An adaptive compensation method and system for electrode wire tension in slow wire EDM

Abstract

This application provides an adaptive compensation method and system for electrode wire tension in slow wire EDM, relating to the field of slow wire EDM technology. The method includes: based on the initial diameter and material of the electrode wire, and constrained by workpiece and cutting parameters, retrieving a historical sample set of slow wire EDM to obtain the target tension of the electrode wire; inputting the target tension into a tension control table and configuring first tension control timing parameters; performing wire diameter loss analysis to obtain wire diameter loss timing information; performing wire tension analysis based on the wire diameter loss timing information and the first tension control timing parameters to obtain wire tension timing information; and performing tension compensation optimization using the wire tension timing information and the target electrode wire tension to obtain second tension control parameter timing information, and then performing electrode wire tension compensation control. This solves the technical problem of poor electrode wire tension control accuracy in existing slow wire EDM processes.

Description

Technical Field

[0001] This invention relates to the field of wire EDM, and more particularly to an adaptive compensation method and system for electrode wire tension in wire EDM. Background Technology

[0002] In slow wire EDM, electrode wire tension is a key parameter affecting machining accuracy, surface quality, and electrode wire life. However, traditional electrode wire tension control often uses fixed parameter presets or simple feedback adjustment based on real-time signals. Fixed parameters cannot adapt to the dynamic changes in tension characteristics caused by electrode wire diameter loss during machining; simple feedback adjustment, due to response lag, cannot compensate for tension fluctuations caused by sudden changes in operating conditions in real time. This neglect of dynamic tension changes results in poor electrode wire tension control accuracy. Summary of the Invention

[0003] This invention addresses the technical problem of poor electrode wire tension control accuracy in existing slow wire EDM processes by providing an adaptive compensation method and system for electrode wire tension in slow wire EDM.

[0004] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0005] In a first aspect, the present invention provides an adaptive compensation method for electrode wire tension in slow wire EDM, comprising:

[0006] Based on the initial diameter and material of the electrode wire, and constrained by workpiece and cutting parameters, the historical sample set of slow wire EDM is retrieved, and the concentrated value of electrode wire tension is statistically analyzed to obtain the target tension of the electrode wire.

[0007] Based on the target tension of the electrode wire, input the tension control table and configure the first tension control timing parameters;

[0008] Based on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire, wire diameter loss analysis is performed to obtain wire diameter loss timing information;

[0009] Based on the filament diameter loss timing information and the first tension control timing parameters, filament tension analysis is performed to obtain filament tension timing information;

[0010] By performing tension compensation optimization based on the wire tension timing information and the target tension of the electrode wire, the timing information of the second tension control parameter is obtained, and the electrode wire tension compensation control is executed.

[0011] Secondly, the present invention provides an adaptive compensation system for electrode wire tension in slow wire EDM, comprising:

[0012] The target tension determination module is used to retrieve the historical sample set of slow wire EDM based on the initial diameter and material of the electrode wire, combined with workpiece parameters and cutting parameters as constraints, perform electrode wire tension concentration value statistics, and obtain the target tension of the electrode wire.

[0013] The first timing parameter configuration module is used to configure the first tension control timing parameters according to the target tension of the electrode wire by inputting the tension control table.

[0014] The loss timing acquisition module is used to perform wire diameter loss analysis based on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire to obtain wire diameter loss timing information;

[0015] The timing information analysis module is used to perform wire tension analysis based on the wire diameter loss timing information and the first tension control timing parameters to obtain wire tension timing information;

[0016] The compensation execution module is used to perform tension compensation optimization by using the wire tension timing information and the target tension of the electrode wire to obtain the timing information of the second tension control parameter and execute the electrode wire tension compensation control.

[0017] The beneficial effects of this invention are:

[0018] Compared to existing technologies, this application firstly retrieves historical sample sets of slow wire EDM based on the initial diameter and material of the electrode wire, combined with workpiece and cutting parameters as constraints. It then performs statistical analysis of the concentrated values ​​of electrode wire tension to obtain the target tension. By filtering historical samples through multi-parameter constraints, a reliable data foundation for the target tension is provided. Secondly, based on the target electrode wire tension, a tension control table is input, and the first tension control timing parameters are configured, ensuring the rationality of the initial control and providing a benchmark for subsequent dynamic compensation based on real-time losses. Thirdly, wire diameter loss analysis is performed based on the initial electrode wire diameter, workpiece parameters, cutting parameters, and target electrode wire tension to obtain wire diameter loss timing information. This enables accurate prediction of electrode wire diameter loss, providing crucial timing data support for subsequent dynamic tension adjustment based on losses. Furthermore, wire tension analysis is performed based on the wire diameter loss timing information and the first tension control timing parameters to obtain wire tension timing information. This enables accurate prediction of the dynamic changes in the actual electrode wire tension, providing crucial dynamic basis for subsequent tension compensation optimization. Finally, by using the wire tension timing information and the target tension of the electrode wire to perform tension compensation optimization, the timing information of the second tension control parameter is obtained, and the electrode wire tension compensation control is performed, so that the tension control parameter can be adaptively adjusted with the processing progress, ensuring that the electrode wire tension is always stable near the target value, thereby improving processing accuracy, reducing the risk of wire breakage, and ultimately achieving efficient and high-precision slow wire EDM processing.

[0019] Through the above technical solution, this application determines the target tension of the electrode wire by conducting multi-dimensional correlation analysis of the initial diameter of the electrode wire, material, workpiece parameters, and cutting parameters, combined with historical sample data mining. Then, it dynamically predicts the electrode wire diameter loss sequence using an LSTM neural network, performs compensation optimization based on the tension change trend, and generates suitable tension control parameters in real time. This improves the accuracy of electrode wire tension control and adapts to various material and working conditions, achieving a synergistic improvement in both high precision and high efficiency. Attached Figure Description

[0020] Figure 1 A flowchart illustrating an adaptive compensation method for electrode wire tension in slow wire EDM provided by the present invention;

[0021] Figure 2 This is a schematic diagram of a slow wire EDM electrode wire tension adaptive compensation system provided by the present invention.

[0022] In the attached diagram, the components represented by each number are as follows:

[0023] The module includes a target tension determination module 11, a first timing parameter configuration module 12, a loss timing acquisition module 13, a timing information analysis module 14, and a compensation execution module 15. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] In the description of this invention, 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 indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0026] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0027] Example 1, as Figure 1 As shown, this embodiment of the invention provides an adaptive compensation method for electrode wire tension in slow wire EDM, comprising:

[0028] S10: Based on the initial diameter and material of the electrode wire, and constrained by the workpiece parameters and cutting parameters, retrieve the historical sample set of slow wire EDM, perform statistical analysis of the concentrated value of electrode wire tension, and obtain the target tension of the electrode wire.

[0029] In slow wire EDM machining, the diversity of workpiece materials (such as metals and alloys), thickness differences (from a few millimeters to tens of millimeters), and cutting speed parameters increase the difficulty of electrode wire tension control. Traditional tension control methods rely heavily on manually preset experience parameters to adapt to different working conditions. This not only has poor versatility but also requires repeated adjustments to avoid wire breakage or accuracy deviations, resulting in high time and labor costs.

[0030] To address the aforementioned issues, this application, based on the initial diameter and material of the electrode wire, and constrained by workpiece and cutting parameters, retrieves historical sample sets of slow wire EDM, performs statistical analysis of the concentrated values ​​of electrode wire tension, and obtains the target tension of the electrode wire.

[0031] Specifically, step S10 in the method includes:

[0032] Obtain initial workpiece parameters, perform correlation sorting on electrode wire tension, and extract workpiece parameters with correlation degree greater than or equal to the correlation degree threshold;

[0033] Obtain initial cutting parameters, perform correlation sorting on electrode wire tension, and extract the cutting parameters whose correlation degree is greater than or equal to the correlation degree threshold;

[0034] Based on the initial diameter of the electrode wire, the material of the electrode wire, the workpiece parameters, and the cutting parameters, an index constraint condition is constructed;

[0035] Retrieve the historical sample set of slow wire cutting that satisfies the index constraints.

[0036] In this embodiment, initial workpiece parameters are first obtained, and correlation sorting is performed on the electrode wire tension to extract workpiece parameters with a correlation degree greater than or equal to the correlation degree threshold. The initial workpiece parameters include parameters such as workpiece thickness and workpiece hardness. Extracting workpiece parameters with a correlation degree greater than or equal to the correlation degree threshold can reduce interference from irrelevant or low correlation degree parameters.

[0037] Secondly, initial cutting parameters are obtained, and correlation sorting is performed on the electrode wire tension to extract cutting parameters with a correlation degree greater than or equal to the correlation degree threshold. These initial cutting parameters include cutting speed, discharge power, flushing fluid pressure, kerf width, and electrode wire turning angle. The correlation degree threshold can be dynamically set according to actual conditions, for example, 0.6. For example, in a method with identical workpiece parameters, correlation sorting is used to filter out cutting parameters with a correlation degree greater than or equal to the correlation degree threshold, ensuring the relevance of subsequent constraint conditions.

[0038] Furthermore, based on the initial diameter of the electrode wire, the electrode wire material, the workpiece parameters, and the cutting parameters, index constraints are constructed to ensure that the retrieved samples are highly matched with the current processing scenario.

[0039] Finally, based on the constructed index constraints, a set of historical samples of slow wire cutting that meet the index constraints is retrieved from the database as the basis for subsequent statistical analysis of target tension.

[0040] Specifically, the step of "obtaining initial workpiece parameters, performing correlation sorting on electrode wire tension, and extracting workpiece parameters with a correlation degree greater than or equal to a correlation degree threshold" includes:

[0041] Using the initial workpiece parameters as independent variables and the electrode wire tension as dependent variables, initial workpiece parameter datasets and electrode wire tension datasets are collected.

[0042] The electrode wire tension dataset is dimensionless to obtain the electrode wire tension feature sequence;

[0043] The initial workpiece parameter dataset is subjected to dimensionless processing of the same attribute parameters to obtain a series of attribute workpiece parameter feature sequences.

[0044] By combining the electrode wire tension characteristic sequence and the several attribute workpiece parameter characteristic sequences, a grey relational degree matrix is ​​constructed, and grey relational analysis is performed to obtain the relational degree of several attributes.

[0045] Based on the aforementioned attribute correlation, the workpiece parameters with correlation greater than or equal to the correlation threshold are extracted from the initial workpiece parameters.

[0046] In this embodiment of the application, the initial workpiece parameters are first used as independent variables and the electrode wire tension is used as dependent variables. Several corresponding initial workpiece parameter and electrode wire tension data are collected from the historical database as initial workpiece parameter dataset and electrode wire tension dataset.

[0047] Secondly, the electrode wire tension dataset is dimensionless to obtain the electrode wire tension feature sequence. Dimensionless processing is necessary because electrode wire tension data may lack comparability due to differences in units and magnitudes; for example, some electrode wire tension data may be in units of N, while others may be in units of kgf. For instance, normalization and other methods can be used to eliminate the influence of dimensions and obtain an electrode wire tension feature sequence of a uniform magnitude.

[0048] Secondly, the initial workpiece parameter dataset is subjected to dimensionless processing of parameters with the same attribute to obtain a feature sequence of several attribute workpiece parameters. Dimensionless processing of parameters with the same attribute is performed because the initial workpiece parameter data contains data with different attributes, such as workpiece thickness and workpiece hardness. Parameters with the same attribute may not be comparable due to differences in units and magnitudes. Therefore, the influence of dimensions is eliminated for each attribute parameter in the initial workpiece parameter dataset through normalization and other methods to obtain a feature sequence of several attribute workpiece parameters of the same magnitude.

[0049] Furthermore, by combining the electrode wire tension characteristic sequence and several attribute workpiece parameter characteristic sequences, a grey relational degree matrix is ​​constructed, and grey relational analysis is performed to obtain the correlation degrees of several attributes. For example, the grey relational degree matrix is ​​a matrix used in grey relational analysis to quantify the degree of correlation between multiple influencing factors (independent variables) and the target variable (dependent variable). The rows and columns typically correspond to the standardized characteristic sequences of different factors and the standardized characteristic sequence of the target variable, respectively. Each element in the matrix represents the correlation degree between a certain factor sequence and the target variable sequence, with a value ranging from 0 to 1. The closer the correlation degree is to 1, the stronger the correlation. The grey relational degree matrix can be constructed by calculating the correlation coefficients between the electrode wire tension characteristic sequence and several attribute workpiece parameter characteristic sequences. The matrix elements represent the degree of correlation between a certain workpiece parameter attribute and the electrode wire tension; the closer the value is to 1, the stronger the correlation, which can intuitively present the influence strength of each factor on the target variable.

[0050] Finally, based on the correlation of several attributes, workpiece parameters with a correlation degree greater than or equal to the correlation threshold are extracted from the initial workpiece parameters. The correlation threshold is a pre-set judgment threshold, such as 0.6, which can be dynamically adjusted by those skilled in the art according to actual scenarios such as processing accuracy requirements and workpiece material characteristics. For example, if the preset correlation threshold is 0.6, workpiece parameters with a correlation degree ≥ 0.6, such as workpiece thickness and workpiece hardness, are selected from the initial workpiece parameters, while low-correlation parameters with a correlation degree lower than the correlation threshold, such as workpiece color, are removed. This process identifies key workpiece parameters that significantly affect electrode wire tension, effectively reducing the interference of low-correlation parameters on constraint construction and ensuring the relevance and reliability of subsequent index constraints.

[0051] Specifically, the phrase "constructing indexed constraint conditions based on the initial diameter of the electrode wire, the electrode wire material, the workpiece parameters, and the cutting parameters" includes:

[0052] Construct workpiece parameter constraint rules:

[0053] Extract the first correlation degree of the workpiece parameters, iterate through the first correlation degree, sum and compare it with the correlation degree to obtain the weight distribution of the workpiece parameters;

[0054] Obtain the sample workpiece parameters, calculate the standard deviation of the same attribute with the workpiece parameters, obtain the standard deviation of multiple attribute workpiece parameters, perform normalization processing on each, and obtain the deviation characteristic values ​​of multiple attribute workpiece parameters.

[0055] Based on the workpiece parameter weight distribution, the weighted summation of the characteristic values ​​of the deviations of the multiple attribute workpiece parameters is obtained to obtain the comprehensive deviation of the workpiece parameters.

[0056] When the standard deviation of the multiple attribute workpiece parameters is less than or equal to the corresponding standard deviation threshold of the workpiece parameters, and the comprehensive deviation of the workpiece parameters is less than or equal to the comprehensive deviation threshold, the workpiece parameter constraints are considered to be satisfied.

[0057] The configuration process for cutting parameter constraints is the same as that for workpiece parameter constraints.

[0058] When the initial diameter of the sample electrode wire is the same as the initial diameter of the electrode wire, and the material of the sample electrode wire is the same as the material of the electrode wire, and both the workpiece parameter constraints and the cutting parameter constraints are satisfied, it is considered that the index constraint condition is satisfied.

[0059] In this embodiment, index constraints are constructed to ensure that the retrieved historical samples match the current scene. The core constraints include workpiece parameter constraints, cutting parameter constraints, and electrode wire attribute constraints. Specifically:

[0060] First, construct workpiece parameter constraint rules, which mainly include weight distribution calculation, deviation characteristic value calculation, comprehensive deviation calculation, and constraint determination.

[0061] The weight distribution calculation involves extracting the first correlation degree of the workpiece parameters, iterating through the first correlation degree, summing it with the total correlation degree, and then comparing the sums to obtain the workpiece parameter weight distribution. For example, if several workpiece parameters and their correlation degrees selected after the correlation degree threshold are: workpiece parameter A has a correlation degree of 0.8, workpiece parameter B has a correlation degree of 0.7, and workpiece parameter C has a correlation degree of 0.6, with a sum of correlation degrees of 2.1, then the weights are calculated as follows: workpiece parameter A = 0.8 / 2.1 = 0.38, workpiece parameter B = 0.7 / 2.1 = 0.33, and workpiece parameter C = 0.6 / 2.1 = 0.28. Workpiece parameters with higher correlation degrees have a more significant impact on tension, resulting in larger calculated weights. This also makes the difference in deviations more prominent in subsequent calculations, ensuring that the matching of key parameters is given priority.

[0062] The deviation characteristic value calculation involves obtaining sample workpiece parameters, calculating the standard deviation of the same attribute against the workpiece parameters, obtaining standard deviations for multiple attribute workpiece parameters, and then performing normalization processing on each to obtain deviation characteristic values ​​for multiple attribute workpiece parameters. For example, sample workpiece parameters refer to workpiece parameters from historical samples. The standard deviation calculation is performed one by one according to parameter type, such as workpiece thickness, workpiece hardness, etc.: the standard deviation is calculated using the values ​​of sample workpiece parameters and workpiece parameters to measure the degree of dispersion between the two. Since different parameters have different units and magnitudes (e.g., thickness is in mm, hardness is in HRC), directly comparing standard deviations is meaningless. Therefore, each standard deviation needs to be normalized, for example, mapped to the 0-1 interval, to obtain deviation characteristic values ​​for multiple attribute workpiece parameters. In this way, the differences between different parameters are standardized, making the deviations of individual parameters comparable, laying the foundation for subsequent comprehensive evaluation.

[0063] The comprehensive deviation calculation is based on the weight distribution of workpiece parameters. It involves a weighted summation of the deviation characteristic values ​​of multiple attribute workpiece parameters. Specifically, each attribute workpiece parameter deviation characteristic value is multiplied by its corresponding attribute parameter weight, and then summed to obtain the comprehensive deviation of the workpiece parameters. For example, if workpiece parameter A has a weight of 0.38 and a deviation characteristic value of 0.2, workpiece parameter B has a weight of 0.33 and a deviation characteristic value of 0.3, and workpiece parameter C has a weight of 0.28 and a deviation characteristic value of 0.35, then the comprehensive deviation of the workpiece parameters = 0.38*0.2 + 0.33*0.3 + 0.28*0.35 = 0.273. The weighted calculation of the comprehensive deviation of the workpiece parameters highlights the influence of workpiece parameters with higher weights, better reflecting the overall matching degree between the sample and the current workpiece parameters.

[0064] The constraint determination is that when the standard deviation of multiple attribute workpiece parameters is less than or equal to the corresponding standard deviation threshold of the workpiece parameters, and the comprehensive deviation of the workpiece parameters is less than or equal to the comprehensive deviation threshold, the workpiece parameter constraint is considered to be satisfied. The standard deviation threshold of the workpiece parameters is the maximum allowable difference for a single parameter in advance, such as 0.35. The comprehensive deviation threshold is the maximum allowable difference for the whole in advance, such as 0.3. Those skilled in the art can set it dynamically according to the actual situation. In this way, it is ensured that the difference of a single attribute workpiece parameter is within an acceptable range, and the situation of a single parameter being qualified but the overall difference being too large is avoided, ensuring the core matching of the sample with the current scenario.

[0065] Secondly, construct the cutting parameter constraint rules. The configuration process of the cutting parameter constraints is the same as that of the workpiece parameter constraints. First, calculate the correlation between the cutting parameters and the tension to determine the weight. Then, calculate the individual deviation and comprehensive deviation between the sample and the current cutting parameters. Finally, determine whether the constraints are met by using the threshold.

[0066] Finally, if the initial diameter of the sample electrode wire is the same as the initial diameter of the electrode wire, the material of the sample electrode wire is the same as the initial material of the electrode wire, and both the workpiece parameter constraints and the cutting parameter constraints are satisfied, it is considered to meet the index constraint conditions. In this way, historical samples that highly match the current processing scenario can be accurately selected, providing a reliable data foundation for subsequent target tension statistics.

[0067] Specifically, the "retrieval of the slow wire cutting history sample set that satisfies the index constraints" includes:

[0068] Retrieve the first-level slow wire cutting historical sample set that satisfies the index constraints;

[0069] Traverse the historical sample set of the first-level slow wire EDM, extract the initial diameter of the sample electrode wire, the material of the sample electrode wire, the parameters of the sample workpiece and the sample cutting parameters, update the index constraints, and obtain the updated index constraint set.

[0070] Traverse the set of updated index constraints to retrieve the historical sample set of secondary slow wire cutting;

[0071] Add the first-level slow wire EDM historical sample set and the second-level slow wire EDM historical sample set to the slow wire EDM historical sample set.

[0072] In this embodiment of the application, while ensuring the relevance of the samples to the current processing scenario, to avoid insufficient data, a two-level retrieval is used to expand the sample size, providing richer data support for subsequent target tension statistics. Specifically:

[0073] First, retrieve the first-level slow wire EDM historical sample set that meets the index constraints. For example, the first-level slow wire EDM historical sample set refers to historical samples that fully meet the index constraints, meaning that the initial diameter and material of the electrode wire in these samples are completely consistent with the current machining, and both workpiece parameter constraints and cutting parameter constraints are satisfied. The first-level slow wire EDM historical sample set is a core sample that highly matches the current machining scenario, and its data has direct reference value, serving as the basis for subsequent analysis.

[0074] Secondly, the historical sample set of Level 1 slow wire EDM is traversed to extract the initial diameter of the sample electrode wire, the sample electrode wire material, the sample workpiece parameters, and the sample cutting parameters. The index constraints are then updated to obtain an updated set of index constraints. For example, for each sample in the historical sample set of Level 1 slow wire EDM, the initial diameter of the sample electrode wire, the sample electrode wire material, the sample workpiece parameters, and the sample cutting parameters are extracted. Based on these parameters, the original index constraints are fine-tuned. For instance, the thresholds of some parameters can be appropriately relaxed, such as lowering the overall deviation threshold. This is because the historical sample set of Level 1 slow wire EDM is highly matched to the current scenario, and its parameter range can be considered a reasonable fluctuation range similar to the current scenario. After updating the index constraints, more samples that are close to the historical samples of Level 1 slow wire EDM but do not fully meet the original strict constraints can be included, thus expanding the search scope.

[0075] Next, the index constraint set is traversed and updated to retrieve the historical sample set of secondary slow wire EDM. The parameter deviation of the historical sample set of secondary slow wire EDM is slightly greater than that of the historical sample set of primary slow wire EDM, and its correlation with the current scenario is weaker than that of the primary sample set, but it still has reference value. In this way, the amount of data is supplemented by the secondary sample set to avoid the deviation of statistical results due to the insufficient number of primary samples.

[0076] Finally, the historical sample sets of first-level and second-level slow wire EDM cutting were added to the historical sample set of slow wire EDM cutting. This not only preserved the highly matched historical sample set of first-level slow wire EDM cutting, but also supplemented the data diversity through the historical sample set of second-level slow wire EDM cutting, balancing the contradiction between sample correlation and quantity, and providing a more reliable data foundation for subsequent target tension statistics.

[0077] Furthermore, the step of "performing statistical analysis of the concentrated tension value of the electrode wire to obtain the target tension of the electrode wire" includes:

[0078] For the historical sample set of the secondary slow wire EDM, perform electrode wire tension concentration value statistics to obtain the target tension of the secondary electrode wire;

[0079] Extract the selected electrode wire tension set from the historical sample set of the first-level slow wire EDM;

[0080] The target tension of the secondary electrode wire is added to the selected electrode wire tension set, and the electrode wire tension set value statistics are performed to obtain the target tension of the electrode wire.

[0081] In this embodiment, the electrode wire tension concentration value is first statistically analyzed on the historical sample set of secondary slow wire EDM to obtain the target tension of the secondary electrode wire. For example, the concentration value is statistically analyzed on all electrode wire tension data in the historical sample set of secondary slow wire EDM, for example, by taking the mean, mode, or median, to obtain the target tension of the secondary electrode wire. The target tension of the secondary electrode wire is a typical electrode wire tension value reflected in the historical sample set of secondary slow wire EDM. This is because although the correlation of the historical sample set of secondary slow wire EDM is slightly weaker, its quantity is usually large, and its concentration value can reflect the general tension pattern under similar scenarios, serving as a supplement to the primary sample.

[0082] Secondly, the selected electrode wire tension set is extracted from the historical sample set of the first-level slow wire EDM. For example, the electrode wire tension value used in actual processing for each sample is directly extracted from the historical sample set of the first-level slow wire EDM to form the selected electrode wire tension set. The historical sample set of the first-level slow wire EDM is highly matched with the current scenario, and the actual electrode wire tension used is verified and valid data, which is the core reference for the target tension.

[0083] Finally, the target tension of the secondary electrode wire is added to the selected electrode wire tension set, and the aggregated value statistics of the electrode wire tension are performed to obtain the target electrode wire tension. For example, the target tension of the secondary electrode wire is added to the selected electrode wire tension set, and the aggregated value statistics are performed again, for example, by calculating the overall mean or mode, to obtain the final target electrode wire tension. The selected electrode wire tensions from the historical sample set of primary slow wire EDM ensure the reliability of the target tension, while the aggregated values ​​from the historical sample set of secondary slow wire EDM supplement its universality. The fused statistical results are both representative and stable, making them more suitable as the target electrode wire tension for current processing, ultimately providing scientific and reliable data for electrode wire tension control.

[0084] In summary, compared to existing technologies, this application, based on the initial diameter and material of the electrode wire, combined with workpiece and cutting parameters as constraints, retrieves historical sample sets of slow wire EDM, performs statistical analysis of concentrated electrode wire tension values, and obtains the target electrode wire tension. Thus, by filtering historical samples through multi-parameter constraints, a reliable data foundation is provided for the target tension.

[0085] S20: Based on the target tension of the electrode wire, input the tension control table and configure the first tension control timing parameters.

[0086] In this embodiment of the application, the target tension of the electrode wire is converted into a specific initial control command, providing basic parameters for tension adjustment. Specifically:

[0087] The target tension of the electrode wire is used as a query condition and input into the tension control table. The corresponding control parameters are then matched and used as the first tension control timing parameters. The tension control table is a mapping table pre-configured by the user based on historical processing experience, equipment characteristics, or experimental data. Its core content is the one-to-one correspondence between the electrode wire tension range and the control parameters. For example, when the target tension of the electrode wire is in the range of 5 to 8 N, the corresponding control parameters are an adjustment frequency of 5 Hz and an initial output intensity of 60%. When the target tension of the electrode wire is in the range of 8 to 12 N, the corresponding control parameters are an adjustment frequency of 8 Hz and an initial output intensity of 75%, etc.

[0088] In this way, by using a preset tension control table, the abstract target tension of the electrode wire can be quickly converted into specific control commands that the equipment can execute. This avoids having to calculate control parameters from scratch for each processing operation, ensuring the rationality of the initial control and providing a benchmark for subsequent dynamic compensation based on real-time losses.

[0089] S30: Perform wire diameter loss analysis based on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire to obtain wire diameter loss timing information.

[0090] During the cutting process, the diameter of the electrode wire gradually decreases over time due to factors such as discharge corrosion and mechanical friction. Traditional tension control often relies on fixed parameter presets or simple feedback adjustment based on real-time tension, without considering the dynamic impact of diameter loss on the mechanical properties of the electrode wire. It cannot adapt to the changes in tension characteristics caused by the gradually decreasing diameter, resulting in the continuous accumulation of deviation between the actual tension and the target value, ultimately leading to insufficient control accuracy and difficulty in meeting the requirements of high-precision machining.

[0091] To address the aforementioned issues, this application performs wire diameter loss analysis based on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire to obtain wire diameter loss timing information.

[0092] Specifically, step S30 in the method includes:

[0093] Based on the preset electrode wire material, the system collects data on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, the initial tension of the electrode wire, and the timing of wire diameter loss.

[0094] Using the wire diameter loss time-series recording data as supervision, and the electrode wire initial diameter recording data, workpiece parameter recording data, cutting parameter recording data, and electrode wire initial tension recording data as input, a long short-term memory neural network is trained to obtain several wire diameter loss sub-analyzers.

[0095] Perform mode integration on the outputs of the plurality of wire diameter loss sub-analyzers to obtain wire diameter loss analyzers, store them in association with the preset electrode wire material, and add them to the wire diameter loss analyzer set.

[0096] By accessing the wire diameter loss analyzer set based on the electrode wire material, the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire are processed to obtain the timing information of the wire diameter loss.

[0097] In this embodiment, the initial diameter of the electrode wire, workpiece parameters, cutting parameters, initial tension of the electrode wire, and wire diameter loss time sequence data are first collected based on the preset electrode wire material. The preset electrode wire material is used because the diameter loss characteristics of the electrode wire are strongly correlated with its material (such as brass, molybdenum wire, tungsten wire, etc.). Different materials have significant differences in physical strength, corrosion resistance, and discharge wear rate, so they need to be classified according to material. For example, based on a preset electrode wire material, such as brass, data such as the initial diameter of the electrode wire (e.g., 0.1mm, 0.15mm), workpiece parameter data (e.g., material hardness, thickness), cutting parameter data (e.g., pulse current, cutting speed), initial tension data of the electrode wire (e.g., tension value at the start of processing), and corresponding wire diameter loss time sequence data are collected. That is, the difference between the measured value of the electrode wire diameter at different time points during the cutting process and the initial diameter reflects the amount of wire diameter loss over time. For example, the loss is 0.002mm at the 10th second and 0.005mm at the 30th second. These data need to be collected in pairs to form a historical dataset for the preset electrode wire material, providing training data for subsequent model training.

[0098] Secondly, using the time-series data of wire diameter loss as supervision, and the initial diameter data of the electrode wire, the workpiece parameter data, the cutting parameter data, and the initial tension data of the electrode wire as input, a Long Short-Term Memory (LSTM) neural network is trained to obtain several wire diameter loss sub-analyzers. Wire diameter loss is a dynamic process that changes over time, and the current loss is related to the historical loss state. As a neural network that excels at processing time-series data, LSTM can effectively capture this time dependency and has higher prediction accuracy than traditional statistical models (such as linear regression). For example, using the initial diameter data of the electrode wire, the workpiece parameter data, the cutting parameter data, and the initial tension data of the electrode wire as input features, and the time-series data of wire diameter loss as supervision, the parameters of the LSTM are continuously adjusted through backpropagation. This allows the model to learn the pattern of wire diameter loss changing over time based on the input parameters, resulting in a wire diameter loss sub-analyzer. Following the same method, several wire diameter loss sub-analyzers are trained using different subsets of training data, different network parameters, or initialization methods. This is to improve the stability of the model and avoid prediction deviations caused by data fluctuations or overfitting in a single model.

[0099] Next, the mode of the outputs from several wire diameter loss sub-analyzers is integrated to obtain a wire diameter loss analyzer, which is then stored in association with a preset electrode wire material and added to the wire diameter loss analyzer set. For example, when several wire diameter loss sub-analyzers predict loss for the same input, they may output slightly different timing results. For instance, wire diameter loss sub-analyzer 1 predicts a loss of 0.005mm at the 30th second, wire diameter loss sub-analyzer 2 predicts a loss of 0.006mm at the 30th second, and wire diameter loss sub-analyzer 3 predicts a loss of 0.005mm at the 30th second. In this case, the mode of 0.005mm is taken as the final output, which can offset the random error of a single sub-model and improve the robustness of the prediction. The integrated wire diameter loss analyzer is then stored in association with the corresponding preset electrode wire material, such as a brass wire diameter loss analyzer, a molybdenum wire diameter loss analyzer, etc., and added to the wire diameter loss analyzer set for easy and quick retrieval based on the actual material later.

[0100] Finally, the wire diameter loss analyzer set is retrieved based on the electrode wire material. The initial diameter of the electrode wire, workpiece parameters, cutting parameters, and target tension of the electrode wire are processed to obtain the wire diameter loss timing information. For example, based on the material of the electrode wire currently being processed, such as brass, the corresponding analyzer is retrieved from the wire diameter loss analyzer set. The initial diameter of the electrode wire being processed, workpiece parameters, cutting parameters, and target tension of the electrode wire are input, and the predicted loss result of the electrode wire diameter over time during the current processing is output. For example, a loss of 0.001 mm at the 5th second, 0.003 mm at the 20th second, and 0.008 mm at the 60th second, etc., are used as the wire diameter loss timing information.

[0101] In summary, compared to existing technologies, this application performs wire diameter loss analysis based on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire to obtain wire diameter loss timing information. This achieves accurate prediction of electrode wire diameter loss, providing crucial timing data support for subsequent dynamic tension adjustment based on loss, and solving the problem that traditional methods struggle to predict loss patterns, leading to lag in tension control.

[0102] S40: Perform wire tension analysis based on the wire diameter loss timing information and the first tension control timing parameters to obtain wire tension timing information.

[0103] The actual tension of the electrode wire changes dynamically with factors such as diameter loss and working condition fluctuations during the cutting process. This causes the deviation from the target tension to accumulate continuously, which not only affects the processing accuracy and surface quality, but may also increase the risk of wire breakage. Therefore, it is necessary to achieve real-time compensation through precise timing analysis and adaptive control.

[0104] To address the aforementioned issues, this application performs wire tension analysis based on the wire diameter loss timing information and the first tension control timing parameters to obtain wire tension timing information.

[0105] Specifically, step S40 in the method includes:

[0106] Constrained by the preset electrode wire material and tension control equipment model, the timing information of wire diameter loss, the timing data of tension control parameters, and the timing data of wire tension are collected.

[0107] Using the silk tension time-series recording data as supervision, and the silk diameter loss recording time-series information and the tension control parameter time-series recording data as input, a long short-term memory neural network is trained to generate a silk tension analyzer, which is stored in association with the preset electrode wire material and added to the silk tension analyzer set.

[0108] The set of wire tension analyzers is accessed by referring to the electrode wire material, and wire tension analysis is performed based on the wire diameter loss timing information and the tension control parameters to obtain wire tension timing information.

[0109] In this embodiment, the wire diameter loss recording time sequence information, tension control parameter time sequence recording data, and wire tension time sequence recording data are first collected under the constraints of the preset electrode wire material and tension control device model. The preset electrode wire material and tension control device model are used as constraints because the actual tension of the electrode wire is not only related to its own diameter loss, but also affected by the response characteristics of the tension control device, such as control accuracy and delay time. Different electrode wire materials (such as brass and tungsten wire) have different mechanical properties (such as elastic modulus and tensile strength). Therefore, the data must be targeted by the dual constraints of material and device model. The wire tension time sequence recording data can be obtained by the actual change of the electrode wire tension over time as measured by the sensor. For example, constrained by the preset electrode wire material and tension control device model, the following time-series information is collected: wire diameter loss record (e.g., loss of 0.001mm at the 5th second, loss of 0.003mm at the 20th second, loss of 0.008mm at the 60th second), tension control parameter time-series record data (e.g., output voltage of 12V at the 20th second, output voltage of 15V at the 60th second), and corresponding wire tension time-series record data (e.g., tension of 5.2N at the 20th second, tension of 4.8N at the 60th second). This data needs to be aligned by timestamp to form a time-series sample library for the preset electrode wire material and tension control device model, providing data support for subsequent model learning.

[0110] Secondly, using the time-series data of wire tension as supervision, and the time-series information of wire diameter loss and the time-series data of tension control parameters as input, a Long Short-Term Memory Neural Network (LSTM) is trained to generate a wire tension analyzer. This analyzer is stored in association with the preset electrode wire material and added to the wire tension analyzer set. This is because the actual change in wire tension is a time-dependent process. The tension at the current moment is not only affected by the current diameter loss and control parameters, but also related to the tension state at the previous moment. As a neural network that is good at handling time-series dependencies, LSTM can effectively capture this dynamic relationship and is more suitable for predicting the tension that changes over time than traditional static models (such as multiple regression). For example, the timing information of wire diameter loss and the timing data of tension control parameters are used as input features of the model, and the timing data of wire tension is used as a supervision signal. The network parameters of the LSTM are continuously optimized through backpropagation, so that the model can predict the corresponding tension timing based on the input timing data. After training, a wire tension analyzer is generated and associated with the corresponding preset electrode wire material, such as a brass wire tension analyzer, a tungsten wire tension analyzer, etc. It is added to the wire tension analyzer set to form a model library classified by material, which can be quickly called according to the actual processing scenario.

[0111] Finally, the wire tension analyzer set is retrieved based on the electrode wire material. Wire tension analysis is performed based on the wire diameter loss timing information and tension control parameters to obtain wire tension timing information. For example, depending on the material of the electrode wire being processed, such as brass, the corresponding analyzer is retrieved from the wire tension analyzer set. The wire diameter loss timing information and tension control parameters are input, and the wire tension timing information during the current processing is predicted and output. For example, the tension is 5.0N at the 5th second, 4.7N at the 20th second, and 4.2N at the 60th second. The wire tension timing information reflects how the tension deviates from the target value over time under the current wire diameter loss timing information and tension control parameters.

[0112] In summary, compared to existing technologies, this application performs wire tension analysis based on the wire diameter loss timing information and the first tension control timing parameters to obtain wire tension timing information. This enables accurate prediction of the actual dynamic changes in electrode wire tension, providing crucial dynamic basis for subsequent tension compensation optimization and solving the problems of traditional methods failing to predict tension change trends and exhibiting compensation lag.

[0113] S50: By using the tension timing information of the wire and the target tension of the electrode wire, tension compensation optimization is performed to obtain the timing information of the second tension control parameter, and electrode wire tension compensation control is performed.

[0114] In this embodiment, the wire tension timing information is the change in the actual tension of the electrode wire over time based on the wire diameter loss timing information and the tension control parameters. For example, the tension is 5.0N at the 5th second, 4.7N at the 20th second, and 4.2N at the 60th second. The target tension of the electrode wire is an ideal tension value, such as 5.0N. By calculating the deviation timing between the wire tension timing information and the target tension of the electrode wire, that is, the difference between the tension information and the target tension at each time point, such as a deviation of -0.3N at the 20th second and a deviation of -0.8N at the 60th second, these deviations reflect the tension deviation from the target tension over time under the initial control parameters. Based on this, electrode wire tension compensation control can be performed.

[0115] Furthermore, tension compensation optimization refers to finding a control parameter adjustment scheme that minimizes tension deviation through optimization algorithms. For example, with the goal of minimizing the deviation between the wire tension timing information and the target tension of the electrode wire, and with adjustable tension control parameters as optimization variables, such as the dynamic change in tension output intensity and adjustment frequency, iterative calculations (e.g., using gradient descent, genetic algorithms) determine the optimal control parameter adjustment value for each time point, ultimately obtaining the timing information of the second tension control parameter. For instance, at the 20th second, the control voltage needs to be increased from the initial 12V to 13V, and at the 60th second, it needs to be increased to 14.5V. Based on the timing information of the second tension control parameter, real-time compensation of the electrode wire tension is achieved.

[0116] In this way, by compensating for optimization, the tension control parameters can be adaptively adjusted according to the processing progress, ensuring that the electrode wire tension is always stable near the target value, thereby improving processing accuracy, reducing the risk of wire breakage, and ultimately achieving efficient and high-precision slow wire EDM processing.

[0117] In summary, the embodiments of this application have at least the following technical effects:

[0118] Compared to existing technologies, this application first uses the initial diameter and material of the electrode wire, combined with workpiece and cutting parameters as constraints, to retrieve historical sample sets of slow wire EDM, perform statistical analysis of the concentrated values ​​of electrode wire tension, and obtain the target tension of the electrode wire. In this way, by filtering historical samples through multi-parameter constraints, a reliable data foundation is provided for the target tension.

[0119] Secondly, this application inputs the target tension of the electrode wire into a tension control table and configures the first tension control timing parameters. In this way, through the preset tension control table, the abstract target tension of the electrode wire is quickly transformed into specific control commands that the equipment can execute, avoiding the need to calculate control parameters from scratch for each processing operation. This ensures the rationality of the initial control and provides a benchmark for subsequent dynamic compensation based on real-time losses.

[0120] Furthermore, this application performs wire diameter loss analysis based on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire to obtain wire diameter loss timing information. This achieves accurate prediction of electrode wire diameter loss, providing crucial timing data support for subsequent dynamic tension adjustment based on loss, and solving the problem that traditional methods struggle to predict loss patterns, leading to lag in tension control.

[0121] Furthermore, this application performs wire tension analysis based on the wire diameter loss timing information and the first tension control timing parameters to obtain wire tension timing information. This enables accurate prediction of the actual dynamic changes in the electrode wire tension, providing crucial dynamic basis for subsequent tension compensation optimization and solving the problems of traditional methods being unable to predict tension change trends and experiencing compensation lag.

[0122] Finally, this application obtains the second tension control parameter timing information by performing tension compensation optimization using the wire tension timing information and the target electrode wire tension, and then performs electrode wire tension compensation control. In this way, through compensation optimization, the tension control parameters can be adaptively adjusted according to the processing progress, ensuring that the electrode wire tension remains stable near the target value, thereby improving processing accuracy, reducing the risk of wire breakage, and ultimately achieving efficient and high-precision slow wire EDM processing.

[0123] Through the above technical solution, this application determines the target tension of the electrode wire by conducting multi-dimensional correlation analysis of the initial diameter of the electrode wire, material, workpiece parameters, and cutting parameters, combined with historical sample data mining. Then, it dynamically predicts the electrode wire diameter loss sequence using an LSTM neural network, performs compensation optimization based on the tension change trend, and generates suitable tension control parameters in real time. This improves the accuracy of electrode wire tension control and adapts to various material and working conditions, achieving a synergistic improvement in both high precision and high efficiency.

[0124] Example 2, as Figure 2 As shown, based on the same inventive concept as the adaptive compensation method for electrode wire tension provided in Embodiment 1, this embodiment of the invention also provides an adaptive compensation system for electrode wire tension in slow wire EDM, comprising:

[0125] The target tension determination module 11 is used to retrieve the historical sample set of slow wire EDM based on the initial diameter and material of the electrode wire, combined with workpiece parameters and cutting parameters as constraints, perform electrode wire tension concentration value statistics, and obtain the target tension of the electrode wire.

[0126] The first timing parameter configuration module 12 is used to configure the first tension control timing parameters according to the target tension of the electrode wire by inputting the tension control table.

[0127] The loss timing acquisition module 13 is used to perform wire diameter loss analysis based on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters and the target tension of the electrode wire to obtain wire diameter loss timing information;

[0128] The timing information analysis module 14 is used to perform wire tension analysis based on the wire diameter loss timing information and the first tension control timing parameters to obtain wire tension timing information;

[0129] The compensation execution module 15 is used to perform tension compensation optimization by using the wire tension timing information and the target tension of the electrode wire to obtain the timing information of the second tension control parameter and to perform electrode wire tension compensation control.

[0130] Specifically, the target tension determination module 11 is used for:

[0131] Obtain initial workpiece parameters, perform correlation sorting on electrode wire tension, and extract workpiece parameters with correlation degree greater than or equal to the correlation degree threshold;

[0132] Obtain initial cutting parameters, perform correlation sorting on electrode wire tension, and extract the cutting parameters whose correlation degree is greater than or equal to the correlation degree threshold;

[0133] Based on the initial diameter of the electrode wire, the material of the electrode wire, the workpiece parameters, and the cutting parameters, an index constraint condition is constructed;

[0134] Retrieve the historical sample set of slow wire cutting that satisfies the index constraints.

[0135] Specifically, the step of "obtaining initial workpiece parameters, performing correlation sorting on electrode wire tension, and extracting workpiece parameters with a correlation degree greater than or equal to a correlation degree threshold" includes:

[0136] Using the initial workpiece parameters as independent variables and the electrode wire tension as dependent variables, initial workpiece parameter datasets and electrode wire tension datasets are collected.

[0137] The electrode wire tension dataset is dimensionless to obtain the electrode wire tension feature sequence;

[0138] The initial workpiece parameter dataset is subjected to dimensionless processing of the same attribute parameters to obtain a series of attribute workpiece parameter feature sequences.

[0139] By combining the electrode wire tension characteristic sequence and the several attribute workpiece parameter characteristic sequences, a grey relational degree matrix is ​​constructed, and grey relational analysis is performed to obtain the relational degree of several attributes.

[0140] Based on the aforementioned attribute correlation, the workpiece parameters with correlation greater than or equal to the correlation threshold are extracted from the initial workpiece parameters.

[0141] Furthermore, the step of "constructing index constraint conditions based on the initial diameter of the electrode wire, the electrode wire material, the workpiece parameters, and the cutting parameters" includes:

[0142] Construct workpiece parameter constraint rules:

[0143] Extract the first correlation degree of the workpiece parameters, iterate through the first correlation degree, sum and compare it with the correlation degree to obtain the weight distribution of the workpiece parameters;

[0144] Obtain the sample workpiece parameters, calculate the standard deviation of the same attribute with the workpiece parameters, obtain the standard deviation of multiple attribute workpiece parameters, perform normalization processing on each, and obtain the deviation characteristic values ​​of multiple attribute workpiece parameters.

[0145] Based on the workpiece parameter weight distribution, the weighted summation of the characteristic values ​​of the deviations of the multiple attribute workpiece parameters is obtained to obtain the comprehensive deviation of the workpiece parameters.

[0146] When the standard deviation of the multiple attribute workpiece parameters is less than or equal to the corresponding standard deviation threshold of the workpiece parameters, and the comprehensive deviation of the workpiece parameters is less than or equal to the comprehensive deviation threshold, the workpiece parameter constraints are considered to be satisfied.

[0147] The configuration process for cutting parameter constraints is the same as that for workpiece parameter constraints.

[0148] When the initial diameter of the sample electrode wire is the same as the initial diameter of the electrode wire, and the material of the sample electrode wire is the same as the material of the electrode wire, and both the workpiece parameter constraints and the cutting parameter constraints are satisfied, it is considered that the index constraint condition is satisfied.

[0149] Further, the "retrieval of the slow wire cutting history sample set that satisfies the index constraints" includes:

[0150] Retrieve the first-level slow wire cutting historical sample set that satisfies the index constraints;

[0151] Traverse the historical sample set of the first-level slow wire EDM, extract the initial diameter of the sample electrode wire, the material of the sample electrode wire, the parameters of the sample workpiece and the sample cutting parameters, update the index constraints, and obtain the updated index constraint set.

[0152] Traverse the set of updated index constraints to retrieve the historical sample set of secondary slow wire cutting;

[0153] Add the first-level slow wire EDM historical sample set and the second-level slow wire EDM historical sample set to the slow wire EDM historical sample set.

[0154] Furthermore, the step of "performing statistical analysis of the concentrated tension value of the electrode wire to obtain the target tension of the electrode wire" includes:

[0155] For the historical sample set of the secondary slow wire EDM, perform electrode wire tension concentration value statistics to obtain the target tension of the secondary electrode wire;

[0156] Extract the selected electrode wire tension set from the historical sample set of the first-level slow wire EDM;

[0157] The target tension of the secondary electrode wire is added to the selected electrode wire tension set, and the electrode wire tension set value statistics are performed to obtain the target tension of the electrode wire.

[0158] The first timing parameter configuration module 12 is specifically used for:

[0159] Based on the target tension of the electrode wire, input the tension control table and configure the first tension control timing parameters.

[0160] The loss timing acquisition module 13 is specifically used for:

[0161] Based on the preset electrode wire material, the system collects data on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, the initial tension of the electrode wire, and the timing of wire diameter loss.

[0162] Using the wire diameter loss time-series recording data as supervision, and the electrode wire initial diameter recording data, workpiece parameter recording data, cutting parameter recording data, and electrode wire initial tension recording data as input, a long short-term memory neural network is trained to obtain several wire diameter loss sub-analyzers.

[0163] Perform mode integration on the outputs of the plurality of wire diameter loss sub-analyzers to obtain wire diameter loss analyzers, store them in association with the preset electrode wire material, and add them to the wire diameter loss analyzer set.

[0164] By accessing the wire diameter loss analyzer set based on the electrode wire material, the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire are processed to obtain the timing information of the wire diameter loss.

[0165] The timing information analysis module 14 is specifically used for:

[0166] Constrained by the preset electrode wire material and tension control equipment model, the timing information of wire diameter loss, the timing data of tension control parameters, and the timing data of wire tension are collected.

[0167] Using the silk tension time-series recording data as supervision, and the silk diameter loss recording time-series information and the tension control parameter time-series recording data as input, a long short-term memory neural network is trained to generate a silk tension analyzer, which is stored in association with the preset electrode wire material and added to the silk tension analyzer set.

[0168] The set of wire tension analyzers is accessed by referring to the electrode wire material, and wire tension analysis is performed based on the wire diameter loss timing information and the tension control parameters to obtain wire tension timing information.

[0169] The compensation execution module 15 is specifically used for:

[0170] By performing tension compensation optimization based on the wire tension timing information and the target tension of the electrode wire, the timing information of the second tension control parameter is obtained, and the electrode wire tension compensation control is executed.

[0171] In summary, the embodiments of this application have at least the following technical effects:

[0172] Compared to existing technologies, this application firstly uses a target tension determination module. Based on the initial diameter and material of the electrode wire, and constrained by workpiece and cutting parameters, it retrieves historical sample sets of slow wire EDM, performs statistical analysis of the concentrated values ​​of electrode wire tension, and obtains the target tension of the electrode wire. By filtering historical samples through multi-parameter constraints, a reliable data foundation for the target tension is provided. Secondly, through a first timing parameter configuration module, based on the target electrode wire tension, the tension control table is input to configure the first tension control timing parameters, ensuring the rationality of the initial control and providing a benchmark for subsequent dynamic compensation based on real-time losses. Thirdly, through a loss timing acquisition module, wire diameter loss analysis is performed based on the initial diameter of the electrode wire, workpiece parameters, cutting parameters, and the target electrode wire tension to obtain wire diameter loss timing information. This enables accurate prediction of electrode wire diameter loss and provides crucial timing data support for subsequent dynamic tension adjustment based on losses. Furthermore, through the timing information analysis module, wire tension analysis is performed based on wire diameter loss timing information and the first tension control timing parameters to obtain wire tension timing information. This enables accurate prediction of the actual dynamic changes in electrode wire tension, providing crucial dynamic basis for subsequent tension compensation optimization. Finally, through the compensation execution module, tension compensation optimization is performed using the wire tension timing information and the target electrode wire tension to obtain the timing information of the second tension control parameters. Electrode wire tension compensation control is then executed, allowing the tension control parameters to adaptively adjust with the processing progress, ensuring the electrode wire tension remains stable near the target value. This improves processing accuracy, reduces the risk of wire breakage, and ultimately achieves efficient and high-precision slow wire EDM. In this way, the accuracy of electrode wire tension control is improved, and it is adaptable to various materials and processing scenarios, achieving a synergistic improvement in both high precision and high efficiency.

[0173] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0174] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0175] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0176] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0177] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0178] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0179] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for adaptive compensation of electrode wire tension in slow wire EDM, characterized in that, include: Based on the initial diameter and material of the electrode wire, and constrained by workpiece and cutting parameters, the historical sample set of slow wire EDM is retrieved, and the concentrated value of electrode wire tension is statistically analyzed to obtain the target tension of the electrode wire. Based on the target tension of the electrode wire, input the tension control table and configure the first tension control timing parameters; Based on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire, wire diameter loss analysis is performed to obtain wire diameter loss timing information; Based on the filament diameter loss timing information and the first tension control timing parameters, filament tension analysis is performed to obtain filament tension timing information; By performing tension compensation optimization based on the wire tension timing information and the target tension of the electrode wire, the timing information of the second tension control parameter is obtained, and the electrode wire tension compensation control is executed. Based on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire, a wire diameter loss analysis is performed to obtain wire diameter loss timing information, including: Based on the preset electrode wire material, the system collects data on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, the initial tension of the electrode wire, and the timing of wire diameter loss. Using the wire diameter loss time-series recording data as supervision, and the electrode wire initial diameter recording data, workpiece parameter recording data, cutting parameter recording data, and electrode wire initial tension recording data as input, a long short-term memory neural network is trained to obtain several wire diameter loss sub-analyzers. Perform mode integration on the outputs of the plurality of wire diameter loss sub-analyzers to obtain wire diameter loss analyzers, store them in association with the preset electrode wire material, and add them to the wire diameter loss analyzer set. By accessing the wire diameter loss analyzer set based on the electrode wire material, the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire are processed to obtain the timing information of the wire diameter loss.

2. The method as described in claim 1, characterized in that, Based on the initial diameter and material of the electrode wire, and constrained by workpiece and cutting parameters, the historical sample set of slow wire EDM is retrieved, and the concentrated value of electrode wire tension is statistically analyzed to obtain the target tension of the electrode wire, including: Obtain initial workpiece parameters, perform correlation sorting on electrode wire tension, and extract workpiece parameters with correlation degree greater than or equal to the correlation degree threshold; Obtain initial cutting parameters, perform correlation sorting on electrode wire tension, and extract the cutting parameters whose correlation degree is greater than or equal to the correlation degree threshold; Based on the initial diameter of the electrode wire, the material of the electrode wire, the workpiece parameters, and the cutting parameters, an index constraint condition is constructed; Retrieve the historical sample set of slow wire cutting that satisfies the index constraints.

3. The method as described in claim 2, characterized in that, Obtain initial workpiece parameters, perform correlation sorting on electrode wire tension, and extract workpiece parameters with a correlation degree greater than or equal to a correlation degree threshold, including: Using the initial workpiece parameters as independent variables and the electrode wire tension as dependent variables, initial workpiece parameter datasets and electrode wire tension datasets are collected. The electrode wire tension dataset is dimensionless to obtain the electrode wire tension feature sequence; The initial workpiece parameter dataset is subjected to dimensionless processing of the same attribute parameters to obtain a series of attribute workpiece parameter feature sequences. By combining the electrode wire tension characteristic sequence and the several attribute workpiece parameter characteristic sequences, a grey relational degree matrix is ​​constructed, and grey relational analysis is performed to obtain the relational degree of several attributes. Based on the aforementioned attribute correlation, the workpiece parameters with correlation greater than or equal to the correlation threshold are extracted from the initial workpiece parameters.

4. The method as described in claim 2, characterized in that, Based on the initial diameter of the electrode wire, the electrode wire material, the workpiece parameters, and the cutting parameters, indexed constraints are constructed, including: Construct workpiece parameter constraint rules: Extract the first correlation degree of the workpiece parameters, iterate through the first correlation degree, sum and compare it with the correlation degree to obtain the weight distribution of the workpiece parameters; Obtain the sample workpiece parameters, calculate the standard deviation of the same attribute with the workpiece parameters, obtain the standard deviation of multiple attribute workpiece parameters, perform normalization processing on each, and obtain the deviation characteristic values ​​of multiple attribute workpiece parameters. Based on the workpiece parameter weight distribution, the weighted summation of the characteristic values ​​of the deviations of the multiple attribute workpiece parameters is obtained to obtain the comprehensive deviation of the workpiece parameters. When the standard deviation of the multiple attribute workpiece parameters is less than or equal to the corresponding standard deviation threshold of the workpiece parameters, and the comprehensive deviation of the workpiece parameters is less than or equal to the comprehensive deviation threshold, the workpiece parameter constraints are considered to be satisfied. The configuration process for cutting parameter constraints is the same as that for workpiece parameter constraints. When the initial diameter of the sample electrode wire is the same as the initial diameter of the electrode wire, and the material of the sample electrode wire is the same as the material of the electrode wire, and both the workpiece parameter constraints and the cutting parameter constraints are satisfied, it is considered that the index constraint condition is satisfied.

5. The method as described in claim 2, characterized in that, Retrieving the historical sample set of slow wire cutting that satisfies the index constraints includes: Retrieve the first-level slow wire cutting historical sample set that satisfies the index constraints; Traverse the historical sample set of the first-level slow wire EDM, extract the initial diameter of the sample electrode wire, the material of the sample electrode wire, the parameters of the sample workpiece and the sample cutting parameters, update the index constraints, and obtain the updated index constraint set. Traverse the set of updated index constraints to retrieve the historical sample set of secondary slow wire cutting; The first-level slow wire EDM historical sample set and the second-level slow wire EDM historical sample set are added to the slow wire EDM historical sample set.

6. The method as described in claim 5, characterized in that, Perform electrode wire tension concentration value statistics to obtain the target tension of the electrode wire, including: For the historical sample set of the secondary slow wire EDM, perform electrode wire tension concentration value statistics to obtain the target tension of the secondary electrode wire; Extract the selected electrode wire tension set from the historical sample set of the first-level slow wire EDM; The target tension of the secondary electrode wire is added to the selected electrode wire tension set, and the electrode wire tension set value statistics are performed to obtain the target tension of the electrode wire.

7. The method as described in claim 1, characterized in that, Based on the filament diameter loss timing information and the first tension control timing parameters, filament tension analysis is performed to obtain filament tension timing information, including: Constrained by the preset electrode wire material and tension control equipment model, the timing information of wire diameter loss, the timing data of tension control parameters, and the timing data of wire tension are collected. Using the silk tension time-series recording data as supervision, and the silk diameter loss recording time-series information and the tension control parameter time-series recording data as input, a long short-term memory neural network is trained to generate a silk tension analyzer, which is stored in association with the preset electrode wire material and added to the silk tension analyzer set. The set of wire tension analyzers is accessed by referring to the electrode wire material, and wire tension analysis is performed based on the wire diameter loss timing information and the tension control parameters to obtain wire tension timing information.

8. A wire EDM electrode tension adaptive compensation system, characterized in that, For performing the method according to any one of claims 1-7, comprising: The target tension determination module is used to retrieve the historical sample set of slow wire EDM based on the initial diameter and material of the electrode wire, combined with workpiece parameters and cutting parameters as constraints, perform electrode wire tension concentration value statistics, and obtain the target tension of the electrode wire. The first timing parameter configuration module is used to configure the first tension control timing parameters according to the target tension of the electrode wire by inputting the tension control table. The loss timing acquisition module is used to perform wire diameter loss analysis based on the initial diameter of the electrode wire, the workpiece parameters, the cutting parameters, and the target tension of the electrode wire to obtain wire diameter loss timing information; The timing information analysis module is used to perform wire tension analysis based on the wire diameter loss timing information and the first tension control timing parameters to obtain wire tension timing information; The compensation execution module is used to perform tension compensation optimization by using the wire tension timing information and the target tension of the electrode wire to obtain the timing information of the second tension control parameter and execute the electrode wire tension compensation control.

Images