Multiple target windows are extracted from multiple processing fault records. The influence values ​​of configuration parameters are accumulated and statistically analyzed for each record according to the aforementioned rules to obtain the final influence value of each configuration parameter. The final influence value of each configuration parameter is then normalized to obtain the historical correlation scores of different configuration parameters.

6. The parameter safety probing method based on controller telemetry according to claim 4, characterized in that, Calculate the reasonableness of the currently selected adjustment parameters, including: First, set the target value for each adjustment parameter to 1; According to the rule that the larger the target value, the smaller the adjustment weight, the adjustment weight of each adjustment parameter is set, where the sum of the adjustment weights is 1. Obtain the historical correlation score and adjustment weight of each currently selected adjustment parameter, multiply the historical correlation score and adjustment weight of each adjustment parameter, sum them up, and calculate the reasonableness of the currently selected adjustment parameter.

7. The parameter safety probing method based on controller telemetry according to claim 6, characterized in that, The process involves iterative verification and parameter fine-tuning until the processing quality meets the standards or a termination condition is triggered, including: If the reasonableness level is less than the preset threshold, it is determined that there are irrelevant control parameters in the adjustment parameters, triggering an abnormality warning and prompting personnel to readjust the parameters; If the reasonableness is not less than the preset threshold and the selected configuration parameters meet the lightweight safety gate verification conditions, the maximum number of iterations is set as the processing test termination condition. If the processing test termination condition is not met, the parameter processing test steps are executed according to the selected configuration parameters. If the processing quality meets the standard after the processing test, the set of parameters is determined to be reasonable and usable. The processing quality meeting the standard means that all key dimensions of the workpiece fall within the preset reasonable tolerance range after processing. During the test, the controller alarm status, shaft load and following error are continuously monitored. When there is a controller alarm or any indicator of shaft load and following error exceeds the preset safety threshold, the test is terminated and an abnormal warning is triggered. If the processing quality still fails to meet the standard after the processing trial, the target values ​​of all adjustment parameters are recalculated based on the processing trial results. The adjustment weights of the adjustment parameters and the rationality of the currently selected adjustment parameters are also recalculated. The judgment result is then re-output. The process of iteratively verifying and fine-tuning parameters is repeated until the processing quality meets the standard or the termination condition is triggered.

8. The parameter safety probing method based on controller telemetry according to claim 7, characterized in that, Recalculate the target values ​​for all adjusted parameters, including: If the machining quality still fails to meet the standard after the machining trial, the machining trial process is divided into several time windows. The actual machining trajectory curves of the cutting tool in the time window of the machining trial process and in the time window of the effective cutting period are extracted respectively, and the machining similarity between the corresponding time windows is calculated. If, in two adjacent time windows during the processing trial, only the value of a certain adjustment parameter B changes, the target value of the adjustment parameter B is increased by |G1-G2| based on the processing similarity G1 corresponding to the previous time window and the processing similarity G2 corresponding to the subsequent time window; the final target value of all adjustment parameters is then calculated.

9. A parameter safety probing method based on controller telemetry according to claim 1, characterized in that, The verification conditions for the lightweight safety door include: the adjustment parameter belongs to a preset set of testable parameters, the adjustment parameter value is within a preset allowable adjustment range, the single adjustment range of the parameter does not exceed a preset range limit, and the controller is in a parameter modification-allowed state.

10. A parameter safety testing system based on controller telemetry, characterized in that, It includes a memory and a processor, as well as a computer program stored in the memory and running on the processor. The processor is coupled to the memory, and when the processor executes the computer program, it implements a parameter security probing method based on controller telemetry as described in any one of claims 1-9.

Images