Micro-hole penetration false rejection prevention method based on transient feature extraction and active closed loop verification

Abstract

The application belongs to the technical field of electric spark machining technology, and provides a micro-hole penetration false negative judgment prevention method based on transient feature extraction and active closed-loop verification, which comprises the following steps: first, collecting the machining state signal before EDM punching, and calculating a dynamic baseline; in the process of machining, the state parameters are monitored in real time, the triggering condition is met, the waveform is locked, the penetration confidence is calculated, and the penetration is classified into three categories: confirmed penetration, ambiguous state and false signal; in the ambiguous state, active verification and final judgment are performed through servo exploration and low-energy detection pulse; finally, the corresponding action is executed, the baseline and threshold are updated, and feedback is given, so that closed-loop self-correction is realized. The application can significantly reduce the false negative rate of EDM micro-hole penetration, improve the success rate of deep and small hole penetration confirmation, realize parameter self-adaptation and machining efficiency.

Description

Technical Field

[0001] This application belongs to the field of electrical discharge machining technology, and in particular relates to a method for preventing false judgments of micro-hole penetration based on transient feature extraction and active closed-loop verification. Background Technology

[0002] Electrical discharge machining (EDM) is a core technology for precision micro-hole machining, widely used in aerospace, fuel nozzles, and other fields. Currently, the industry generally uses real-time acquisition of machining status parameters such as electrode feed rate, discharge voltage, or effective discharge percentage, with a preset fixed amplitude threshold as the basis for penetration judgment, to achieve machining start and stop control.

[0003] However, current methods rely on passive discrimination logic based on a single steady-state amplitude, which cannot distinguish between true penetration and chip interference or false state fluctuations caused by secondary discharge. This can easily lead to misjudgment of penetration and over-etching of the workpiece, making it difficult to meet the high-precision machining requirements of deep microholes. Summary of the Invention

[0004] This application provides a method for preventing false judgments of micro-hole penetration based on transient feature extraction and active closed-loop verification. The device, terminal equipment, and storage medium can solve the problems of current methods that rely on passive discrimination logic with a single steady-state amplitude, which cannot distinguish between true penetration and chip interference or false state fluctuations caused by secondary discharge, easily leading to false judgments of penetration and over-etching of the workpiece, and making it difficult to meet the high-precision processing requirements of deep micro-holes.

[0005] In a first aspect, embodiments of this application provide a method for preventing false judgments of micro-hole penetration based on transient feature extraction and active closed-loop verification, comprising: S1, collecting a processing state sampling sequence under a stable state before processing, and calculating a dynamic baseline of the processing state under the current processing conditions based on the sampling sequence; the processing state sampling sequence includes at least one or more combinations of electrode feed rate, effective value of discharge voltage, effective value of discharge current, and percentage of effective discharge pulses; S2, collecting the current processing state signal in real time at a fixed sampling interval during processing, calculating time-domain change feature parameters based on the processing state dynamic baseline and the current processing state signal, the time-domain change feature parameters including at least the instantaneous feed rate change rate and the real-time state deviation; when the instantaneous feed rate change rate and the real-time state deviation meet a preset trigger condition, freezing the state waveform segment before and after the current moment, recording the trigger moment, and jumping to the locked state; S3, in the locked state, based on the preset integration time window, the trigger moment, and the integration time window... The real-time processing status signal inside the mouth and the processing status dynamic baseline are used to calculate the penetration confidence score; the penetration confidence score is compared with the preset judgment criteria to obtain the first-level judgment result; the first-level judgment result includes at least the confirmed penetration state, the ambiguous state, and the pseudo-signal state; S4, when the first-level judgment result is the ambiguous state, the main processing pulse is paused according to the preset timing sequence and active detection verification is performed: first, the servo feed probe action is performed and a low-energy probe pulse is applied, the processing status response signal after the probe action is collected, and the final judgment result is generated according to the comparison result between the probe response signal and the processing status dynamic baseline; wherein, the final judgment result includes confirmed penetration and non-penetration; S5, according to the final judgment result, the corresponding stop discharge tool lifting or compensation discharge pulse output action is performed, and the processing status dynamic baseline and the trigger threshold used in the preset trigger conditions are updated according to the status data of this processing process, and the updated dynamic baseline and trigger threshold are fed back to step S1 for the next processing cycle.

[0006] In one possible implementation of the first aspect, the step S1 above, calculating the dynamic baseline of the processing state under the current processing conditions based on the sampling sequence, specifically includes:

[0007] Calculate the dynamic baseline of the processing state based on the sampling sequence. :

[0008]

[0009] Where N is the number of sampling points; Let be the instantaneous value of the processing state feature quantity at the i-th sampling time.

[0010] Optionally, in another possible implementation of the first aspect, the calculation of time-domain variation characteristic parameters based on the processing state dynamic baseline and the current processing state signal in step S2 above specifically includes:

[0011] Based on the dynamic baseline of the processing state and current processing status signal Calculate the instantaneous rate of change of feed rate :

[0012]

[0013] and real-time state deviation :

[0014]

[0015] in, The fixed sampling interval is t, and the current sampling time is t.

[0016] Optionally, in another possible implementation of the first aspect, the instantaneous feed rate change rate and the real-time state deviation in step S2 above satisfy a preset trigger condition, specifically including:

[0017] Preset rising edge slope threshold and falling edge slope threshold ;

[0018] When the instantaneous feed rate change rate Greater than the rising edge slope threshold or the instantaneous rate of change of feed speed Less than the falling edge slope threshold , or the absolute value of the real-time state deviation When the percentage is greater than 30%, the preset triggering condition is determined to be met.

[0019] Optionally, in another possible implementation of the first aspect, the calculation of the penetration confidence score in step S3 above specifically includes:

[0020] According to the trigger time and the integration time window The integral area of ​​the signal deviation from the dynamic baseline of the processing state within the calculation window is calculated. :

[0021]

[0022] in, For integration variables, This refers to the real-time processing status signal within the integration time window;

[0023] Calculate the signal within the integration time window Normalized variance ;

[0024] According to the state deviation integral area The normalized variance and preset minimum effective penetration area threshold Calculate the penetration confidence score:

[0025]

[0026] in, The penetration confidence score is given.

[0027] Optionally, in another possible implementation of the first aspect, the step S3 above, comparing the penetration confidence score with a preset judgment criterion to obtain a first-level judgment result, specifically includes:

[0028] A first threshold and a second threshold are preset, and the first threshold is greater than the second threshold;

[0029] When the penetration confidence score is greater than or equal to the first threshold, the first-level determination result is a confirmed penetration state;

[0030] When the penetration confidence score is less than the first threshold and greater than or equal to the second threshold, the first-level determination result is in an ambiguous state.

[0031] When the penetration confidence score is less than the second threshold, the first-level determination result is a pseudo-signal state.

[0032] Optionally, in another possible implementation of the first aspect, step S4 above, which involves pausing the main processing pulse according to a preset timing sequence and performing active detection verification, specifically includes:

[0033] The peak current of the low-energy detection pulse is set to a preset percentage of the peak current of the main processing pulse.

[0034] According to the preset detection pulse emission time Calculate the servo feed trial start time :

[0035]

[0036] in, This is the servo feed advance;

[0037] At the time of the servo feed probe start Initiate electrode quantitative servo feed at the time of detection pulse emission. The low-energy detection pulse is applied, and the processing status response signal is acquired after waiting for a preset stabilization time.

[0038] Optionally, in another possible implementation of the first aspect, updating the dynamic baseline of the processing state in step S5 above specifically includes:

[0039] After the current hole is machined, collect M open-circuit state signal values ​​during the machining gap when there is no discharge output and calculate their average value. Where M is the preset quantity and ;

[0040] Based on the preset forgetting factor The dynamic baseline of the current processing state is adaptively updated, and the calculation formula is as follows:

[0041]

[0042] in, This is the current dynamic baseline value for the processing status. This is the updated dynamic baseline value for the processing status.

[0043] Optionally, in another possible implementation of the first aspect, the trigger threshold used in updating the preset trigger condition in step S5 above specifically includes:

[0044] Statistical continuous processing of the most recent The number of times the first-level determination result of each hole is in an ambiguous state. ,in This is the preset total number of statistics;

[0045] Based on adaptive step size coefficient The above and stated Calculate the updated slope trigger threshold :

[0046]

[0047] in, The current slope trigger threshold.

[0048] Secondly, embodiments of this application provide a micro-hole penetration anti-false judgment device based on transient feature extraction and active closed-loop verification, comprising: a signal acquisition module, used to acquire a processing state sampling sequence in a stable state before processing, and to acquire the current processing state signal in real time during processing at a fixed sampling interval; a baseline and feature calculation module, used to calculate a processing state dynamic baseline based on the sampling sequence, and to calculate time-domain change feature parameters based on the current processing state signal and the processing state dynamic baseline; and a state machine control module, used to trigger a locking state based on the time-domain change feature parameters, calculate a penetration confidence score in the locking state, and output a value after comparing the penetration confidence score with a preset judgment criterion. The system generates a first-level judgment result; when the first-level judgment result is in an ambiguous state, it outputs an active verification start command; the active verification execution module receives the active verification start command, pauses the main machining pulse according to a preset timing sequence, executes servo feed probing and applies low-energy detection pulses, collects machining status response signals and feeds them back to the state machine control module so that the state machine control module can generate the final judgment result; the adaptive update and execution module executes stop discharge tool lifting or compensated discharge pulse output according to the final judgment result, updates the machining status dynamic baseline and the trigger threshold used in the preset judgment criteria according to the current machining status data, and feeds back the updated parameters to the baseline and feature calculation module.

[0049] Thirdly, embodiments of this application provide a terminal device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the aforementioned method for preventing false judgments of micro-hole penetration based on transient feature extraction and active loop closure verification.

[0050] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the aforementioned method for preventing false judgments of micro-hole penetration based on transient feature extraction and active loop closure verification.

[0051] Beneficial Effects: This application addresses the pain point of penetration misjudgment in EDM deep micro-hole machining. Through a discrimination architecture combining transient feature extraction and active closed-loop verification, it achieves a comprehensive improvement in penetration detection accuracy and machining stability. This solution upgrades penetration discrimination from the traditional single steady-state amplitude threshold to a multi-dimensional confidence discrimination based on slope-priority triggering, integral energy, and variance weighting. It can capture the instantaneous state change during penetration within a microsecond window, effectively distinguishing true penetration features from false signals caused by random noise and chip fluctuations, significantly reducing the false detection rate and avoiding the workpiece over-etching and hole diameter out-of-tolerance problems easily caused by traditional solutions. Based on this, this solution constructs a three-level state machine discrimination mechanism. For fuzzy state conditions with insufficient confidence, it actively distinguishes true penetration from false deceleration caused by secondary discharge and chip accumulation through servo feed probing and low-energy detection pulse linkage verification. This significantly improves the success rate of one-time penetration confirmation for micro-holes with large aspect ratios, overcoming the fundamental limitation of passive sensing solutions that cannot eliminate physical interference. Meanwhile, this solution can adapt to the drift of working conditions such as electrode wear and working fluid contamination by updating the dynamic baseline forgetting factor and self-tuning the threshold based on the frequency of fuzzy states. It does not require frequent manual calibration, effectively reduces the output of invalid discharge pulses, and has the advantages of high reliability and low loss in batch continuous processing. It has good engineering practical value and industrial promotion prospects. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application, 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a flowchart illustrating a micro-hole penetration prevention method based on transient feature extraction and active loop closure verification provided in an embodiment of this application.

[0054] Figure 2 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application;

[0055] Figure 3 This is a schematic diagram of a micro-hole penetration prevention device based on transient feature extraction and active closed-loop verification provided in an embodiment of this application. Detailed Implementation

[0056] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0057] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0058] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0059] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0060] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0061] The following description, with reference to the accompanying drawings, details a method, apparatus, terminal device, and storage medium for preventing false positives in micro-hole penetration based on transient feature extraction and active closed-loop verification provided in this application.

[0062] It should be noted that, in the embodiments of this application, "micro-hole" refers to the machining of through holes or blind holes with a diameter of millimeters or less, and is particularly suitable for deep micro-hole EDM with a depth-to-diameter ratio greater than 5:1. The "machining status signal" in the embodiments of this application can be one or more combinations of electrode feed rate, effective discharge voltage, effective discharge current, and effective discharge pulse percentage. The specific signal can be flexibly selected according to the machining method and sensor configuration, and this application is not limited to this.

[0063] For example, in an electrical discharge machining (EDM) scenario, the system acquires the electrode feed speed signal through a servo motor encoder, collects the effective values ​​of discharge voltage and current through a voltage / current Hall sensor, and counts the percentage of effective discharge pulses through a discharge pulse counting circuit. When the electrode penetrates the workpiece, the feed speed jumps to a positive or negative direction, and at the same time, the gap voltage rises sharply and the percentage of effective discharge decreases. These characteristics can all be used as the basis for penetration determination.

[0064] Figure 1 The illustration shows a flowchart of a micro-hole penetration prevention method based on transient feature extraction and active loop closure verification provided in an embodiment of this application.

[0065] like Figure 1 As shown, the micropore penetration prevention method based on transient feature extraction and active loop closure verification includes the following steps:

[0066] S1. Collect a processing state sampling sequence under stable conditions before processing, and calculate the processing state dynamic baseline under the current processing conditions based on the sampling sequence; the processing state sampling sequence includes at least one or more combinations of electrode feed rate, effective value of discharge voltage, effective value of discharge current, and percentage of effective discharge pulses.

[0067] Before machining begins, the system performs initialization operations. Specifically, under no-load conditions where the workpiece is clamped, the tool electrode is in place, the working fluid circulation is stable, and no main machining pulse is applied, a set of machining status signal samples are continuously collected by sensors as a characteristic quantity sampling sequence under the current working condition.

[0068] It should be noted that the processing state sampling sequence in the embodiments of this application includes at least one or more combinations of electrode feed speed, effective value of discharge voltage, effective value of discharge current, and percentage of effective discharge pulses. Those skilled in the art can select one or more of these combinations according to the actual sensor configuration and processing requirements, and this application is not limited thereto.

[0069] In the embodiments of this application, the dynamic baseline is not a fixed constant, but an adaptive reference value recalculated for each hole before machining or even for each batch of material conditions. Its purpose is to eliminate the zero-point drift of the signal caused by factors such as different electrode wear, batch differences of workpiece materials, and fluctuations in working fluid temperature or conductivity, so as to provide a reliable reference system for subsequent real-time deviation calculation.

[0070] Furthermore, in this embodiment of the application, the step S1 above, which calculates the dynamic baseline of the processing state under the current processing conditions based on the sampling sequence, specifically includes:

[0071] Calculate the dynamic baseline of the processing state based on the sampling sequence. :

[0072]

[0073] Where N is the number of sampling points; The instantaneous value of the processing state characteristic at the i-th sampling time is denoted as , which can be any normalized characterization value among the encoder reading of electrode feed speed, effective value of discharge voltage, effective value of discharge current, or percentage of effective discharge pulses.

[0074] As one possible implementation, before machining begins (or before machining the first hole of each workpiece), the system continuously collects a preset number of machining status signal samples in a stable state without discharge output. The number of sampling points can be set according to the signal stability requirements and the system's computing power, typically using a sample size sufficient to cover multiple fluctuation cycles or sufficient to characterize the steady-state mean.

[0075] It should be noted that the dynamic baseline is calculated using an arithmetic mean in this embodiment, which has the advantages of being simple to calculate, having good real-time performance, and effectively suppressing the impact of random noise on baseline estimation. For scenarios where there are periodic fluctuations in the signal, the number of sampling points should be sufficient to cover an integer number of fluctuation periods to ensure that the mean is unbiased. This application is not limited to this.

[0076] In this embodiment, when electrode feed rate is used as a characteristic quantity, if the workpiece surface has slight unevenness, the initial feed rate of the electrode will fluctuate slightly. By re-acquiring the baseline value of the hole position before machining each hole, the system can adaptively eliminate the influence of surface condition differences.

[0077] S2. During the processing, the current processing status signal is collected in real time at a fixed sampling interval. The time-domain change characteristic parameters are calculated based on the processing status dynamic baseline and the current processing status signal. The time-domain change characteristic parameters include at least the instantaneous feed rate change rate and the real-time status deviation. When the instantaneous feed rate change rate and the real-time status deviation meet the preset trigger conditions, the state waveform segment before and after the current moment is frozen, the trigger moment is recorded, and the process jumps to the locked state.

[0078] It should be noted that the time-domain variation characteristic parameters in the embodiments of this application include at least the instantaneous feed rate change rate and the real-time state deviation. The instantaneous feed rate change rate characterizes the drastic change of the signal within an extremely short time interval. Its physical meaning lies in capturing the unique signal abrupt change at the moment of penetration—the sudden increase or decrease in feed rate when the electrode exits the workpiece, both manifested as a drastic jump in the signal on a microsecond to millisecond time scale. The real-time state deviation reflects the percentage deviation of the current signal from the baseline in terms of amplitude.

[0079] In this embodiment of the application, when the system detects that the instantaneous feed rate change rate and the real-time state deviation meet the preset trigger conditions, it immediately freezes the state waveform segment before and after the current moment, records the trigger moment, and jumps to the locked state. The locked state means that after the system detects a suspected penetration signal, it does not immediately make a final decision, but enters a transient intermediate state in which the waveform is frozen, so as to gain a time window for subsequent integration discrimination and active verification.

[0080] Furthermore, in this embodiment, the calculation of time-domain variation characteristic parameters based on the processing state dynamic baseline and the current processing state signal in step S2 specifically includes:

[0081] Based on the dynamic baseline of the processing state and current processing status signal Calculate the instantaneous rate of change of feed rate :

[0082]

[0083] and real-time state deviation :

[0084]

[0085] in, The fixed sampling interval is t, and the current sampling time is t.

[0086] It should be noted that the instantaneous rate of change in this embodiment is calculated using the first-order forward differential, and its physical meaning lies in the rate of change of the quantized signal between two adjacent sampling moments. At the moment of penetration, a sudden change in feed velocity or a jump in discharge state both manifest as a sharp increase in the instantaneous rate of change (positive or negative). Therefore, the instantaneous rate of change is one of the most sensitive indicators for capturing the moment of penetration.

[0087] It should be noted that the real-time state deviation in the embodiments of this application represents the degree of deviation of the current signal from the dynamic baseline, expressed as a percentage, and its function is to provide a penetration criterion from the amplitude dimension.

[0088] In this embodiment of the application, during EDM micro-hole processing, the feed rate is relatively stable and the voltage waveform is stable when the hole has not penetrated; at the moment of penetration, the feed rate changes abruptly and the voltage jumps to the open circuit value. The system simultaneously monitors two parameters: the instantaneous rate of change and the real-time deviation. The instantaneous rate of change captures the steepness of the change (i.e., whether the change is "sudden" enough), and the real-time deviation captures the absolute magnitude of the change (i.e., whether the change is "large" enough). The two parameters corroborate and complement each other.

[0089] Furthermore, in this embodiment of the application, the instantaneous feed rate change rate and the real-time state deviation in step S2 above satisfy a preset trigger condition, specifically including:

[0090] Preset rising edge slope threshold and falling edge slope threshold ;

[0091] When the instantaneous feed rate change rate Greater than the rising edge slope threshold or the instantaneous rate of change of feed speed Less than the falling edge slope threshold , or the absolute value of the real-time state deviation When the percentage is greater than 30%, the preset triggering condition is determined to be met.

[0092] It should be noted that the rising edge slope threshold and falling edge slope threshold correspond to the trigger thresholds for positive and negative steep changes in the signal, respectively. In EDM drilling scenarios, the feed rate during penetration may exhibit a positive jump (acceleration due to a sudden load drop) or a negative jump (instant deceleration due to exit burrs or secondary discharge). By setting thresholds in both directions simultaneously, the system can cover penetration characteristics in different changing directions.

[0093] It should be noted that in this embodiment, the instantaneous rate of change is used as the first priority triggering condition, while the amplitude deviation is used as a backup redundancy triggering condition. This design is because the instantaneous rate of change reflects the "suddenness" of the change, which is the most essential physical characteristic of the moment of penetration; while the amplitude deviation reflects the "amplitude" of the change, and can be used as a supplementary criterion in scenarios where the signal changes slowly but the amplitude has deviated significantly from the baseline.

[0094] It should be noted that the 30% amplitude deviation threshold in the embodiments of this application is only an illustrative example. In actual applications, it can be adjusted according to specific materials, processing parameters and signal characteristics. This application is not limited thereto.

[0095] For example, in an EDM drilling scenario, the moment the electrode penetrates the bottom of the workpiece, the forward resistance suddenly disappears, and the feed rate generates a steep rising edge, which the system immediately detects. And trigger a lock. Even if the speed change is relatively gentle in some special cases and the slope threshold is not triggered, the system will still trigger a lock as long as the absolute value of the real-time deviation exceeds 30%, forming a double protection.

[0096] Once locked, the system does not rely on a single instantaneous amplitude or rate of change for judgment. Instead, it performs integral analysis on the signal waveform within a preset integral time window before and after the trigger moment to calculate the penetration confidence score.

[0097] S3. In the locked state, a penetration confidence score is calculated based on a preset integration time window, the trigger time, the real-time processing status signal within the integration time window, and the processing status dynamic baseline; the penetration confidence score is compared with a preset judgment criterion to obtain a first-level judgment result; the first-level judgment result includes at least a confirmed penetration state, an ambiguous state, and a pseudo-signal state;

[0098] It should be noted that the penetration confidence score in this application embodiment is a quantitative index that integrates two dimensions: the cumulative energy of the signal deviating from the baseline and the stability of signal fluctuation. Its core purpose is that even if the signal spike at the moment of penetration disappears quickly due to chip interference, secondary discharge or instantaneous blockage by debris, as long as the energy dissipation characteristics of the spike are significant enough, the system can still give a high confidence penetration judgment.

[0099] In this embodiment, the penetration confidence score is compared with a preset judgment criterion to obtain a first-level judgment result. The first-level judgment result in this embodiment includes at least three states: confirmed penetration state, ambiguous state, and pseudo-signal state. This three-state design breaks through the binary logic of the traditional either-or approach, enabling the system to adopt differentiated processing strategies based on different signal confidence levels.

[0100] Furthermore, in this embodiment of the application, the calculation of the penetration confidence score in step S3 above specifically includes:

[0101] According to the trigger time and the integration time window The integral area of ​​the signal deviation from the dynamic baseline of the processing state within the calculation window is calculated. :

[0102]

[0103] in, For integration variables, This refers to the real-time processing status signal within the integration time window;

[0104] Calculate the signal within the integration time window Normalized variance ;

[0105] According to the state deviation integral area The normalized variance and preset minimum effective penetration area threshold Calculate the penetration confidence score:

[0106]

[0107] in, The penetration confidence score is given.

[0108] It should be noted that after entering the locked state, the system opens a preset integration time window. Within this window, the system does not rely on a single instantaneous amplitude for judgment, but performs integration analysis on the signal waveform within a preset time window before and after the trigger moment.

[0109] It should be noted that the state deviation integral area in this embodiment refers to the cumulative deviation of the signal from the baseline within a preset time window after the trigger moment. Its physical meaning lies in quantifying the total energy dissipated by the signal at the moment of penetration—when penetration occurs, the electrode exits the workpiece, causing drastic changes in the feed rate and electrical parameters. This physical process leaves an integrable signal footprint within the time window. Even if this footprint presents as a brief spike in time, its integral area can still be effectively measured.

[0110] It should be noted that normalized variance is used to measure the severity of signal fluctuations within the integration time window. Its purpose is to suppress the interference of random noise spikes on confidence assessment: if a signal spike is purely caused by noise or occasional interference, the signal before and after it often exhibits irregular and drastic fluctuations with a large variance; while a true penetration signal, although it may also disappear quickly, shows a certain physical continuity in the signal changes before and after it, with a relatively small variance. By using variance as a penalty factor for confidence, true penetration and noise interference can be effectively distinguished.

[0111] It should be noted that, in the embodiments of this application... It is a process parameter related to specific processing conditions (electrode material and size, workpiece material and thickness, discharge parameters, working fluid state, etc.), which can be pre-calibrated through a small number of process experiments and stored in the processing database. Different working conditions correspond to different thresholds, and the system automatically calls the corresponding thresholds according to the process parameters before processing. This application is not limited to this.

[0112] In this embodiment of the application, during EDM machining of deep microholes, the feed rate spike generated at the moment of penetration may rapidly decay due to instantaneous debris blockage, but the integral area corresponding to this spike is sufficient to characterize the signal dissipation during the penetration process. At this time, although the real-time amplitude of the signal drops rapidly after the spike, the integral area value has been effectively recorded and used in the confidence score.

[0113] Furthermore, in this embodiment of the application, the step S3 above, comparing the penetration confidence score with a preset judgment criterion to obtain a first-level judgment result, specifically includes:

[0114] A first threshold and a second threshold are preset, and the first threshold is greater than the second threshold;

[0115] When the penetration confidence score is greater than or equal to the first threshold, the first-level determination result is a confirmed penetration state;

[0116] When the penetration confidence score is less than the first threshold and greater than or equal to the second threshold, the first-level determination result is in an ambiguous state.

[0117] When the penetration confidence score is less than the second threshold, the first-level determination result is a pseudo-signal state.

[0118] It should be noted that the system presets a first threshold and a second threshold, with the first threshold being greater than the second threshold. These two thresholds together constitute a three-interval judgment system.

[0119] when At this point, the first-level judgment result is "confirmed penetration state". At this time, the signal characteristics are clear and the reliability is strong. The system directly judges it as confirmed penetration and stops discharging and raising the knife, without the need for additional verification.

[0120] when At this point, the first-level judgment result is "ambiguous state". At this time, the signal has partial penetration characteristics but is not enough to be certain. The system enters the active verification stage and performs secondary confirmation through servo feed probing and low-energy detection pulses.

[0121] when At this point, the first-level judgment result is "pseudo-signal state". At this time, the signal characteristics obviously do not conform to the physical laws of penetration, and are likely to originate from noise interference or non-penetrating factors (such as debris obstruction, electrode vibration, instantaneous short circuit, etc.). The system judges it as a pseudo-signal, unlocks it, and continues processing.

[0122] It should be noted that the three-state partitioning in this embodiment enables the system to adopt differentiated processing strategies based on different signal confidence levels, avoiding arbitrary decisions in boundary situations by traditional binary logic. The width of the fuzzy state interval determines the frequency with which the system enters the active verification stage—a wider interval means more opportunities for active verification and more cautious judgment, but active verification itself requires time resources; a narrower interval means fewer opportunities for active verification and higher processing efficiency, but may lead to incorrect judgments in boundary situations. This application achieves a balance between judgment accuracy and processing efficiency by reasonably setting two thresholds.

[0123] It should be noted that the specific values ​​of the first threshold and the second threshold need to be pre-calibrated through process experiments based on specific processing conditions (electrode material and size, workpiece material and thickness, discharge parameters, sensor configuration, etc.) and can be stored in the process database for use under different working conditions. This application is not limited to this.

[0124] S4. When the first-level judgment result is ambiguous, the main machining pulse is paused according to a preset timing sequence and active detection verification is performed: first, a servo feed probe action is performed and a low-energy probe pulse is applied, the machining status response signal after the probe action is collected, and the final judgment result is generated based on the comparison result between the probe response signal and the dynamic baseline of the machining status; wherein, the final judgment result includes confirmed penetration and non-penetration;

[0125] When the primary judgment result is ambiguous, the system pauses the main machining pulse according to a preset timing sequence and performs active detection verification. Specifically, a servo feed probe is first executed, and a low-energy probe pulse is applied. The machining status response signal after the probe is applied is collected, and the final judgment result is generated based on the comparison between the probe response signal and the dynamic baseline of the machining status. The final judgment result includes confirmed penetration and non-penetration.

[0126] It should be noted that the active detection closed-loop verification in this embodiment refers to the system actively applying physical intervention and applying a low-energy detection pulse with distinct characteristics when the passive sensing signal is insufficient to make a definite judgment. Secondary judgment criteria are obtained by observing the system's response to this active stimulus. This mechanism upgrades the traditional one-way passive sensing to a closed-loop verification of "sensing-intervention-re-sensing".

[0127] Furthermore, in this embodiment, step S4 above, which involves pausing the main processing pulse according to a preset timing and performing active detection verification, specifically includes:

[0128] The peak current of the low-energy detection pulse is set to a preset percentage of the peak current of the main processing pulse.

[0129] According to the preset detection pulse emission time Calculate the servo feed trial start time :

[0130]

[0131] in, This is the servo feed advance;

[0132] At the time of the servo feed probe start Initiate electrode quantitative servo feed at the time of detection pulse emission. The low-energy detection pulse is applied, and the processing status response signal is acquired after waiting for a preset stabilization time.

[0133] It should be noted that the preset percentage needs to be calibrated comprehensively based on the electrode size, machining gap and sensor sensitivity to ensure that the detection pulse has sufficient resolution to distinguish the gap state (open circuit / normal / short circuit) while not causing significant material erosion to avoid damaging the machined hole wall. This application is not limited to this.

[0134] It should be noted that the aforementioned lead time is used to compensate for the mechanical response delay between the issuance of the command and the actual execution of the servo system (including the acceleration time of the servo motor, the time to eliminate backlash in the lead screw drive, etc.). Its specific value needs to be pre-calibrated for different equipment models to ensure that the electrode has reached the target feed position when the probe pulse is applied.

[0135] It should be noted that the purpose of the preset stabilization time in this embodiment is to avoid the influence of servo feed mechanical oscillation and probe pulse transient effects, ensuring that the acquired response signal truly reflects the stable state of the gap. This stabilization time needs to be set according to the damping characteristics of the servo system and the transient response time of the discharge circuit, and this application is not limited to this.

[0136] For example, at the instant the electrode penetrates the bottom of the workpiece, the resistance in front disappears. If the system is in an ambiguity state at this moment, servo quantitative feed is initiated. If the electrode has already penetrated, it will not touch the bottom material after feeding a certain distance, the gap voltage remains high, and the detection pulse can establish a normal discharge. If the electrode has not penetrated, the electrode tip will touch the bottom of the workpiece after feeding, the gap will be short-circuited, and the detection pulse cannot establish a normal discharge. The system can make a final judgment based on this difference.

[0137] It should be noted that the final decision logic is as follows: if the deviation of the detection response signal from the dynamic baseline is greater than a preset threshold, it is forcibly determined as "confirmed penetration"; otherwise, it is determined as "not penetrated," the system is unlocked, the main processing pulse is restored, and a compensation discharge pulse is automatically added. The number of compensation pulses can be obtained from a process database based on factors such as workpiece thickness and electrode wear; this application is not limited to this.

[0138] S5. Execute the corresponding stop discharge and tool lifting or compensation discharge pulse output action according to the final determination result, and update the dynamic baseline of the processing state and the trigger threshold used in the preset trigger conditions according to the status data of this processing process. Feed back the updated dynamic baseline and trigger threshold to step S1 for use in the next processing cycle.

[0139] It should be noted that the closed-loop feedback in this application embodiment includes two levels of adaptive updates: First, dynamic baseline updates—as processing continues, electrodes wear out, the working fluid becomes increasingly turbid, and the temperature rises. These factors cause the signal baseline to drift slowly. By re-acquiring the open-circuit state signal and updating the baseline with weights during each processing interval, the system can always remain adapted to the current working conditions. Second, adaptive adjustment of the trigger threshold—if the system frequently enters an ambiguity state, it indicates that the current trigger threshold is set too tight or too loose. The background automatically adjusts the threshold based on the frequency of ambiguity state triggers, thereby achieving self-tuning of sensitivity.

[0140] In this embodiment, during continuous batch processing in EDM, as the number of holes processed accumulates, the electrodes gradually wear out, the working fluid becomes increasingly contaminated, and the temperature rises, all of which cause the baseline of signals such as feed rate to drift slowly. If the baseline remains fixed, the calculation of real-time deviation will increasingly deviate from the actual situation, eventually leading to the failure of the judgment criteria. This application updates the baseline and threshold after each hole is processed, enabling the system to follow the slow changes in operating conditions and always maintain accurate judgment capabilities.

[0141] Furthermore, in this embodiment of the application, updating the dynamic baseline of the processing state in step S5 specifically includes:

[0142] After the current hole is machined, collect M open-circuit state signal values ​​during the machining gap when there is no discharge output and calculate their average value. Where M is the preset quantity and ;

[0143] Based on the preset forgetting factor The dynamic baseline of the current processing state is adaptively updated, and the calculation formula is as follows:

[0144]

[0145] in, This is the current dynamic baseline value for the processing status. This is the updated dynamic baseline value for the processing status.

[0146] It should be noted that the forgetting factor in the embodiments of this application... The forgetting factor is a weighting coefficient between 0 and 1, used to control the influence of historical baseline values ​​on the current baseline update result. The closer the forgetting factor is to 1, the greater the weight of historical baseline values, the smoother the update, and the stronger the resistance to sudden interference, but the slower the response to slow drift; the closer the forgetting factor is to 0, the greater the weight of the current measurement value, and the faster the response, but the more susceptible it is to occasional noise. This application achieves a balance between interference resistance and response speed by reasonably setting the forgetting factor, but this application is not limited to this.

[0147] In this embodiment, during continuous batch EDM processing, as the number of holes is accumulated, electrode wear gradually leads to a slow overall change in the feed rate signal. The working fluid also becomes increasingly contaminated, and rising temperatures affect the baseline electrical parameters. This application updates the baseline using a forgetting factor after each hole is processed, ensuring that the baseline follows the slow changes in electrode wear and working fluid conditions, thus maintaining an accurate description of the current processing conditions.

[0148] Furthermore, in this embodiment of the application, the step S5 above, which updates the trigger threshold used in the preset triggering condition, specifically includes:

[0149] Statistical continuous processing of the most recent The number of times the first-level determination result of each hole is in an ambiguous state. ,in This is the preset total number of statistics;

[0150] Based on adaptive step size coefficient The above and stated Calculate the updated slope trigger threshold :

[0151]

[0152] in, The current slope trigger threshold.

[0153] It should be noted that the adaptive step size coefficient in the embodiments of this application... This step size coefficient is used to control the magnitude and speed of threshold adjustment. A larger step size coefficient results in a larger adjustment magnitude each time and a faster system response to changes in operating conditions, but may cause oscillations; a smaller step size coefficient results in a smoother adjustment and a more stable system, but a slower response speed. This application achieves smooth tracking of threshold changes to operating conditions by reasonably setting the step size coefficient.

[0154] It should be noted that if the frequency of fuzzy states is high, the threshold will automatically decrease (become more relaxed) to lower the threshold for entering the fuzzy state and reduce the frequency of subsequent active verification calls. If the frequency of fuzzy states is low (the threshold may be too loose, causing situations that should enter the fuzzy state to be directly judged as confirmations or false signals), then bidirectional adaptation can be achieved by setting a lower threshold or a reverse adjustment mechanism. In practical applications, bidirectional adaptive adjustment of the threshold can be achieved by combining upper and lower limit constraints, but this application is not limited to this.

[0155] In this embodiment, during continuous processing, as electrode wear intensifies, the fluctuation range of the feed rate gradually increases, and the originally set slope trigger threshold may become too tight, leading to frequent entry into the fuzzy state. This application automatically lowers the slope trigger threshold by statistically analyzing the frequency of fuzzy state triggers, enabling the system to adapt to the new operating conditions after electrode wear and maintain stable trigger sensitivity.

[0156] This application provides a method for preventing false judgments of micro-hole penetration based on transient feature extraction and active closed-loop verification. First, the processing status signal before EDM drilling is acquired, and a dynamic baseline is calculated. During processing, status parameters are monitored in real time, and the waveform is locked when trigger conditions are met. The penetration confidence is calculated and categorized into three types: confirmed penetration, ambiguous state, and pseudo-signal. In the ambiguous state, active verification and final judgment are achieved through servo probing and low-energy probe pulses. Finally, corresponding actions are executed to update the baseline and threshold and provide feedback, achieving closed-loop self-correction. This application can significantly reduce the false judgment rate of EDM micro-hole penetration, improve the success rate of deep micro-hole penetration confirmation, and achieve parameter adaptation and processing efficiency improvement.

[0157] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0158] To implement the above embodiments, this application also proposes a terminal device.

[0159] Figure 2 This is a schematic diagram of the structure of a terminal device according to an embodiment of this application.

[0160] like Figure 2 As shown, the terminal device 200 includes:

[0161] The system includes a memory 210 and at least one processor 220, and a bus 230 connecting different components (including the memory 210 and the processor 220). The memory 210 stores a computer program, and when the processor 220 executes the program, it implements the micro-hole penetration anti-false judgment method based on transient feature extraction and active loop closure verification described in the embodiments of this application.

[0162] Bus 230 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0163] Terminal device 200 typically includes various electronically readable media. These media can be any available media that can be accessed by terminal device 200, including volatile and non-volatile media, removable and non-removable media.

[0164] Memory 210 may also include computer system readable media in the form of volatile memory, such as RAM 240 and / or cache 250. Terminal device 200 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 260 may be used to read and write non-removable, non-volatile magnetic media (… Figure 2 Not shown; usually referred to as a "hard drive"). Although Figure 2 As not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 230 via one or more data media interfaces. Memory 210 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application.

[0165] A program / utility 280 having a set (at least one) of program modules 270 may be stored in, for example, memory 210. Such program modules 270 include—but are not limited to—an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 270 typically perform the functions and / or methods described in the embodiments of this application.

[0166] Terminal device 200 can also communicate with one or more external devices 290 (e.g., keyboard, pointing device, display 291, etc.), and with one or more devices that enable a user to interact with terminal device 200, and / or with any device that enables terminal device 200 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 292. Furthermore, terminal device 200 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 293. As shown, network adapter 293 communicates with other modules of terminal device 200 via bus 230. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with terminal device 200, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0167] The processor 220 performs various functional applications and data processing by running programs stored in the memory 210.

[0168] It should be noted that the implementation process and technical principles of the terminal device in this embodiment are explained in the foregoing description of a micro-hole penetration prevention method based on transient feature extraction and active closed-loop verification in this application embodiment, and will not be repeated here.

[0169] Corresponding to the above embodiment, a micropore penetration prevention method based on transient feature extraction and active loop closure verification is provided. Figure 3 The diagram shows a structural block diagram of a micro-hole penetration prevention device based on transient feature extraction and active closed-loop verification provided in an embodiment of this application. For ease of explanation, only the parts related to the embodiment of this application are shown.

[0170] Reference Figure 3 The device 300 includes:

[0171] The signal acquisition module 301 is used to acquire the processing state sampling sequence under the stable state before processing, and to acquire the current processing state signal in real time during the processing process at a fixed sampling interval;

[0172] The baseline and feature calculation module 302 is used to calculate the dynamic baseline of the processing state based on the sampling sequence, and to calculate the time-domain change feature parameters based on the current processing state signal and the dynamic baseline of the processing state.

[0173] The state machine control module 303 is used to trigger a locked state based on the time-domain change characteristic parameters, calculate the penetration confidence score in the locked state, compare the penetration confidence score with a preset judgment criterion, and output a first-level judgment result; when the first-level judgment result is an ambiguous state, it outputs an active verification start command.

[0174] The active verification execution module 304 is used to receive the active verification start command, pause the main machining pulse according to the preset timing sequence, perform servo feed probing and apply low-energy detection pulse, collect machining status response signals and feed them back to the state machine control module so that the state machine control module can generate the final judgment result.

[0175] The adaptive update and execution module 305 is used to execute stop discharge and tool lifting or compensate discharge pulse output according to the final judgment result, and update the dynamic baseline of the processing status and the trigger threshold used in the preset judgment criteria according to the current processing status data, and feed back the updated parameters to the baseline and feature calculation module.

[0176] In practical use, the micro-hole penetration prevention device based on transient feature extraction and active closed-loop verification provided in this application embodiment can be configured in any terminal device to execute the aforementioned micro-hole penetration prevention method based on transient feature extraction and active closed-loop verification.

[0177] This application provides a micro-hole penetration anti-false judgment device based on transient feature extraction and active closed-loop verification. First, it collects the processing status signal before EDM drilling and calculates the dynamic baseline. During processing, it monitors the status parameters in real time, locks the waveform when trigger conditions are met, calculates the penetration confidence and classifies it into three categories: confirmed penetration, ambiguous state, and pseudo-signal. In the ambiguous state, it actively verifies and makes a final judgment through servo probing and low-energy detection pulses. Finally, it executes the corresponding action, updates the baseline and threshold, and provides feedback, achieving closed-loop self-correction. This application can significantly reduce the false judgment rate of EDM micro-hole penetration, improve the success rate of deep micro-hole penetration confirmation, and achieve parameter adaptation and processing efficiency improvement.

[0178] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0179] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0180] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0181] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0182] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some regions, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0183] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0184] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0185] In the embodiments provided in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0186] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0187] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for preventing false positives in micropore penetration based on transient feature extraction and active loop closure verification, characterized in that, Includes the following steps: S1. Collect a processing state sampling sequence under stable conditions before processing, and calculate the processing state dynamic baseline under the current processing conditions based on the sampling sequence; the processing state sampling sequence includes at least one or more combinations of electrode feed rate, effective value of discharge voltage, effective value of discharge current, and percentage of effective discharge pulses. S2. During the processing, the current processing status signal is collected in real time at a fixed sampling interval. The time-domain change characteristic parameters are calculated based on the processing status dynamic baseline and the current processing status signal. The time-domain change characteristic parameters include at least the instantaneous feed rate change rate and the real-time status deviation. When the instantaneous feed rate change rate and the real-time status deviation meet the preset trigger conditions, the state waveform segment before and after the current moment is frozen, the trigger moment is recorded, and the process jumps to the locked state. S3. In the locked state, a penetration confidence score is calculated based on a preset integration time window, the trigger time, the real-time processing status signal within the integration time window, and the processing status dynamic baseline; the penetration confidence score is compared with a preset judgment criterion to obtain a first-level judgment result; the first-level judgment result includes at least a confirmed penetration state, an ambiguous state, and a pseudo-signal state; S4. When the first-level judgment result is ambiguous, the main machining pulse is paused according to a preset timing sequence and active detection verification is performed: first, a servo feed probe action is performed and a low-energy probe pulse is applied, the machining status response signal after the probe action is collected, and the final judgment result is generated based on the comparison result between the probe response signal and the dynamic baseline of the machining status; wherein, the final judgment result includes confirmed penetration and non-penetration; S5. Execute the corresponding stop discharge and tool lifting or compensation discharge pulse output action according to the final determination result, and update the dynamic baseline of the processing state and the trigger threshold used in the preset trigger conditions according to the status data of this processing process. Feed back the updated dynamic baseline and trigger threshold to step S1 for use in the next processing cycle.

2. The method according to claim 1, characterized in that, The step S1, which involves calculating the dynamic baseline of the processing state under the current processing conditions based on the sampling sequence, specifically includes: Calculate the dynamic baseline of the processing state based on the sampling sequence. : ； Where N is the number of sampling points; Let be the instantaneous value of the processing state feature quantity at the i-th sampling time.

3. The method according to claim 2, characterized in that, The step S2, which calculates the time-domain variation characteristic parameters based on the dynamic baseline of the processing state and the current processing state signal, specifically includes: Based on the dynamic baseline of the processing state and current processing status signal Calculate the instantaneous rate of change of feed rate : ； and real-time state deviation : ； in, The fixed sampling interval is t, and the current sampling time is t.

4. The method according to claim 3, characterized in that, The instantaneous feed rate change rate and the real-time state deviation in step S2 satisfy preset trigger conditions, specifically including: Preset rising edge slope threshold and falling edge slope threshold ; When the instantaneous feed rate change rate Greater than the rising edge slope threshold or the instantaneous rate of change of feed speed Less than the falling edge slope threshold , or the absolute value of the real-time state deviation When the percentage is greater than 30%, the preset triggering condition is determined to be met.

5. The method according to claim 4, characterized in that, The calculation of the penetration confidence score in step S3 specifically includes: According to the trigger time and the integration time window The integral area of ​​the signal deviation from the dynamic baseline of the processing state within the calculation window is calculated. : ； in, For integration variables, This refers to the real-time processing status signal within the integration time window; Calculate the signal within the integration time window Normalized variance ; According to the state deviation integral area The normalized variance and preset minimum effective penetration area threshold Calculate the penetration confidence score: ； in, The penetration confidence score is given.

6. The method according to claim 5, characterized in that, Step S3, which compares the penetration confidence score with a preset judgment criterion to obtain a first-level judgment result, specifically includes: A first threshold and a second threshold are preset, and the first threshold is greater than the second threshold; When the penetration confidence score is greater than or equal to the first threshold, the first-level determination result is a confirmed penetration state; When the penetration confidence score is less than the first threshold and greater than or equal to the second threshold, the first-level determination result is in an ambiguous state. When the penetration confidence score is less than the second threshold, the first-level determination result is a pseudo-signal state.

7. The method according to claim 6, characterized in that, The step S4, which involves pausing the main processing pulse according to a preset timing sequence and performing active detection verification, specifically includes: The peak current of the low-energy detection pulse is set to a preset percentage of the peak current of the main processing pulse. According to the preset detection pulse emission time Calculate the servo feed trial start time : ； in, This is the servo feed advance; At the time of the servo feed probe start Initiate electrode quantitative servo feed at the time of detection pulse emission. The low-energy detection pulse is applied, and the processing status response signal is acquired after waiting for a preset stabilization time.

8. The method according to claim 7, characterized in that, Updating the dynamic baseline of the processing state in step S5 specifically includes: After the current hole is machined, collect M open-circuit state signal values ​​during the machining gap when there is no discharge output and calculate their average value. Where M is the preset quantity and ; Based on the preset forgetting factor The dynamic baseline of the current processing state is adaptively updated, and the calculation formula is as follows: ； in, This is the current dynamic baseline value for the processing status. This is the updated dynamic baseline value for the processing status.

9. The method according to claim 8, characterized in that, The step S5 of updating the trigger threshold used in the preset trigger conditions specifically includes: Statistical continuous processing of the most recent The number of times the first-level determination result of each hole is in an ambiguous state. ,in This is the preset total number of statistics; Based on adaptive step size coefficient The above and stated Calculate the updated slope trigger threshold : ； in, The current slope trigger threshold.

10. A micro-hole penetration prevention device based on transient feature extraction and active loop closure verification, applied to the method described in any one of claims 1-9, characterized in that, include: The signal acquisition module is used to acquire the processing status sampling sequence under the stable state before processing, and to acquire the current processing status signal in real time during the processing at a fixed sampling interval; The baseline and feature calculation module is used to calculate the dynamic baseline of the processing state based on the sampling sequence, and to calculate the time-domain change feature parameters based on the current processing state signal and the dynamic baseline of the processing state. The state machine control module is used to trigger a locked state based on the time-domain change characteristic parameters, calculate the penetration confidence score in the locked state, and output a first-level judgment result after comparing the penetration confidence score with a preset judgment criterion. When the first-level determination result is ambiguous, an active verification start command is output. The active verification execution module is used to receive the active verification start command, pause the main machining pulse according to the preset timing sequence, perform servo feed probing and apply low-energy detection pulse, collect machining status response signals and feed them back to the state machine control module so that the state machine control module can generate the final judgment result. The adaptive update and execution module is used to execute stop discharge and tool lifting or compensate discharge pulse output according to the final judgment result, and update the dynamic baseline of the processing status and the trigger threshold used in the preset judgment criteria according to the current processing status data, and feed back the updated parameters to the baseline and feature calculation module.

Images