Bayesian online detection method and device for penetration state of electric spark piercing machining

Abstract

The application relates to a Bayesian online detection method and device for a piercing state of electric spark piercing machining, wherein the method comprises the following steps: dynamically updating a probability statistical model constructed by original signals in an electric spark piercing machining process by calculating an expected value of a posterior probability distribution of a key parameter; modeling and calculating a duration of a current machining stage based on the dynamically updated probability statistical model; quantifying and identifying a current machining stage of the electric spark piercing based on the duration of the machining stage, and online detecting a time point when the machining stage is changed from a blind hole machining stage to a piercing machining stage according to a quantification result, so as to generate an online detection result of a piercing time point. According to the embodiment of the application, a tool electrode feeding speed in a machining process is taken as a characteristic signal, a dynamically updated probability statistical model is established to describe the electric spark piercing machining state, model parameters are identified and quantified, and an instantaneous piercing state is identified, so that self-adaptive detection of the piercing time point with high robustness is realized.

Description

Technical Field

[0001] This application relates to the field of special processing technology, and in particular to a Bayesian online detection method and device for the penetration state of electrical discharge machining. Background Technology

[0002] Suitable for use on difficult-to-machine materials such as titanium alloys and nickel-based superalloys High-speed electrical discharge machining (EDM) technology has been widely applied in numerous industrial fields, including aero-engines, the automotive industry, and precision molds. The EDM process typically consists of three stages: blind hole machining, through-hole machining, and over-hole machining. When using a tubular tool electrode with internal flushing fluid for EDM, the flushing fluid flows out from the outlet after penetrating the bottom of the hole, reducing the machining stability and efficiency of residual workpiece material at the outlet. Because the axial length loss and radial profile loss of the machining electrode are difficult to calculate accurately, a large through-hole machining distance can lead to machining back damage, which can have serious consequences, especially for machining film cooling holes on aero-engine turbine blades.

[0003] At the moment of penetration, the abrupt changes in the working fluid flow field morphology and chip removal state within the machining gap lead to alterations in the discharge state and the servo motion state of the electrode. Existing research has analyzed changes in characteristic signals to determine penetration, including gap voltage, electrode displacement, back pressure of the internal working fluid, pulse duration, peak current, and acoustic signals. Current identification techniques primarily include deterministic models such as threshold judgment, support vector machine classification, neural network classification, and machining state diagram classification. Data-driven deterministic models often rely on the validity of data labels and are suitable for EDM processes with stable machining states and clear penetration signal characteristics.

[0004] However, among the penetration detection methods in related technologies for electrical discharge machining, the more robust penetration detection models are still lacking due to their high computational complexity, difficulty in improving real-time performance, and failure to consider the interference of unstable phenomena such as poor chip removal and electrode short circuits that occur randomly during the machining process on the penetration judgment accuracy. Therefore, there is still a lack of penetration detection methods with both good real-time performance and robustness, which urgently need to be improved. Summary of the Invention

[0005] This application provides a Bayesian online detection method and apparatus for the penetration state of electrical discharge machining (EDM) to address the problem that in related EDM penetration detection methods, robust penetration detection models suffer from high computational complexity, difficulty in improving real-time performance, and do not consider the interference of random unstable phenomena such as poor chip removal and electrode short circuits during machining on the penetration judgment accuracy. Therefore, there is still a lack of a penetration detection method with both good real-time performance and robustness.

[0006] The first aspect of this application provides a Bayesian online detection method for the penetration state of electrical discharge machining (EDM), comprising the following steps: dynamically updating a probabilistic statistical model constructed from the original signals during EDM by calculating the expected value of the posterior probability distribution of key parameters, thereby obtaining a dynamically updated probabilistic statistical model; calculating the duration of the current machining stage based on the dynamically updated probabilistic statistical model; quantifying and identifying the current machining stage of EDM based on the duration of the machining stage, thereby obtaining a quantification result; and detecting online the moment when the machining stage changes from the blind hole machining stage to the machining penetration stage based on the quantification result, thereby generating an online detection result of the machining penetration moment.

[0007] Optionally, in one embodiment of this application, the raw signal includes at least one or more combinations of the average voltage signal of the machining gap, the feed rate signal of the tool electrode servo axis, and the internal flushing fluid pressure signal.

[0008] Optionally, in one embodiment of this application, the step of dynamically updating the probabilistic statistical model constructed from the original signals during the electrical discharge machining process by calculating the expected value of the posterior probability distribution of the key parameters includes: calculating the posterior probability distribution of the key parameters using Bayesian inference based on the original signals during the electrical discharge machining process; and iteratively updating the posterior probability distribution using model hyperparameters to calculate the expected value of the posterior probability distribution.

[0009] Optionally, in one embodiment of this application, the modeling calculation of the duration of the processing stage includes: based on the expected value of the posterior probability distribution, when the key parameter is detected to meet a preset stability condition, the duration of the processing stage is increased sequentially from zero to generate the increased duration of the processing stage; when the key parameter is detected to meet a preset change condition, the duration of the processing stage is reset to zero to generate a new duration of the processing stage.

[0010] Optionally, in one embodiment of this application, after generating the online detection result of the machining penetration time, the method further includes: based on the detection result of the machining penetration time, controlling the electrode position to feed downwards by an over-penetration distance that meets preset conditions; and adjusting the exit of the electrical discharge drilling according to the over-penetration distance to generate the final electrical discharge drilling machining result.

[0011] A second aspect of this application provides a Bayesian online detection device for the penetration state of electrical discharge machining (EDM), comprising: an update module, configured to dynamically update a probabilistic statistical model constructed from the original signals during EDM by calculating the expected value of the posterior probability distribution of key parameters, thereby obtaining a dynamically updated probabilistic statistical model; a calculation module, configured to model and calculate the duration of the current machining stage based on the dynamically updated probabilistic statistical model; and a detection module, configured to quantify and identify the current machining stage of EDM based on the duration of the machining stage, thereby obtaining a quantification result, and online detect the moment when the machining stage changes from the blind hole machining stage to the machining penetration stage based on the quantification result, thereby generating an online detection result of the machining penetration moment.

[0012] Optionally, in one embodiment of this application, the raw signal includes at least one or more combinations of the average voltage signal of the machining gap, the feed rate signal of the tool electrode servo axis, and the internal flushing fluid pressure signal.

[0013] Optionally, in one embodiment of this application, the updating module includes: a calculation unit, used to calculate the posterior probability distribution of the key parameters using Bayesian inference based on the original signals during the electrical discharge machining process; and an updating unit, used to iteratively update the posterior probability distribution using model hyperparameters to calculate the expected value of the posterior probability distribution.

[0014] Optionally, in one embodiment of this application, the calculation module includes: an increasing unit, configured to, based on the expected value of the posterior probability distribution, gradually increase the duration of the processing stage from zero when the key parameter is detected to meet a preset stability condition, so as to generate an increased duration of the processing stage; and a generating unit, configured to, when the key parameter is detected to meet a preset change condition, reset the duration of the processing stage to zero, so as to generate a new duration of the processing stage.

[0015] Optionally, in one embodiment of this application, it further includes: a control module, configured to, after generating the online detection result of the processing penetration time, control the electrode position to feed downwards by an over-penetration distance that meets preset conditions based on the detection result of the processing penetration time; and a correction module, configured to adjust the exit of the electrical discharge drilling according to the over-penetration distance to generate the final electrical discharge drilling processing result.

[0016] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the program to implement the Bayesian online detection method for the penetration state of electrical discharge machining as described in the above embodiments.

[0017] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the Bayesian online detection method for the penetration state of electrical discharge machining as described above.

[0018] A fifth aspect of this application provides a computer program product that stores a computer program that, when executed by a processor, implements the Bayesian online detection method for the penetration state of electrical discharge machining as described above.

[0019] This application embodiment uses the tool electrode feed rate during the sampling process as a feature signal to establish a dynamically updated probabilistic statistical model to describe the EDM drilling state. By identifying the quantified model parameters and recognizing the instantaneous penetration state, it achieves highly robust adaptive detection of the penetration moment. This solves the problems in related EDM drilling penetration detection methods, such as the high computational complexity of robust penetration detection models, difficulty in improving real-time performance, and the lack of consideration for the interference of random unstable phenomena during processing, such as poor chip removal and electrode short circuits, on the penetration judgment accuracy. Ultimately, it addresses the lack of a penetration detection method with both excellent real-time performance and robustness.

[0020] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0021] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0022] Figure 1 This is a flowchart of a Bayesian online detection method for the penetration state of electrical discharge machining according to an embodiment of this application;

[0023] Figure 2 This is a diagram illustrating the update process of a feed rate statistical model according to an embodiment of this application;

[0024] Figure 3 This is a graph showing the variation of the duration of a processing stage under different processing stages according to an embodiment of this application;

[0025] Figure 4 This is a flowchart of an electrical discharge machining process using a Bayesian online penetration detection method according to an embodiment of this application;

[0026] Figure 5 This is an example diagram illustrating the detection and identification of a sudden change in processing state and the moment of penetration according to an embodiment of this application;

[0027] Figure 6 This is a schematic diagram of a Bayesian online detection device for the penetration state of electrical discharge machining according to an embodiment of this application;

[0028] Figure 7 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0029] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0030] The following describes a Bayesian online detection method and apparatus for the penetration state of electrical discharge machining (EDM) according to embodiments of this application, with reference to the accompanying drawings. Regarding the EDM penetration detection methods mentioned in the background art, robust penetration detection models suffer from high computational complexity, difficulty in improving real-time performance, and do not consider the interference of random instability phenomena such as poor chip removal and electrode short circuits during machining on the penetration judgment accuracy. Therefore, a penetration detection method with both excellent real-time performance and robustness is still lacking. This application provides a Bayesian online detection method for the penetration state of EDM. In this method, the tool electrode feed rate during machining is sampled as a feature signal to establish a dynamically updated probabilistic statistical model to describe the EDM state. By identifying the quantified model parameters and recognizing the instantaneous penetration state, highly robust adaptive detection of the penetration moment is achieved. Therefore, this solves the problems in the penetration detection methods of electrical discharge machining in related technologies, such as the high computational complexity of the penetration detection model, the difficulty in improving real-time performance, and the fact that it does not take into account the interference of unstable phenomena such as poor chip removal and electrode short circuits that occur randomly during the machining process on the penetration judgment accuracy. There is still a lack of penetration detection methods with both good real-time performance and robustness.

[0031] Specifically, Figure 1 This is a schematic flowchart of a Bayesian online detection method for the penetration state of electrical discharge machining provided in an embodiment of this application.

[0032] like Figure 1 As shown, the Bayesian online detection method for the penetration state of electrical discharge machining includes the following steps:

[0033] In step S101, the expected value of the posterior probability distribution of key parameters is calculated, and the probabilistic statistical model constructed from the original signals during the electrical discharge machining process is dynamically updated to obtain the dynamically updated probabilistic statistical model.

[0034] It is understood that the probabilistic statistical model in the embodiments of this application includes a probabilistic statistical model in which the feed rate v follows a t-distribution.

[0035] In actual implementation, the probability statistical model following the t-distribution in the embodiments of this application is as follows:

[0036]

[0037] Where τ is the degree of freedom of the t-distribution; the mean μ n This reflects the machining rate and electrode wear rate of blind holes via electrical discharge machining (EDM), and its magnitude is related to the machining material, machining electrical specifications, and servo control parameters; variance This reflects the speed fluctuations, the control process, and the noise in the signal sampling.

[0038] The model parameters in equation (1) are updated in real time during the processing to establish a statistical model applicable to each processing condition. As the speed signal v is continuously sampled, the model parameters (μ) are calculated successively. n and The statistical model is adaptively adjusted based on the posterior probability distribution of the given information.

[0039]

[0040] Wherein, the left side represents the posterior probability density of the distribution to be updated, and the first term on the right side represents the feed rate v obtained from the new observation. t The likelihood probability in the current statistical model is calculated and simplified using the probability density function of the t-distribution as follows:

[0041]

[0042] Furthermore, such as Figure 2 As shown, the dynamic update in this embodiment includes hyperparameter updates {α0,β0,γ0} based on Bayesian inference and model parameter updates, specifically using a normal-inverse gamma distribution as the mean μ. n and variance The initial prior distribution is:

[0043]

[0044]

[0045] Among them, the hyperparameter γ0 reflects the sensitivity of the mean μ n The confidence level of the estimate; α0, β0 are the variances, respectively. Shape and scale parameters of the distribution.

[0046] Therefore, the probability density of the joint distribution in equation (2) can be calculated from the probability density functions of the normal distribution and the gamma distribution:

[0047]

[0048] Substituting equations (3) and (4) into equation (2), the posterior probability distribution of the feed rate statistical model parameters is calculated as shown in equation (5):

[0049]

[0050] This allows for the observation of a new feed rate v. t Subsequently, through iterative updates of hyperparameters {α,β,γ}, the expected value of the normal-inverse gamma distribution is used as the adjusted statistical model parameter to complete the update of the feed rate statistical model parameters.

[0051] This application embodiment can use a probabilistic statistical model to handle random instability during processing, improving the robustness of penetration detection. Simultaneously, it eliminates the need for threshold setting, effectively avoiding false negatives and negatives caused by threshold-based detection. By sampling the tool electrode feed rate during processing as a feature signal, a dynamically updated probabilistic statistical model is established to describe the EDM drilling state. By identifying quantified model parameters and instantaneous penetration states, highly robust adaptive detection of penetration moments is achieved.

[0052] Optionally, in one embodiment of this application, the raw signal includes at least one or more combinations of the average voltage signal of the machining gap, the feed rate signal of the tool electrode servo axis, and the internal flushing fluid pressure signal.

[0053] In this embodiment, the original signal is one or more combinations of the average voltage signal of the machining gap, the feed speed signal of the tool electrode servo axis, and the internal flushing fluid pressure signal. The established probability and statistical model is one of Gaussian distribution, inverse Gaussian distribution, and t-distribution.

[0054] Optionally, in one embodiment of this application, the probabilistic statistical model constructed from the original signals during the electrical discharge machining process is dynamically updated by calculating the expected value of the posterior probability distribution of the key parameters, including: calculating the posterior probability distribution of the key parameters using Bayesian inference based on the original signals during the electrical discharge machining process; and iteratively updating the posterior probability distribution using model hyperparameters to calculate the expected value of the posterior probability distribution.

[0055] In actual implementation, this embodiment can calculate the posterior probability distribution of key parameters based on Bayesian inference using the original signal obtained from the current sampling, and then use the expected value of the posterior probability distribution as the updated value of the key parameters. To improve computational efficiency, the posterior probability distribution of the key parameters is controlled by the model hyperparameters, which can be directly iteratively updated, avoiding the complex model inference solution process.

[0056] In step S102, the duration of the current processing stage is calculated based on a dynamically updated probabilistic statistical model.

[0057] It is understandable that, such as Figure 3 As shown, the modeling calculations in this embodiment include modeling calculations based on the Bayesian online change point detection algorithm.

[0058] In actual implementation, the embodiments of this application can use a dynamically updated probabilistic statistical model to update and calculate the duration of the processing stage during the acquisition cycle of the raw signal processing, using a Bayesian online change point detection algorithm.

[0059] After observing the new feed rate v t+1 After that, r t+1 joint probability distribution The joint probability distribution at the previous time step The product of the predicted probability from the feed rate statistical model and the prior probability of machining time:

[0060]

[0061] Where, r t This refers to the duration of the machining phase. During the hole machining phase, as the feed rate statistical model is continuously updated, r... t A linear increase, with brief periods of instability causing r to... t A brief decrease, but after it disappears, it does not affect r. t Subsequent growth.

[0062] Optionally, in one embodiment of this application, modeling and calculating the duration of the processing stage includes: based on the expected value of the posterior probability distribution, when it is detected that the key parameter meets the preset stability condition, the duration of the processing stage is increased from zero to generate the increased duration of the processing stage; when it is detected that the key parameter meets the preset change condition, the duration of the processing stage is reset to zero to generate a new duration of the processing stage.

[0063] It is understood that the duration of the processing stage in this application embodiment can be: the number of times the original signal is sampled from the beginning of processing or from the moment of the last change in processing state.

[0064] As one possible implementation method, embodiments of this application can be based on the expected value of the posterior probability distribution. When the key parameters remain stable during the processing stage, the duration of the processing stage starts from 0 and increases gradually. When the key parameters change significantly, the duration of the processing stage drops to 0 instantaneously to generate a new duration of the processing stage.

[0065] It should be noted that the preset stability conditions and preset change conditions can be set by those skilled in the art according to the actual situation.

[0066] In step S103, based on the duration of the processing stage, the current processing stage of the electrical discharge drilling is quantitatively identified to obtain a quantitative result. Based on the quantitative result, the moment when the processing stage changes from the blind hole processing stage to the processing penetration stage is detected online to generate an online detection result of the processing penetration moment.

[0067] It is understood that the processing stages in the embodiments of this application specifically include: blind hole processing stage, through-hole processing stage and over-through-hole processing stage.

[0068] In actual implementation, the embodiments of this application can quantitatively identify the current processing stage of electrical discharge drilling based on the duration of the processing stage, and perform online identification of the processing penetration moment, that is, online detection of the moment when the processing stage changes from the blind hole processing stage to the processing penetration stage, so as to generate online detection results of the processing penetration moment, thereby achieving highly robust adaptive detection of the penetration moment.

[0069] The embodiments of this application do not require threshold judgment, which can effectively avoid the missed detection and false detection of penetration caused by threshold judgment, and has significant advantages compared with existing penetration detection methods.

[0070] Optionally, in one embodiment of this application, after generating the detection result of the machining penetration time, the method further includes: based on the detection result of the machining penetration time, controlling the electrode position to feed downwards by an overpenetration distance that meets preset conditions; and adjusting the exit of the electrical discharge drilling according to the overpenetration distance to generate the final electrical discharge drilling machining result.

[0071] In actual implementation, such as Figure 4 As shown, the embodiment of this application includes a processing stage with a feed rate signal, lasting for a duration r. t The calculation and update of the feed rate statistical model. When the duration r of the penetration machining stage is detected. t When the detection length is greater than the recognition length h, that is, when h+1>r is detected... tWhen the time is ≥h, it is determined that a machining penetration phenomenon has occurred. After identifying and recording the penetration time, the control electrode position continues to feed downwards for a certain distance beyond the penetration distance to perform hole exit trimming to generate the final EDM drilling result.

[0072] The embodiments of this application can effectively reduce misjudgments caused by instability during normal processing, have high accuracy, and are suitable for perforation detection under different workpiece tilt angles.

[0073] In the implementation example, a larger identification length h during the penetration processing stage helps to eliminate false penetration judgments caused by transient instability, thus improving the robustness of penetration judgment. Conversely, a smaller identification length h during the penetration processing stage results in lower detection latency and stronger real-time performance. In this embodiment, the identification length during the penetration processing stage is set to 3 times the sampling period (h = 3) to capture the penetration moment. Figure 5 As shown. Figure 5 b is the step size r t The probability distribution diagram shows that, compared to the stable blind hole machining stage, the step size r... t The probability distribution band is wider during the penetration and over-penetration processing stages. Calculate p(r) t The probability distribution profile of (=h) is shown in the figure. Figure 5 As shown in Figure c, the first peak of the penetration probability occurs during the initial stage of processing when the step size increases, which can be eliminated by setting a processing time of no less than a certain duration. Brief processing instability will not cause an increase in the penetration probability. The second peak of the penetration probability occurs during the penetration processing stage, thus enabling the capture of the penetration moment.

[0074] It should be noted that the preset conditions can be set by those skilled in the art according to the actual situation, and no specific restrictions are imposed here.

[0075] The Bayesian online detection method for the penetration state of electrical discharge machining (EDM) proposed in this application can use the tool electrode feed rate during machining as a feature signal to establish a dynamically updated probabilistic statistical model to describe the EDM machining state. By identifying the quantified model parameters and recognizing the instantaneous penetration state, highly robust adaptive detection of the penetration moment is achieved. This solves the problem that in related EDM penetration detection methods, more robust penetration detection models suffer from high computational complexity, making it difficult to improve real-time performance. Furthermore, these methods do not consider the interference of random unstable phenomena such as poor chip removal and electrode short circuits during machining on the penetration judgment accuracy, and still lack a penetration detection method with both excellent real-time performance and robustness.

[0076] Next, referring to the accompanying drawings, a Bayesian online detection device for the penetration state of electrical discharge machining according to an embodiment of this application is described.

[0077] Figure 6This is a schematic diagram of the Bayesian online detection device for the penetration state of electrical discharge machining according to an embodiment of this application.

[0078] like Figure 6 As shown, the Bayesian online detection device 10 for the penetration state of electrical discharge machining includes: an update module 100, a calculation module 200, and a detection module 300.

[0079] Specifically, the update module 100 is used to dynamically update the probabilistic statistical model constructed from the original signals during the electrical discharge machining process by calculating the expected value of the posterior probability distribution of key parameters, thereby obtaining a dynamically updated probabilistic statistical model.

[0080] The calculation module 200 is used to model and calculate the duration of the current processing stage based on a dynamically updated probabilistic statistical model.

[0081] The detection module 300 is used to quantitatively identify the current processing stage of the electrical discharge drilling based on the duration of the processing stage, so as to obtain the quantitative result, and to detect the moment when the processing stage changes from the blind hole processing stage to the processing penetration stage online based on the quantitative result, so as to generate the online detection result of the processing penetration moment.

[0082] Optionally, in one embodiment of this application, the raw signal includes at least one or more combinations of the average voltage signal of the machining gap, the feed rate signal of the tool electrode servo axis, and the internal flushing fluid pressure signal.

[0083] Optionally, in one embodiment of this application, the update module 100 includes a calculation unit and an update unit.

[0084] The computing unit is used to calculate the posterior probability distribution of key parameters based on the original signals during the electrical discharge machining process using Bayesian inference.

[0085] The update unit is used to iteratively update the posterior probability distribution using model hyperparameters in order to calculate the expected value of the posterior probability distribution.

[0086] Optionally, in one embodiment of this application, the calculation module 200 includes an addition unit and a generation unit.

[0087] The addition unit is used to increase the duration of the processing stage from zero sequentially based on the expected value of the posterior probability distribution, when the key parameters are detected to meet the preset stability conditions, so as to generate the duration of the increased processing stage.

[0088] The generation unit is used to reset the duration of the processing stage to zero when it detects that the key parameters meet the preset change conditions, so as to generate a new duration of the processing stage.

[0089] Optionally, in one embodiment of this application, the Bayesian online detection device 10 for the penetration state of electrical discharge machining further includes a control module and a correction module.

[0090] The control module is used to control the electrode position to advance downwards by a predetermined overpenetration distance based on the online detection result of the processing penetration moment after generating the online detection result of the processing penetration moment.

[0091] The correction module is used to adjust the exit of the EDM (Electrical Discharge Machining) based on the over-penetration distance to generate the final EDM machining result.

[0092] It should be noted that the explanation of the aforementioned embodiment of the Bayesian online detection method for the penetration state of electrical discharge machining also applies to the Bayesian online detection device for the penetration state of electrical discharge machining in this embodiment, and will not be repeated here.

[0093] The Bayesian online detection device for the penetration state of electrical discharge machining (EDM) proposed in this application can use the tool electrode feed rate during machining as a feature signal to establish a dynamically updated probabilistic statistical model to describe the EDM machining state. By identifying the quantified model parameters and recognizing the instantaneous penetration state, it achieves highly robust adaptive detection of the penetration moment. This solves the problem in related EDM penetration detection methods where robust penetration detection models suffer from high computational complexity, difficulty in improving real-time performance, and a lack of consideration for the interference of random unstable phenomena such as poor chip removal and electrode short circuits during machining on the penetration judgment accuracy. Ultimately, a penetration detection method with both excellent real-time performance and robustness is still lacking.

[0094] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include:

[0095] The memory 701, the processor 702, and the computer program stored on the memory 701 and capable of running on the processor 702.

[0096] When the processor 702 executes the program, it implements the Bayesian online detection method for the penetration state of electrical discharge machining provided in the above embodiments.

[0097] Furthermore, electronic devices also include:

[0098] Communication interface 703 is used for communication between memory 701 and processor 702.

[0099] The memory 701 is used to store computer programs that can run on the processor 702.

[0100] The memory 701 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0101] If the memory 701, processor 702, and communication interface 703 are implemented independently, then the communication interface 703, memory 701, and processor 702 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0102] Optionally, in a specific implementation, if the memory 701, processor 702, and communication interface 703 are integrated on a single chip, then the memory 701, processor 702, and communication interface 703 can communicate with each other through an internal interface.

[0103] The processor 702 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0104] This embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the Bayesian online detection method for the penetration state of electrical discharge machining as described above.

[0105] This application also provides a computer program product storing a computer program that, when executed by a processor, implements the Bayesian online detection method for the penetration state of electrical discharge machining as described above.

[0106] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0107] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0108] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0109] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0110] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0111] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0112] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0113] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A Bayesian online detection method for the penetration state of electrical discharge machining, characterized in that, Includes the following steps: By calculating the expected value of the posterior probability distribution of key parameters, the probabilistic statistical model constructed from the original signals during the electrical discharge machining process is dynamically updated, resulting in a dynamically updated probabilistic statistical model. Based on the dynamically updated probabilistic statistical model, the duration of the current processing stage is modeled and calculated; Based on the duration of the processing stage, the current processing stage of the electrical discharge drilling is quantitatively identified to obtain a quantitative result. Based on the quantitative result, the moment when the processing stage changes from the blind hole processing stage to the processing penetration stage is detected online to generate an online detection result of the processing penetration moment. The step of dynamically updating the probabilistic statistical model constructed from the original signals during the electrical discharge machining process by calculating the expected value of the posterior probability distribution of the key parameters includes: calculating the posterior probability distribution of the key parameters using Bayesian inference based on the original signals during the electrical discharge machining process; and iteratively updating the posterior probability distribution using model hyperparameters to calculate the expected value of the posterior probability distribution. The modeling calculation of the duration of the processing stage includes: based on the expected value of the posterior probability distribution, when the key parameter is detected to meet the preset stability condition, the duration of the processing stage is increased from zero to generate the increased duration of the processing stage; when the key parameter is detected to meet the preset change condition, the duration of the processing stage is reset to zero to generate a new duration of the processing stage. After generating the online detection result of the machining penetration moment, the method further includes: based on the online detection result of the machining penetration moment, controlling the electrode position to feed downwards by an over-penetration distance that meets preset conditions; and adjusting the exit of the electrical discharge drilling according to the over-penetration distance to generate the final electrical discharge drilling machining result.

2. The method according to claim 1, characterized in that, The raw signals include at least one or more combinations of the average voltage signal of the machining gap, the feed rate signal of the tool electrode servo axis, and the internal flushing fluid pressure signal.

3. A Bayesian online detection device for the penetration state of electrical discharge machining, characterized in that, The Bayesian online detection method for the penetration state of electrical discharge machining as described in any one of claims 1-2 includes: The update module is used to dynamically update the probabilistic statistical model constructed from the original signals during the electrical discharge machining process by calculating the expected value of the posterior probability distribution of key parameters, thus obtaining a dynamically updated probabilistic statistical model. The calculation module is used to model and calculate the duration of the current processing stage based on the dynamically updated probabilistic statistical model. The detection module is used to quantitatively identify the current processing stage of the electrical discharge drilling based on the duration of the processing stage, so as to obtain a quantitative result, and to detect online the moment when the processing stage changes from the blind hole processing stage to the processing penetration stage based on the quantitative result, so as to generate an online detection result of the processing penetration moment. The update module includes: a calculation unit, used to calculate the posterior probability distribution of the key parameters based on the original signals during the electrical discharge machining process using Bayesian inference; and an update unit, used to iteratively update the posterior probability distribution using model hyperparameters to calculate the expected value of the posterior probability distribution. The calculation module includes: an increment unit, used to increment the duration of the processing stage from zero sequentially based on the expected value of the posterior probability distribution, when the key parameter is detected to meet a preset stability condition, so as to generate the increased duration of the processing stage; and a generation unit, used to reset the duration of the processing stage to zero when the key parameter is detected to meet a preset change condition, so as to generate a new duration of the processing stage. It also includes: a control module, used to control the electrode position to feed downwards by an over-penetration distance that meets preset conditions based on the online detection result of the processing penetration time after generating the online detection result of the processing penetration time; and a correction module, used to adjust the exit of the electrical discharge drilling according to the over-penetration distance to generate the final electrical discharge drilling processing result.

4. The apparatus according to claim 3, characterized in that, The raw signals include at least one or more combinations of the average voltage signal of the machining gap, the feed rate signal of the tool electrode servo axis, and the internal flushing fluid pressure signal.

5. An electronic device, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the Bayesian online detection method for the penetration state of electrical discharge machining as described in any one of claims 1-2.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the Bayesian online detection method for the penetration state of electrical discharge machining as described in any one of claims 1-2.

7. A computer program product, comprising a computer program, characterized in that, The computer program is executed to implement the Bayesian online detection method for the penetration state of electrical discharge machining as described in any one of claims 1-2.

Images