Machining path self-adaptive control system and method of electric spark cutting machine tool

Abstract

The invention provides a self-adaptive control system and method for a machining path of an electric spark cutting machine tool, and the method comprises the steps: obtaining initial data of the electric spark cutting machine tool and operation state data in a real-time machining process, and carrying out the initial correction of a preset theoretical machining path based on initial offset data in the initial data, obtaining an initial processing path; on the basis of the operation state data and state data in the initial machining path, real-time dynamic change data is determined, matching is conducted on the basis of the real-time dynamic change data in combination with a preset error type library, and a current error type and a corresponding error magnitude are obtained; and performing fuzzy reasoning based on the current error type and the error magnitude to obtain a target path compensation parameter, and adjusting a corresponding parameter of the initial processing path based on the path compensation parameter to obtain a target processing path. The technical problem that the machining precision is insufficient due to the fact that the machining path is only based on theoretical parameters and does not consider inherent errors and dynamic accumulative errors is effectively solved.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to an adaptive control system and method for machining paths of an electrical discharge machining (EDM) machine tool. Background Technology

[0002] In the machining process of electrical discharge machining (EDM) machines, the machining path control system is a core component to ensure machining accuracy and efficiency. Most existing EDM machines employ a multi-axis motion control structure, and their machining path control process is as follows: a fixed-format machining path file is generated using pre-designed CAD software; this file is then converted into an executable NC program by a dedicated program generator; and the machine tool then drives each axis and electrode to perform machining operations according to this NC program.

[0003] However, existing control methods generate machining paths based solely on theoretical design parameters, neglecting various errors and dynamic changes during the machining process. On one hand, the mechanical manufacturing process of machine tools inevitably involves offset errors between the electrode fixture and the axis control center, as well as deviations between the electrode's reference point and the actual machining surface. Machine tools equipped with head yaw axes (such as the B-axis) amplify these errors due to electrode yaw movements, directly affecting machining accuracy. On the other hand, dynamic factors such as electrode wear and temperature changes during machining introduce new errors. These errors accumulate, causing the deviation between the machining path and the theoretical path to gradually increase, ultimately reducing workpiece machining accuracy and making it difficult to meet the requirements of high-precision, high-quality machining. Summary of the Invention

[0004] This invention provides an adaptive control system and method for machining paths of an electrical discharge machining (EDM) machine tool, which solves the technical problem of insufficient machining accuracy caused by machining paths based solely on theoretical parameters without considering inherent errors and dynamic cumulative errors. Ultimately, it ensures the machining accuracy of the EDM machine tool and meets the requirements for high-precision and high-quality machining.

[0005] In a first aspect, the present invention provides an adaptive control method for the machining path of an electrical discharge machining (EDM) machine tool, comprising: The initial data of the electrical discharge machining machine tool and the real-time operating status data during the machining process are acquired. Based on the initial offset data in the initial data, the preset theoretical machining path is initially corrected to obtain the initial machining path. The initial data also includes the initial status data corresponding to the operating status data. Based on the running status data and the status data in the initial processing path, the real-time dynamic change data is determined, and the real-time dynamic change data is matched with the preset error type library to obtain the current error type and the corresponding error magnitude; the current error type includes the current dominant error type and the current auxiliary error type. Based on the current error type and magnitude, fuzzy reasoning is performed to obtain the target path compensation parameters. Based on the path compensation parameters, the corresponding parameters of the initial processing path are adjusted to obtain the target processing path. The target machining path is merged with the preset NC program template file to obtain the NC program adapted to the EDM machine tool, and the workpiece machining operation is performed according to the target machining path based on the NC program.

[0006] In a second aspect, the present invention also provides a machining path adaptive control system for an electrical discharge machining (EDM) machine tool, applied to the machining path adaptive control method for an EDM machine tool as described in the first aspect; the machining path adaptive control system for the EDM machine tool includes: The path static correction module is used to acquire the initial data of the EDM machine tool and the real-time operating status data during the machining process, and to perform initial correction on the preset theoretical machining path based on the initial offset data in the initial data to obtain the initial machining path; the initial data also includes the initial status data corresponding to the operating status data. The error type determination module is used to determine the real-time dynamic change data based on the running status data and the status data in the initial processing path, and to match the real-time dynamic change data with a preset error type library to obtain the current error type and the corresponding error magnitude; the current error type includes the current dominant error type and the current auxiliary error type; The path dynamic adjustment module is used to perform fuzzy reasoning based on the current error type and error magnitude to obtain the target path compensation parameters, and to adjust the corresponding parameters of the initial processing path based on the path compensation parameters to obtain the target processing path; The execution output module is used to merge the target machining path with the preset NC program template file to obtain the execution NC program adapted to the EDM machine tool, and to perform workpiece machining operations according to the target machining path based on the execution NC program.

[0007] Thirdly, the present invention also provides an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby realizing the adaptive control method for the machining path of the electrical discharge machining tool as described above.

[0008] Fourthly, the present invention also provides a non-transitory computer-readable storage medium storing a computer software program, which, when executed by a processor, implements the adaptive control method for the machining path of any of the above-described electrical discharge machining tools.

[0009] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the adaptive control method for the machining path of an electrical discharge machining tool as described above.

[0010] The adaptive control method for machining paths of an EDM machine tool provided in this invention obtains an initial machining path by acquiring initial data of the EDM machine tool and real-time operating status data during the machining process, and using initial offset data from the initial data to initially correct the theoretical machining path. This method specifically compensates for inherent errors in the machine tool's mechanical manufacturing process, such as the offset error between the electrode fixture and the axis control center, and the deviation between the electrode's own reference point and the actual machining surface. Furthermore, it obtains real-time change data by performing difference processing between the operating status data and the initial status data, and combines this with preset error types to complete data matching and fusion. This accurately identifies the current error type and magnitude introduced by dynamic factors such as electrode wear and temperature changes during the machining process, and clarifies the priority compensation targets. Subsequently, based on the identified current error type and magnitude, path compensation parameters are obtained through fuzzy reasoning. The corresponding parameters of the initial machining path are adjusted, realizing real-time compensation for the amplification effect of inherent errors and dynamic cumulative errors, forming a target adaptive machining path. Finally, the target adaptive machining path is merged with the preset NC program template file to obtain the NC program adapted to the machine tool, driving the machine tool to process according to the target adaptive machining path. This realizes the real-time identification, compensation and execution of dynamic errors from the correction of initial inherent errors. It effectively solves the technical problem of insufficient machining accuracy caused by machining paths based only on theoretical parameters without considering inherent errors and dynamic cumulative errors. Ultimately, it ensures the machining accuracy of the EDM machine tool and meets the requirements of high-precision and high-quality machining. Attached Figure Description

[0011] Figure 1 This is a flowchart illustrating the adaptive control method for the machining path of an electrical discharge machine tool provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the machining path adaptive control system of the electrical discharge machine tool provided in an embodiment of the present invention; Figure 3 An embodiment diagram of the electronic device provided in this invention; Figure 4 An embodiment diagram of a computer-readable storage medium provided in accordance with the present invention. Detailed Implementation

[0012] 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.

[0013] See Figure 1 , Figure 1 This is a flowchart illustrating the adaptive control method for the machining path of an electrical discharge machining (EDM) machine tool provided by the present invention. In this embodiment, the execution entity of the adaptive control method for the machining path of the EDM machine tool is an adaptive control system. Therefore, the adaptive control method for the machining path of the EDM machine tool includes: Step 10: Obtain the initial data of the EDM machine tool and the real-time operating status data during the machining process, and make initial corrections to the preset theoretical machining path based on the initial offset data in the initial data to obtain the initial machining path; the initial data also includes the initial status data corresponding to the operating status data.

[0014] Optionally, the adaptive control system first collects the initial data of the EDM machine tool based on sensors deployed at various locations on the EDM machine tool. This initial data is the inherent static data of the EDM machine tool. The initial data includes initial offset data and initial state data. The initial offset data refers to the positional deviation values ​​between the electrode fixture and the control center of each axis, and the positional deviation values ​​between the electrode reference point and the actual machined surface, obtained by precision measuring equipment (such as laser interferometer, coordinate measuring machine, etc.) during the factory or commissioning phase of the EDM machine tool. The initial state data refers to the position data of each axis, the ambient temperature data (i.e., the initial room temperature of the machining environment, such as 25℃), and the initial contact pressure data between the electrode and the workpiece (i.e., 0, indicating no contact) when the EDM machine tool is in standby mode before machining begins. At the same time, it also acquires real-time operating status data of the EDM machine tool, such as real-time position data of each axis of the machine tool collected by the grating sensor, discharge current fluctuation value collected by the current and voltage sensors (used to indirectly reflect the electrode wear state and discharge gap changes, such as discharge current fluctuation amplitude and discharge voltage fluctuation amplitude), ambient temperature data collected by the temperature sensor (used to capture thermal deformation error, such as the ambient temperature rising from 25℃ to 30℃ during processing), and contact pressure data between the electrode and the workpiece collected by the pressure sensor (used to determine whether there is abnormal collision caused by path deviation, such as the normal processing contact pressure is 5N, which rises to 20N in abnormal situations).

[0015] Furthermore, after acquiring the initial bias data from the initial data, the adaptive control system performs an initial correction to the preset theoretical machining path based on the initial bias data. The theoretical machining path refers to the set of ideal machining path coordinates generated by CAD software based on the workpiece design drawings, without considering any errors (e.g., the theoretical machining path coordinates of a certain workpiece are...). , , , (where X, Y, and Z are linear axis coordinates, and B is the head yaw axis coordinate). In the initial correction process, the deviation values ​​between the electrode fixture and each axis control center, and the deviation values ​​between the electrode reference point and the actual machining surface in the initial offset data are compensated with the corresponding axis coordinates and electrode machining position coordinates in the theoretical machining path to obtain the corrected initial machining path coordinate set.

[0016] In one embodiment, the initial data acquired by the adaptive control system of an EDM machine tool with a B-axis (head yaw axis) is as follows: the initial offset data is the deviation value between the electrode fixture and the X-axis control center. mm, deviation from the Y-axis control center mm, deviation from the Z-axis control center mm, the deviation between the electrode reference point and the actual machined surface (B-axis angle deviation); initial state data is the initial position of the X-axis. mm, initial position of the Y-axis mm, initial Z-axis position mm, initial angle of B-axis Initial ambient temperature ℃, initial contact pressure N. The coordinates of a certain feature point in the preset theoretical processing path are ( mm, mm, mm, ).

[0017] The adaptive control system performs initial corrections based on the initial bias data, and the correction formula is as follows: ; ; ; ; Substituting the data, we get: ; ; ; ; The final initial machining path coordinates corresponding to this feature point are (9.98mm, 7.99mm, 4.995mm, 29.985°). After all feature points are corrected, a complete initial machining path is formed.

[0018] Step 20: Based on the running status data and the status data in the initial processing path, determine the real-time dynamic change data, and match the real-time dynamic change data with the preset error type library to obtain the current error type and the corresponding error magnitude; the current error type includes the current dominant error type and the current auxiliary error type.

[0019] Optionally, after acquiring real-time operating status data and the initial machining path, the adaptive control system first performs difference processing on the operating status data and the status data in the initial machining path. This involves subtracting the real-time acquired values ​​of each parameter in the operating status data from the corrected status data in the initial machining path to obtain the change in each parameter, i.e., the real-time dynamic change data. This real-time dynamic change data includes the dynamic change in the position of each axis (real-time position data of a machine tool axis minus the initial position data of that axis), the dynamic change in discharge current fluctuation (real-time discharge current fluctuation value minus the initial discharge current reference value; since there is no discharge in the initial state, the reference value is zero), the dynamic change in ambient temperature (real-time ambient temperature data minus the initial ambient temperature data), and the dynamic change in contact pressure (real-time contact pressure data minus the initial contact pressure data).

[0020] Furthermore, after acquiring real-time dynamic change data, the adaptive control system matches the real-time dynamic change data with a preset error type library to obtain the current error type and its corresponding error magnitude, as detailed in steps 201-206. The error type library is a pre-stored database of various error characteristics and their corresponding error types, stored by the adaptive control system. The error types in the library include mechanical errors, thermal deformation errors, electrode wear errors, and collision errors. Each error type corresponds to a range of characteristic parameters and an error magnitude classification standard (e.g., the characteristic range of dynamic changes in shaft position for mechanical errors, the characteristic range of dynamic changes in ambient temperature for thermal deformation errors, etc.). The final current error type includes the current dominant error type and the current auxiliary error type, along with their corresponding error magnitude.

[0021] Continuing with the above embodiments, the adaptive control system collects the following operating status data during real-time machining: real-time position of the machine tool's X-axis. Y-axis real-time position Z-axis real-time position B-axis real-time angle Real-time discharge current fluctuation value A (Initial discharge current reference value) A), Real-time ambient temperature Tactual = 28℃, Real-time contact pressure Factual = 6N.

[0022] Based on this, the dynamic change in the X-axis position is obtained by performing difference processing on the real-time dynamic change data: Similarly, we can conclude that , , Dynamic changes in discharge current fluctuations Dynamic changes in ambient temperature Dynamic change in contact pressure .

[0023] Step 30: Perform fuzzy reasoning based on the current error type and error magnitude to obtain the target path compensation parameters, and adjust the corresponding parameters of the initial processing path based on the path compensation parameters to obtain the target processing path.

[0024] Optionally, after determining the current error type and error magnitude, the adaptive control system performs fuzzy inference based on the current dominant error type and the current auxiliary error type and their corresponding error magnitudes to obtain the target path compensation parameters, as described in steps 301-305.

[0025] Furthermore, after completing the fuzzy inference, the corresponding parameters of the initial machining path are adjusted according to the obtained path compensation parameters. That is, the compensation amount of each axis in the target path compensation parameters is added to the corresponding axis coordinate in the initial machining path to obtain the adjusted target machining path coordinate set, i.e., the target machining path.

[0026] Step 40: Merge the target machining path with the preset NC program template file to obtain the NC program adapted to the EDM machine tool, and perform workpiece machining operations according to the target machining path based on the NC program.

[0027] Optionally, after obtaining the target machining path, the adaptive control system merges it with a preset NC program template file to generate an NC program adapted to the EDM machine tool. The preset NC program template file is a framework file that is pre-stored by the adaptive control system and conforms to the NC program syntax specification of the EDM machine tool. This file contains basic machine tool operation instructions (such as spindle start instructions, discharge start instructions, and feed rate setting instructions) and reserved path coordinate insertion positions (i.e., placeholders in the template used to fill the machining path coordinates). The specific fusion process is as follows: First, the target machining path is parsed, and all coordinate point data (the set of X, Y, Z, and B axis coordinates arranged in the machining order) is extracted. Next, the extracted coordinate point data is filled into the path coordinate insertion positions in the preset NC program template file one by one according to the NC program syntax specification. At the same time, according to the machining requirements of the target machining path (such as feed rate and discharge parameters), the corresponding basic instruction parameters in the template file are adjusted (such as setting the speed value in the feed rate instruction to a value that adapts to the target path). Finally, the NC program file after filling and adjustment is syntax-checked to ensure that there are no syntax errors. After the check passes, the NC program adapted to the EDM machine tool is obtained.

[0028] Furthermore, after the adaptive control system receives the generated NC program, it sends the program to the motion control module of the EDM machine tool. The machine tool then executes the workpiece machining operation according to the target machining path based on the NC program. In other words, the machine tool motion control module receives the instructions in the NC program and drives each axis (X, Y, Z, B axes) to move according to the coordinates of the target machining path. At the same time, it controls the electrodes to perform electrical discharge cutting with the workpiece according to the discharge parameters set in the program, until the machining of the entire workpiece is completed.

[0029] This invention, through acquiring initial data and real-time operational status data of an EDM machine tool, and using initial offset data from the initial data to initially correct the theoretical machining path, obtains an initial machining path. This specifically compensates for inherent errors in the machine tool's mechanical manufacturing process, such as the offset error between the electrode fixture and the axis control center, and the deviation between the electrode's own reference point and the actual machining surface. Real-time change data is obtained by subtracting the operational status data from the initial status data. Combined with preset error types, data matching and fusion are completed to accurately identify the current error type and magnitude introduced during machining due to dynamic factors such as electrode wear and temperature changes, clarifying the priority compensation targets. Subsequently, based on the identified current error... The error type and magnitude are obtained through fuzzy inference to obtain path compensation parameters. The corresponding parameters of the initial machining path are adjusted to realize real-time compensation for the amplification effect of inherent errors and dynamic cumulative errors, forming a target adaptive machining path. Finally, the target adaptive machining path is integrated with the preset NC program template file to obtain the NC program adapted to the machine tool, which drives the machine tool to process according to the target adaptive machining path. This realizes the real-time identification, compensation and execution of dynamic errors from the correction of initial inherent errors to the correction of inherent errors. It effectively solves the technical problem of insufficient machining accuracy caused by machining paths based only on theoretical parameters without considering inherent errors and dynamic cumulative errors. Ultimately, it ensures the machining accuracy of the EDM machine tool and meets the requirements of high precision and high quality machining.

[0030] In one embodiment, steps 201-206 include: Step 201: Based on the real-time dynamic change data, compare it one by one with the error type library, and filter out the error types that have a matching relationship with each real-time dynamic change data in the feature dimension to obtain the preliminary error type association group.

[0031] Optionally, the adaptive control system will calculate the real-time dynamic change data ( , , , , , , The data is compared one by one with the characteristic dimension parameters of various errors in the error type library. If a real-time dynamic data parameter belongs to the characteristic dimension parameter of a certain type of error, then the real-time dynamic data is associated with the corresponding error type. After traversing all real-time dynamic data parameters, all error types with established associations are summarized to form a preliminary error type association group.

[0032] Furthermore, the error type library stores four error types: mechanical error, thermal deformation error, electrode wear error, and collision error. Each error type corresponds to at least one characteristic dimension parameter (a characteristic dimension parameter refers to a real-time dynamic change data parameter that can directly reflect the generation and change of the error type). The specific correspondence is as follows: the characteristic dimension parameter of mechanical error is the deviation value of each axis position ( , , , The characteristic dimension parameter of thermal deformation error is the dynamic change in ambient temperature. ) and the position deviation values ​​of each axis (due to temperature changes causing deformation of the mechanical structure, which in turn leads to axis position offset); the characteristic dimension parameter of electrode wear error is the dynamic change of discharge current fluctuation ( The characteristic dimension parameter of collision error is the dynamic change in contact pressure. ).

[0033] Step 202: Based on the consistency judgment between the characteristic change trend corresponding to each error type in the preliminary error type association group and the actual change trend of each real-time dynamic change data, an error type trend matching group is obtained.

[0034] Optionally, the adaptive control system extracts the characteristic change trend corresponding to each error type based on the obtained preliminary error type association group (this characteristic change trend refers to the regular change direction of the characteristic dimension parameter as the error intensity increases when the error type occurs, which is summarized based on a large amount of processing experimental data). Specifically, the characteristic change trend of mechanical error is: as the error intensity increases, the absolute value of the position deviation of each axis increases; the characteristic change trend of thermal deformation error is: as the error intensity increases, the dynamic change of ambient temperature increases, and the absolute value of the position deviation of each axis also increases; the characteristic change trend of electrode wear error is: as the error intensity increases, the dynamic change of discharge current fluctuation increases; and the characteristic change trend of collision error is: as the error intensity increases, the dynamic change of contact pressure increases.

[0035] Next, based on the actual changing trends of the acquired real-time dynamic data (the actual changing trend refers to the direction of numerical change of the real-time dynamic data from the start of processing to the current moment, which can be obtained by comparing the current data with three or more continuously collected historical data sets), the consistency between the characteristic changing trend of each error type in the preliminary error type association group and the actual changing trend of the corresponding characteristic dimension parameter is judged: if the actual changing trends of all characteristic dimension parameters of an error type are consistent with the characteristic changing trend of that error type, then the error type is retained; if the actual changing trend of at least one characteristic dimension parameter is inconsistent with the characteristic changing trend, then the error type is removed. The final retained error types form the error type trend matching group.

[0036] Step 203: Based on each error type in the error type trend matching group, analyze the correlation between the impact of each real-time dynamic change data on the error type, and based on the analysis results, count the change magnitude of the impact of each error type on the real-time dynamic change data. Sort the error types in the error type trend matching group according to the change magnitude to obtain the error type impact ranking result.

[0037] Optionally, the adaptive control system analyzes the correlation between various real-time dynamic data and their impact on each error type in the error type trend matching group (this correlation refers to the "direct impact" or "indirect impact" of real-time dynamic data parameters on the error type; direct impact means that the parameter is a core characteristic dimension parameter of the error type, directly determining whether an error occurs; indirect impact means that the parameter is a non-core characteristic dimension parameter of the error type, only having a slight adjustment effect on the degree of error after it occurs), and analyzes the correlation between mechanical errors: the position deviation values ​​of each axis ( , , , This has a direct impact on it (core feature dimension parameters). , , It has no effect on it (non-feature dimension parameter); correlation analysis of its impact on thermal deformation error: The positional deviation of each axis has a direct impact on it (core feature dimension parameter), while the positional deviation of each axis has an indirect impact on it (non-core feature dimension parameter, which only reflects the error result). , It has no effect on it; correlation analysis of the influence on electrode wear error: The core feature dimension parameter has a direct impact on it, while other parameters have no effect on it; correlation analysis of the impact on collision error: It has a direct impact on the core feature dimension parameters, while other parameters have no effect on it.

[0038] Next, for each type of error, the real-time dynamic data parameters that directly affect it are extracted, and the magnitude of change of these parameters is calculated (the magnitude of change refers to the absolute value of the current value of the parameter; since the units of each parameter are different, there is no need for uniform conversion, and the degree of influence is measured only based on its own absolute value). For mechanical errors, the average value of the absolute values ​​of the position deviations of each axis is taken as the magnitude of change, and the formula is: The B-axis angular deviation needs to be converted to a linear deviation using the following formula: ;in, The equivalent diameter of the electrode is taken here. , which are the machine tool default parameters; for thermal deformation error: take The absolute value is used as the range of change, i.e. For electrode wear error: take The absolute value is used as the range of change, i.e. For collision error: take The absolute value is used as the range of change, i.e. Finally, the error types in the error type trend matching group are sorted according to the order of magnitude of change from largest to smallest to obtain the sorting result of the influence of error type.

[0039] Step 204: Based on the impact of error type on the sorting results and the preset real-time processing accuracy requirements, an adaptability evaluation is performed on each sorted error type to obtain an error type adaptability classification result; the error type adaptability classification result includes adaptable error types and non-adaptable error types.

[0040] Optionally, the adaptive control system affects the sorting results based on the determined error type. First, it obtains the preset real-time machining accuracy requirements (real-time machining accuracy requirements refer to the allowable error range corresponding to the currently machined workpiece, which is preset by the operator in the system according to the workpiece design drawings, such as: the maximum allowable mechanical error is 0.02mm, and the allowable thermal deformation error is...). Maximum temperature 8℃, electrode wear error permissible Maximum value 1.5A, collision error allowed The maximum value is 15N, and this requirement includes a threshold for the allowable variation range for each error type (the allowable variation range threshold refers to the maximum permissible value of the directly affecting parameter corresponding to the error type; exceeding this threshold will result in substandard machining accuracy). Then, for each error type in the error type impact ranking results, the variation range of its directly affecting parameter is compared with the corresponding allowable variation range threshold to perform an adaptation assessment: if the variation range is ≤ the allowable variation range threshold, the error type will not cause the machining accuracy to exceed the requirement and is assessed as an adaptable error type; if the variation range is > the allowable variation range threshold, the error type will cause the machining accuracy to fail and is assessed as a non-adaptable error type. Finally, all assessment results are summarized to form an error type adaptation classification result, where adaptable error types refer to error types that currently will not cause the machining accuracy to exceed the standard, and non-adaptable error types refer to error types that currently will cause the machining accuracy to exceed the standard.

[0041] Step 205: Based on the error type adaptability classification results, analyze the interaction relationship between adaptability error types, and determine the adaptability error type with the highest impact on machining accuracy as the current dominant error type according to the analysis results, and determine the other adaptability error types as the current auxiliary error types.

[0042] Optionally, the adaptive control system analyzes the interaction relationships between the adaptation level error types in the determined error type adaptation classification results (interaction relationship refers to whether there is a causal relationship or an independent relationship between different error types; a causal relationship means that the occurrence of one error will lead to the occurrence of another error, and an independent relationship means that the occurrence of two errors has no direct causal relationship) based on the interaction relationship table stored in the error type library (based on the summary of processing mechanism). For example, the specific interaction relationships in this table include thermal deformation error and mechanical error: thermal deformation error (temperature change) will lead to mechanical structure deformation, which will aggravate mechanical error (axis position deviation), which is a causal relationship (thermal deformation error is the cause, and mechanical error is the reinforcing factor of the effect); electrode wear error and mechanical error: electrode wear will lead to changes in the discharge gap, which may slightly affect the axis motion accuracy, which is a weak causal relationship (small degree of influence); collision error and other errors: collision error did not occur ( (Not reaching the threshold), which is independent of other errors; thermal deformation error and electrode wear error: have no direct causal relationship, and are independent.

[0043] Next, the impact percentage of each adapter level error type on machining accuracy is calculated (the impact percentage refers to the proportion of machining deviation caused by that error type to the total machining deviation of all adapter level error types; the total machining deviation is the weighted sum of the changes in parameters directly affected by each error type, with the weights based on the error's impact coefficient on machining accuracy, and the impact coefficient is determined by industry machining experiment data, such as the mechanical error impact coefficient). thermal deformation error Electrode wear error Collision error (because If the threshold is not reached, the influence coefficient is 0). The formula is as follows: ;in, For the first The percentage of impact of each error type; For the first Influence coefficients of various error types; For the first The type of error directly affects the magnitude of parameter change; The number of adaptation-level error types is determined. Finally, the adaptation-level error type with the highest impact is identified as the current dominant error type, and the remaining adaptation-level error types are identified as the current auxiliary error types.

[0044] Step 206: Based on the magnitude of the change in the real-time dynamic data corresponding to each error type and the preset error magnitude rules, calculate the error magnitude of the current dominant error type and the error magnitude of the current auxiliary error type.

[0045] Optionally, after determining the current dominant error type and the current auxiliary error type, the adaptive control system extracts a preset error magnitude rule from the error type library (the error magnitude rule refers to the specific range of mild, moderate, and severe errors for each error type based on the magnitude of change of the parameters directly affected by it; this range is determined based on machining accuracy requirements and experimental data). The specific rule is: Mechanical error: directly affects the average magnitude of change of the parameters. 0.005mm≤ <0.01mm is considered a slight error, 0.01mm ≤ <0.015mm is considered a moderate error. ≥0.015mm is considered a specific gravity error; for thermal deformation error: 2℃≤ <5℃ is considered a slight error; 5℃≤ A temperature below 8℃ is considered a moderate error. ≥8℃ is considered a severe error; for electrode wear error: 0.5A≤ <1.0A is considered a slight error, 1.0A ≤ <1.5A is considered a moderate error. ≥1.5A is considered a severe error; for collision errors: 15N≤ <25N is considered a slight error; 25N≤ <35N is considered a moderate error. ≥35N is considered a severe error. Next, for the current dominant error type (mechanical error) and the current auxiliary error type (thermal deformation error, electrode wear error), the variation amplitude of their directly affecting parameters is substituted into the corresponding error magnitude rules to determine their respective error magnitudes.

[0046] The embodiments of the present invention improve the accuracy of error type identification, reduce the error magnitude judgment bias, and avoid the error compensation parameter error caused by misjudgment by layered screening and analysis, from initial association to trend matching, and then to influence ranking, adaptability classification, and determination of dominant / auxiliary type and magnitude.

[0047] In one embodiment, the process of steps 301-305 includes: Step 301: Based on the current dominant error type, the current auxiliary error type, and the corresponding error magnitude, determine the physical carrier shared by different error types, and construct an error direct coupling correlation factor based on the characteristics of the shared physical carrier and the error magnitude of each error type. The error direct coupling correlation factor is used to quantify the degree of direct coupling influence between different error types due to the shared physical carrier. Direct coupling influence refers to the superposition or amplification of error magnitudes caused by multiple types of errors sharing the same physical carrier.

[0048] Optionally, the adaptive control system, based on the determined current dominant error type, current auxiliary error type, and corresponding error magnitude, first analyzes the physical carriers upon which each error type depends. These physical carriers refer to the specific components or structures in the EDM machine tool that bear the generation and transmission of errors, such as linear axes (X, Y, Z axes), head yaw axis (B-axis), electrodes, and electrode fixtures. It then filters out the physical carriers shared by different error types. Specifically, for mechanical errors: the physical carriers are the linear axes (X, Y, Z axes) and the B-axis (because mechanical errors originate from the offset between the axis control center and the fixture, and the electrode reference deviation, directly reflected in the axis movement and electrode position); for thermal deformation errors: the physical carriers are the linear axes (X, Y, Z axes) and the B-axis (because temperature changes cause thermal expansion and contraction of the axis structure, leading to axis position shifts); for electrode wear errors: the physical carrier is the electrode (because wear occurs directly on the electrode machining surface, affecting the discharge gap and machining path). Therefore, the shared physical carriers are determined to be the linear axes (X, Y, Z axes), the B-axis (shared by mechanical error and thermal deformation error), and the electrodes (only electrode wear error depends on them separately, with no sharing). The core shared physical carriers are the linear axes and the B-axis.

[0049] Subsequently, based on the characteristics of the shared physical carrier (including the carrier's coefficient of thermal expansion, mechanical stiffness, and motion accuracy level), these characteristics are pre-stored in the system database, such as the coefficients of thermal expansion of the X / Y / Z axes. B-axis thermal expansion coefficient Shaft mechanical stiffness N / μm) and the error magnitude of each error type (e.g., mechanical error: slight, mm; Heat distortion error: slight. ℃; Electrode wear error: slight. A) Construct an error direct coupling correlation factor. This factor is used to quantify the direct coupling effect between mechanical errors and thermal deformation errors caused by shared shaft type carriers (e.g., shaft offset caused by thermal deformation will be superimposed on shaft deviation caused by mechanical errors). Its calculation formula is as follows: ;in, For the first On a shared physical carrier (axis), the first The direct coupling correlation factor between auxiliary errors (thermal deformation errors) and dominant errors (mechanical errors); For the first Thermal expansion coefficient of each axis (unit: / ℃); The temperature change of thermal deformation error (unit: °C); For the first Effective motion length of each axis (unit: such as the effective length of the X-axis Y-axis Z-axis B-axis equivalent linear length , (Electrode radius) (Processing angle) mm); For the first Average deviation of mechanical error on each axis (unit: mm, e.g., X-axis) , mm, Y-axis , mm, Z-axis , mm, B-axis , mm); The coupling effect correction factor (determined based on shaft mechanical stiffness, stiffness) hour (According to preset settings).

[0050] Step 302: Based on the error direct coupling correlation factor, and combining the current dominant error type, the current auxiliary error type and the corresponding error magnitude, analyze the comprehensive influence trend of different error types on the processing path deviation, and obtain the initial path compensation direction.

[0051] Optionally, the adaptive control system analyzes the comprehensive influence trend of different error types on the machining path deviation based on the determined error direct coupling correlation factor, combined with the current dominant error type, the current auxiliary error type and the corresponding error magnitude. The machining path deviation refers to the coordinate difference between the real-time machining path and the initial machining path, and finally determines the initial path compensation direction, as shown in steps 3021-3023.

[0052] Step 303: Based on the initial path compensation direction and the characteristics of the current dominant error type and the current auxiliary error type, a judgment is made to obtain the potential risk points and corresponding potential risk values ​​of indirect coupling effects; indirect coupling effects refer to the compensation operation of a certain type of error indirectly causing another type of new error that is not directly related to the original error.

[0053] Optionally, the adaptive control system, based on the obtained initial path compensation direction, and in conjunction with the characteristics of the current dominant error type and the current auxiliary error type (mechanical error characteristics: directly related to axis position, compensation requires precise matching of axis deviation; thermal deformation error characteristics: dynamically fluctuates with temperature changes, compensation requires reserving temperature margin; electrode wear error characteristics: related to discharge gap, excessive compensation can easily lead to discharge instability), makes a judgment to obtain the potential risk points and corresponding potential risk values ​​of indirect coupling effects. Among them, indirect coupling effects refer to compensation operations for one type of error that do not directly act on the physical carrier of another type of error, but due to the correlation of the machine tool system, indirectly cause new errors that are not directly related to the original error (such as overcompensating the Z-axis to offset mechanical errors, which may lead to abnormal contact pressure between the electrode and the workpiece, causing collision error risk), as detailed in steps 3031-3034.

[0054] Step 304: Based on the potential risk points affected by indirect coupling and their corresponding potential risk values, sort the different potential risk points to obtain risk avoidance priorities.

[0055] Optionally, the adaptive control system prioritizes potential risk points based on the identified indirect coupling effects and their corresponding potential risk values. The risk avoidance priority is based on the risk value of each risk point and the severity of its consequences (severity is categorized as minor, moderate, severe, and catastrophic, with corresponding weighting coefficients W=1, 3, 5, and 10, respectively). Therefore, the severity of the consequences caused by each risk point and its corresponding weighting coefficient are first determined. For example, for risk point 1 (Z-axis overcompensation → discharge fluctuation): the consequences are unstable discharge, decreased processing efficiency, and slight deviation in surface quality; the severity of the consequences is minor. For risk point 2 (X-axis overcompensation → collision): the consequence is that the electrode collides with the workpiece, resulting in electrode damage and workpiece scrapping. The severity of the consequence is severe. For risk point 3 (B-axis compensation → surface roughness): the consequence is that the surface roughness of the machined surface exceeds the tolerance, requiring subsequent grinding. The severity of the consequence is moderate. Then, the risk priority index for each risk point is calculated using the following formula: ;in, Let i be the potential risk value of the i-th risk point; Let be the priority index of the i-th risk point. The larger the index value, the higher the risk aversion priority. Finally, the risk points are sorted from largest to smallest according to their priority indices to obtain the risk aversion priority.

[0056] Step 305: Adjust the compensation parameters based on the error direct coupling correlation factor, the initial path compensation direction, the potential risk points of indirect coupling influence, and the risk avoidance priority to obtain the target path compensation parameters.

[0057] Optionally, after obtaining the error direct coupling correlation factor, the initial path compensation direction, the potential risk points of indirect coupling influence, and the risk avoidance priority, the adaptive control system adjusts the compensation parameters to finally obtain the target path compensation parameters, as described in steps 3051-3054.

[0058] The embodiments of the present invention achieve dynamic optimization of compensation parameters by error coupling quantification (accurately identifying the moderate coupling effect of thermal deformation and mechanical error), direction judgment (avoiding compensation direction errors), risk prediction (early identification of risks such as collisions and discharge fluctuations), and dual-dimensional calculation (coupling adjustment + risk correction), thereby improving compensation accuracy and effectively avoiding the loss of processing precision caused by compensation deviation.

[0059] In one embodiment, the process of steps 3021-3023 includes: Step 3021: If the error direct coupling correlation factor is greater than or equal to the preset correlation factor threshold, and the current dominant error type and the current auxiliary error type cause the same direction of the processing path deviation, then the direction opposite to the deviation direction is determined as the initial path compensation direction.

[0060] Optionally, the adaptive control system has a pre-set correlation factor threshold. This threshold is used to determine whether the direct coupling correlation factor of the error has reached a critical value that requires serious consideration of the influence of auxiliary errors. This threshold is set based on a large amount of machining experimental data and the law of error coupling influence, and is stored in the system database. In the shared physical carrier of shafts for EDM machine tools, the correlation factor threshold... (when When the auxiliary error is coupled with the dominant error, the effect needs to be included in the compensation direction judgment; when (At this time, the impact of auxiliary errors is not a primary consideration). Therefore, after determining the correlation factor threshold, a judgment operation is performed. First, the relationship between the magnitude of the direct coupling correlation factor of each axis error and the preset correlation factor threshold is determined. To determine if this is true, further analyze whether the direction of the machining path deviation caused by the current dominant error type (e.g., mechanical error) and the current auxiliary error type (e.g., thermal deformation error) is the same. Same deviation direction means that both types of errors cause the machining path to shift in the same coordinate change direction (e.g., both shift in the "coordinate decreasing direction" or both shift in the "coordinate increasing direction"). This is determined by analyzing the influence trends of the two types of errors (the influence trend of mechanical error is shift in the "coordinate decreasing direction," and the influence trend of thermal deformation error is shift in the "coordinate increasing direction"). If both of the above judgments are true, then... If the deviation directions are the same, then the direction opposite to the deviation direction is determined as the initial path compensation direction. The direction opposite to the deviation direction means: if both types of errors cause the machining path to shift in the direction of decreasing coordinates, then the compensation direction is the direction of increasing coordinates; if they both shift in the direction of increasing coordinates, then the compensation direction is the direction of decreasing coordinates, so as to cancel out the deviation after the two types of errors are superimposed.

[0061] Step 3022: If the error direct coupling correlation factor is greater than the correlation factor threshold, and the current dominant error type and the current auxiliary error type cause the processing path deviation direction to be opposite, then the direction opposite to the deviation direction corresponding to the error type with larger error magnitude is determined as the initial path compensation direction.

[0062] Optionally, the adaptive control system again checks if the direct coupling correlation factor of the error is greater than the correlation factor threshold. If the result is yes, it checks whether the machining path deviation directions caused by the current dominant error type and the current auxiliary error type are opposite (i.e., one type of error shifts towards the direction of coordinate decrease, and the other type shifts towards the direction of coordinate increase). If yes, it first compares the magnitudes of the two types of errors. The comparison of error magnitudes is based on comparing the quantified values ​​of the deviation degree corresponding to the two types of errors. For the quantified value of the deviation degree of the dominant error (mechanical error): the average of the absolute values ​​of the position deviations of each axis is taken; for the quantified value of the deviation degree of the auxiliary error (thermal deformation error): the average of the linear offsets caused by thermal deformation of each axis is taken. Then, the direction opposite to the deviation direction corresponding to the error type with a larger error magnitude (i.e., the error type with a larger deviation degree quantified value) is determined as the initial path compensation direction to preferentially offset the error deviation with a greater impact.

[0063] Step 3023: If the error direct coupling correlation factor is less than the correlation factor threshold, then the direction opposite to the processing path deviation direction caused by the current dominant error type is directly determined as the initial path compensation direction.

[0064] Optionally, the adaptive control system first performs a judgment on whether the direct coupling correlation factor of the error is less than the correlation factor threshold, that is, it judges the direct coupling correlation factor of the error of each axis. Is it valid? If the result is yes, it indicates that the coupling effect of the auxiliary error (thermal deformation error) on the dominant error (mechanical error) is small. Therefore, it is unnecessary to consider the deviation direction of the auxiliary error; only the deviation of the dominant error needs to be compensated. Thus, the direction opposite to the machining path deviation caused by the current dominant error type is directly determined as the initial path compensation direction. That is, if the dominant error causes the machining path to shift in the direction of decreasing coordinates, the compensation direction is the direction of increasing coordinates; if it shifts in the direction of increasing coordinates, the compensation direction is the direction of decreasing coordinates.

[0065] The embodiments of the present invention first determine whether auxiliary errors need to be considered based on the coupling factor, then select the judgment logic according to whether the deviation directions are the same, and finally determine the direction by combining the error magnitude or dominant attribute, forming a three-dimensional judgment system of coupling degree-direction relationship-error priority, which improves the accuracy of compensation direction judgment, avoids the problem of the more compensation is performed, the greater the deviation is due to the wrong direction, and also expands the scope of scene adaptation.

[0066] In one embodiment, the process of steps 3031-3034 includes: Step 3031: Based on the initial path compensation direction and the functions of the EDM machine tool, identify all feasible compensation operations that can realize the initial path compensation direction to obtain a set of compensation operations.

[0067] Optionally, the adaptive control system first acquires the functions of the EDM machine tool, specifically including independent motion control functions for each axis (linear motion of the X, Y, and Z axes, and angular rotation of the B axis), electrode discharge parameter adjustment functions (discharge current and discharge voltage control), and contact pressure monitoring and feedback functions (real-time monitoring of the contact pressure between the electrode and the workpiece and adjusting the axis motion). Then, based on the determined initial path compensation direction and the acquired functions of the EDM machine tool, it analyzes feasible compensation operations that can achieve compensation in that direction, i.e., specific operations supported by the machine tool functions that can shift the machining path towards the initial compensation direction, combined with the functions of each axis. The correspondence between motion functions and compensation directions is determined, including X-axis + direction compensation: feasible operations include positive linear feed on the X-axis (adjustable single feed amount) and continuous positive feed on the X-axis (adjustable feed speed); Y-axis + direction compensation: feasible operations include positive linear feed on the Y-axis and continuous positive feed on the Y-axis; Z-axis + direction compensation: feasible operations include positive linear feed on the Z-axis (used to adjust the vertical position of the electrode to adapt to directions away from the workpiece) and intermittent positive feed on the Z-axis (to avoid excessive single feed); B-axis + direction compensation: feasible operations include positive angular rotation on the B-axis (adjustable single rotation angle) and continuous positive rotation on the B-axis. Finally, duplicate or overlapping operations (such as non-overlapping similar feed operations on different axes) are removed from all feasible compensation operations to form a set of compensation operations.

[0068] Step 3032: Based on the set of compensation operations and the characteristics of the current dominant error type and the current auxiliary error type, analyze the impact of each compensation operation on other error types during execution to obtain the operation error correlation table.

[0069] Optionally, the adaptive control system first extracts the characteristics of the current dominant error type and the previous auxiliary error type, including mechanical error characteristics: caused by axis control center offset and electrode reference deviation, directly related to axis position accuracy; compensation operations that change the axis position may exacerbate or alleviate this error; thermal deformation error characteristics: caused by changes in ambient temperature leading to axis thermal expansion and contraction, related to changes in axis length; compensation operations that involve long-term axis movement (such as continuous feed) may exacerbate thermal deformation due to axis frictional heat generation; electrode wear error characteristics: caused by discharge loss on the electrode machining surface, related to the discharge gap; compensation operations that change the Z-axis position (adjusting the distance between the electrode and the workpiece) may affect the discharge gap, thereby changing the wear rate. Then, the characteristics of the extracted current dominant error type and current auxiliary error type are analyzed for each operation in the determined set of compensation operations, and their impact on the above three error types during execution is categorized into three types: exacerbation, mitigation, and no impact. Specifically, "exacerbation" refers to an increase in the quantified value of the deviation of a certain error type after the operation is executed (e.g., excessive positive feed in the Z-axis leads to an excessively large discharge gap, exacerbating electrode wear error); "mitigation" refers to a decrease in the quantified value of the deviation of a certain error type after the operation is executed (e.g., positive feed in the X-axis offsets the negative deviation of mechanical error, mitigating mechanical error); and "no effect" refers to no significant change in the quantified value of the deviation of a certain error type after the operation is executed (e.g., B-axis rotation operation has no direct effect on thermal deformation error). Subsequently, an operation error correlation table is constructed using the dimensions of compensation operation-error type-impact result, clearly presenting the correlation between each operation and each error type.

[0070] Step 3033: Based on the operation error association table and the influence range of each type of triggered error, determine and filter whether the influence range of the triggered error type overlaps with the current processing area to obtain a subset of high-association-risk operations; high-association-risk operations refer to compensation operations that overlap.

[0071] Optionally, after obtaining the operation error correlation table, the adaptive control system determines whether the influence range of each type of induced error in the operation error correlation table is amplified by the compensated operation. The induced error type refers to the error type aggravated by the compensated operation in the operation error correlation table, and the influence range refers to the spatial range of the induced error type's impact on the machine tool processing area (e.g., the influence range of thermal deformation error is the effective motion area of ​​each axis, and the influence range of electrode wear error is the discharge area between the electrode and the workpiece). Therefore, in the process of judging and screening whether the influence range of the induced error type overlaps with the current processing area, the current processing area refers to the specific area where the workpiece is being processed at the current moment, determined by the processing path coordinates (e.g., the current processing area coordinate range: X∈[9]). [~12mm], Y∈[7~10mm], Z∈[4~6mm], B∈[29~32°]), and in the judgment and screening process, first determine the type of error caused by the operation and its range of influence, and then determine whether the range of influence of the type of error caused overlaps with the current processing area. The overlap means that the range of influence intersects with the spatial coordinates of the current processing area (e.g., the effective movement area of ​​the X-axis affected by thermal deformation error [0~500mm] overlaps with the current processing area X∈[9~12mm]). If it is determined that there is an overlap, the operation may cause an error in the current processing area and it is included in the high-association-risk operation subset; if the result is no (no overlap), the error caused by the operation does not affect the current processing and is not included in the subset.

[0072] Step 3034: Based on the subset of highly correlated risk operations and the execution frequency of each highly correlated risk operation, determine the risk contribution value of each operation, and determine the combination of compensation operation and error type corresponding to operation whose risk contribution value is greater than the preset contribution threshold as potential risk points with indirect coupling effect, and determine the potential risk value corresponding to the potential risk point based on the potential risk point and the risk contribution value.

[0073] Optionally, the adaptive control system, based on the determined subset of highly correlated risk operations, first determines the execution frequency of each highly correlated risk operation. This execution frequency refers to the number of times or duration a particular highly correlated risk operation needs to be executed in the current processing stage (e.g., operation 2 needs to be executed continuously for 5 minutes in the current processing stage; the execution frequency is expressed as a percentage of duration, ranging from 0% to 100%). Then, based on the execution frequency, it determines the risk contribution value of each operation. This risk contribution value refers to the combined execution frequency of the operation and the severity of the error type it triggers, used to quantify the error risk caused by the operation. The severity is determined based on the weight of the error type's impact on processing accuracy (e.g., mechanical error weight). thermal deformation error Electrode wear error (Pre-stored in the system), its risk contribution value is calculated using the following formula: ;in, For the first Risk contribution value of each highly correlated risk operation; For the first The execution frequency of each operation (expressed as a decimal, e.g., if the execution time accounts for 30%). ); For the first The type of error triggered for each operation weights ( (Including thermal deformation error, electrode wear error, etc.)

[0074] Next, the execution frequency of each highly correlated risk operation (determined based on the complexity of the current processing path and compensation requirements) is substituted into the formula to calculate the risk contribution value, and this risk contribution value is compared with a preset contribution threshold (the preset contribution threshold). Based on the required machining accuracy, exceeding this threshold indicates a higher likelihood of operational risks. A comparison is then made to obtain... If this operation, combined with the corresponding triggered error type, is identified as a potential risk point of indirect coupling effect, then... If the risk is low, the portfolio is not included in the potential risk category.

[0075] Finally, based on the identified potential risk points, their corresponding potential risk values ​​are determined. These potential risk values ​​quantify the impact of the triggered error type on current machining accuracy, machining efficiency, and equipment safety if the error type actually occurs, and range from 0 to 10 (0 indicates no impact, 10 indicates extremely severe impact). Therefore, the impact intensity of the triggered error type is first determined, based on the degree of damage the error type causes to the machining results. For example, collision errors (which would damage both the electrode and the workpiece if they occur) have the highest impact intensity and are pre-stored in the impact intensity table of the adaptive control system (collision errors). Electrode wear error thermal deformation error Next, the importance weight of the current processing step is determined, based on its criticality within the overall workpiece processing flow. Critical steps (such as forming steps with a precision requirement of ±0.005mm) have specific weights. =2, weight of general processes (such as roughing processes). =1, and finally, combining the risk contribution value, the formula for calculating the potential risk value is: ,in, Let i be the potential risk value of the i-th potential risk point; The influence intensity of the error type k corresponding to the i-th potential risk point; This is a normalized coefficient for the importance weight of the process (to ensure that the weight does not excessively amplify the risk value). An upper limit constraint is imposed on the risk value (to avoid the calculated result exceeding 10, ensuring the value is within the range of 0 to 10). Finally, based on each identified potential risk point, the potential risk value is calculated by substituting it into the formula, ultimately forming a correspondence between potential risk points and potential risk values.

[0076] In this embodiment of the invention, by full operation identification (covering all feasible compensation operations), full error correlation analysis (corresponding to dominant and auxiliary errors), regional overlap judgment (focusing on the current processing area), and risk contribution quantification (combining execution frequency and error weight), the coverage of potential risk identification is improved, the processing accuracy deviation caused by the failure to identify hidden risks is avoided, unnecessary risk avoidance operations are reduced, and processing efficiency is improved.

[0077] In one embodiment, the process of steps 3051-3054 includes: Step 3051: Based on the error direct coupling correlation factor, the initial path compensation direction and risk avoidance priority, combined with the processing technology limitations of the EDM machine tool, determine the parameter adjustment range.

[0078] Optionally, the adaptive control system first acquires the machining process constraints of the EDM machine tool, namely, the boundary values ​​of compensation parameters set based on the material properties of the workpiece (e.g., hardness of cemented carbide HRC60), machining accuracy level (IT5 level), and machine tool hardware performance (e.g., minimum feed unit of 0.0001mm for each axis, maximum feed speed of 10mm / min). This includes the minimum allowable value (to avoid undercompensation) and maximum allowable value (to avoid overcompensation) for each axis compensation. Then, it directly couples these values ​​with the error correlation factor (the larger the coupling factor, the more adjustment space needs to be reserved for the compensation amount to cope with the coupling effect), and the initial path compensation direction (each axis + direction, reference...). The adjustment range for each axis parameter is determined by combining the adjustment range (which must be consistent with the direction, i.e., all compensation values ​​are positive) and the risk avoidance priority (axis with high-priority risks need to have their upper limit of the adjustment range reduced to lower the risk). First, the minimum adjustment value is determined: the maximum value of the absolute value of the mechanical error × 0.8 (ensuring at least 80% of the dominant error is offset) and the minimum feed unit of the machine tool. Then, the maximum adjustment value is determined: the minimum of (absolute value of mechanical error + thermal deformation offset) × 1.2 (combined with the coupling factor, reserving a 20% coupling effect margin), the maximum allowable value of the machining process limit, and the risk safety threshold × 0.9 (high-priority risk axes need to be reduced by an additional 10%). Finally, the parameter adjustment range for each axis compensation amount ([minimum adjustment value, maximum adjustment value]) is formed, ensuring that the adjustment range covers both error compensation requirements and meets process limitations and risk avoidance requirements.

[0079] Step 3052: Based on the parameter adjustment range and the error magnitude change trend of the current dominant error type and the current auxiliary error type, determine the candidate path compensation parameter set.

[0080] Optionally, after determining the parameter adjustment range, the adaptive control system combines the error magnitude change trends of the current dominant error type and the current auxiliary error type to finally determine the candidate path compensation parameter group. Specifically, in the process of steps 30521-30524, the error magnitude change trend refers to the error change slope obtained by linear fitting based on the error data (such as the deviation value of mechanical error and thermal deformation error) collected at 5 consecutive processing times. It is divided into an upward trend (slope > 0, error aggravation), a stable trend (slope ≈ 0, error stability), and a downward trend (slope < 0, error mitigation), which is used to determine whether the compensation parameters need to be biased towards the upper or lower limit of the adjustment range.

[0081] Step 3053: Based on the candidate path compensation parameter group and the potential risk points of indirect coupling influence, analyze the degree of influence of the error type corresponding to each potential risk point for each candidate path compensation parameter, and obtain the comprehensive risk influence degree of each candidate path compensation parameter.

[0082] Optionally, after determining the candidate path compensation parameter set, the adaptive control system, in conjunction with the potential risk points of indirect coupling influence, analyzes the impact of the error type corresponding to each potential risk point for each candidate path compensation parameter. In obtaining the comprehensive risk impact of each candidate path compensation parameter, this comprehensive risk impact refers to the quantification of the overall impact of a certain candidate path compensation parameter set on the errors caused by all potential risk points (risk point 1: Z-axis discharge fluctuation, risk point 2: X-axis collision, risk point 3: B-axis surface roughness) during execution. The value range is... (0 represents no impact, 10 represents extreme impact). Therefore, when analyzing the impact of each candidate parameter group on each potential risk point, the deviation ratio between the candidate parameter value and the risk safety threshold is first calculated. The larger the deviation ratio, the greater the impact. The formula is: ;in, The degree of influence of the m-th candidate group on the k-th risk point; Candidate parameter values; The risk safety threshold is used as the basis; then, the average of the impact levels of each risk point is taken to calculate the overall risk impact level, and the formula is as follows: Finally, all candidate path compensation parameters are calculated sequentially to obtain the comprehensive risk impact of each candidate path compensation parameter, forming a correspondence between candidate groups and comprehensive risk values.

[0083] Step 3054: Based on the comprehensive risk impact of each candidate path compensation parameter and the error compensation requirements of the EDM machine tool, evaluate each candidate parameter and select the candidate path compensation parameter with the lowest comprehensive risk impact that can meet the error compensation requirements as the target path compensation parameter.

[0084] Optionally, the adaptive control system first obtains the error compensation requirements of the EDM machine tool. These requirements refer to the minimum compensation effect that the current machining process must meet, i.e., the residual amount of the dominant error after compensation ≤ 0.002 mm (residual mechanical error) and the cumulative amount of auxiliary errors ≤ 0.003 mm (residual amount of thermal deformation + electrode wear), derived from the precision requirements of the workpiece design drawings. Then, considering the comprehensive risk impact of each candidate path compensation parameter, the system evaluates each candidate parameter. This evaluation includes assessing the error compensation requirements, calculating the residual error after compensation for each candidate group, and determining whether the requirements are met (residual amount ≤ threshold); and ranking the comprehensive risk impact, sorting the candidate groups that meet the compensation requirements by comprehensive risk value from smallest to largest, and selecting the candidate group with the lowest risk value. If multiple candidate groups meet the requirements and have the same risk value, the system further compares the "matching degree between compensation amount and error coupling amount" (matching degree = compensation amount / (mechanical error + thermal deformation offset), the closer to 1, the higher the matching degree), and selects the candidate group with the highest matching degree as the target path compensation parameter.

[0085] This invention achieves dynamic optimization of compensation parameters while improving the stability of processing deviations and reducing the probability of risk events by using multi-dimensional parameter range constraints (coupling factor + risk + process), trend-based candidate group generation (based on error change slope), risk quantification analysis (comprehensive risk value calculation), and demand matching screening (residual amount assessment).

[0086] In one embodiment, the process of steps 30521-30524 includes: Step 30521: Based on the current dominant error type and the current auxiliary error type, and combined with the time series data of the error magnitude of the two error types within adjacent preset time periods, calculate the average difference of the error magnitude at adjacent times to obtain the trend of error magnitude change.

[0087] Optionally, the adaptive control system, based on the current dominant error type and the current auxiliary error type, and combining the time-series data of the error magnitudes of the two error types within adjacent preset time periods, calculates the average difference in error magnitudes between adjacent time points. The adjacent preset time period refers to a fixed time interval set for collecting the error magnitude time-series data, determined based on the processing rhythm and error change rate of the EDM machine tool, such as a preset time period of 1 minute (i.e., collecting error data once every 1 minute). The error magnitude time-series data refers to the actual deviation values ​​between the current dominant error type (mechanical error) and the current auxiliary error type (thermal deformation error, electrode wear error), recorded chronologically within adjacent preset time periods. Therefore, in the calculation process, the time-series data is first extracted, obtaining five time-series data points for the dominant error (mechanical error). (Unit: mm / °) and auxiliary error (heat distortion error) Electrode wear error The time series data is used; then, the difference in error magnitude between adjacent time points is calculated. For each type of error, the adjacent difference is calculated by subtracting the previous time point from the subsequent time point data. For example, the adjacent difference for mechanical errors is... , , , (A total of 4 adjacent differences); then calculate the mean of the differences, taking the arithmetic mean of adjacent differences for each type of error, using the formula: ;in, For error type, This embodiment represents the number of time-series data acquisitions. Therefore, the denominator is 4; finally, determine the trend of change based on the mean of the differences. The positive or negative sign determines the trend: if (Set a minimum effective threshold to avoid misjudging minor fluctuations), indicating an upward trend; if If, then it is a downward trend; if If the value is constant, then the trend is stable. Ultimately, the trend of error magnitude variation for each error type is obtained.

[0088] Step 30522: Based on the trend of error magnitude change and the mean difference between the two error types, determine the error magnitude change trend coefficient of the current dominant error type and the error magnitude change trend coefficient of the current auxiliary error type.

[0089] Optionally, in the process of determining the error magnitude change trend coefficients for the current dominant error type and the current auxiliary error type based on the determined error magnitude change trend and the mean difference between the two error types, the adaptive control system uses the error magnitude change trend coefficient as an indicator to quantify the strength of the error type change trend. The value range is 0.8 to 1.2; the closer the coefficient is to 1.2, the stronger the upward trend; the closer it is to 0.8, the stronger the downward trend; a value of 1.0 indicates a stable trend, which is used for subsequent adjustment of the candidate parameter benchmark values. Therefore, the error magnitude change trend coefficient is determined based on the error magnitude change trend and the mean difference. The process is as follows: If it is an upward trend, then ;in, The preset maximum average difference (based on the machine tool error fluctuation limit, in this embodiment) is the maximum average difference value. (Ensure the coefficient does not exceed 1.2); if the trend is downward, then... Ensure the coefficient is not lower than 0.8; if the trend is stable, then... (Since the mean difference is close to 0, the trend is not obviously strong or weak).

[0090] Subsequently, for the current auxiliary error type, the average of the trend coefficients of all auxiliary errors (thermal deformation error, electrode wear error) is taken as the error magnitude change trend coefficient of the current auxiliary error type. To avoid the influence of a single auxiliary error on the deviation; the trend coefficient of the error magnitude of the current dominant error type (mechanical error) is directly denoted as... Ultimately, we obtained... and

[0091] Step 30523: Based on the error magnitude change trend coefficient of the current dominant error type and the error magnitude change trend coefficient of the current auxiliary error type, and combined with the median of the parameter adjustment range, determine the candidate parameter benchmark value.

[0092] Optionally, after determining the error magnitude trend coefficients of the current dominant error type and the current auxiliary error type, the adaptive control system, in conjunction with the median of the parameter adjustment range (i.e., the median of the adjustment range of each axis compensation determined in step 3051), reflects the center position of the adjustment range. In determining the candidate parameter reference values, the combined influence of the dominant and auxiliary errors must also be considered. Therefore, the candidate parameter reference values... The calculation formula is: ;in, The weights for the current dominant error type (set based on the impact of error on machining accuracy, with mechanical error as the dominant error). ); The weights for the current auxiliary error type (comprehensive weights for auxiliary errors). ); This is the coefficient representing the trend of error magnitude change for the current dominant error type; This is the error magnitude trend coefficient for the current auxiliary error type. This formula, through weight allocation, balances the influence of the trends of the dominant and auxiliary errors on the benchmark value, ensuring that the benchmark value closely matches the actual error change requirements. Finally, the candidate parameter benchmark values ​​for each axis are calculated. .

[0093] Step 30524: Divide the parameter adjustment range based on the candidate parameter baseline value to obtain multiple equidistant sub-intervals, and uniformly select a preset number of parameter values ​​in the sub-interval containing the candidate parameter baseline value and the two adjacent sub-intervals to obtain the candidate path compensation parameter group.

[0094] Optionally, the adaptive control system divides the parameter adjustment range according to the determined candidate parameter reference values ​​to obtain multiple equidistant sub-intervals, where each equidistant sub-interval refers to the parameter adjustment range determined in step 3051. Divide the system evenly into several intervals of equal length. For example, the number of intervals can be preset to 5 (balancing coverage and parameter accuracy; 5 intervals can form 4 spaces, ensuring sufficient sub-interval refinement). The length of each sub-interval is... That is, when dividing equidistant subintervals, the subinterval length is used as the basis. from We begin by dividing the data sequentially, resulting in 5 sub-intervals. , , , , Next, locate the sub-interval where the candidate parameter baseline value is located, i.e., determine the value obtained in step 30523. The target sub-interval is identified as the baseline sub-interval. Next, parameter values ​​are selected, with one parameter value pre-selected for each target sub-interval (a total of three sub-intervals: baseline sub-interval + preceding adjacent sub-interval + following adjacent sub-interval). The selected position is the median of each sub-interval (ensuring the parameter value is representative and avoiding selection of interval endpoints). Finally, the parameter groups are combined, with the three parameter values ​​selected for each axis combined in the order of X-axis parameter - Y-axis parameter - Z-axis parameter - B-axis parameter to form three candidate path compensation parameter groups (each group contains compensation amounts for four axes). Ultimately, the candidate path compensation parameter groups are obtained, ensuring that the parameter groups cover the area surrounding the baseline value, balancing compensation needs with risk control.

[0095] This invention, through time-series data trend analysis, trend coefficient quantification, dynamic adjustment of benchmark values, and directional sub-interval selection (focusing on three key sub-intervals surrounding the benchmark value, rather than random selection across the entire range), ensures that the candidate parameter set covers full compensation strength from low to benchmark to high, and highly matches the error change trend. This improves the matching degree of the candidate parameter set to actual error requirements, and enhances the accuracy and representativeness of the candidate parameter set.

[0096] Furthermore, the adaptive control system for the machining path of the EDM machine tool provided by the present invention will be described below. The adaptive control system for the machining path of the EDM machine tool described below can be referred to in correspondence with the adaptive control method for the machining path of the EDM machine tool described above.

[0097] Optional, refer to Figure 2 , Figure 2 This is a schematic diagram of the machining path adaptive control system for an EDM machine tool provided by the present invention. The machining path adaptive control system for the EDM machine tool includes: The path static correction module 210 is used to acquire the initial data of the EDM machine tool and the real-time operating status data during the machining process, and to perform initial correction on the preset theoretical machining path based on the initial offset data in the initial data to obtain the initial machining path; the initial data also includes the initial status data corresponding to the operating status data. The error type determination module 220 is used to determine the real-time dynamic change data based on the running status data and the status data in the initial processing path, and to match the real-time dynamic change data with a preset error type library to obtain the current error type and the corresponding error magnitude; the current error type includes the current dominant error type and the current auxiliary error type. The path dynamic adjustment module 230 is used to perform fuzzy reasoning based on the current error type and error magnitude to obtain the target path compensation parameters, and to adjust the corresponding parameters of the initial processing path based on the path compensation parameters to obtain the target processing path. The execution output module 240 is used to merge the target machining path with the preset NC program template file to obtain the execution NC program adapted to the EDM machine tool, and to perform workpiece machining operations according to the target machining path based on the execution NC program.

[0098] This invention, through acquiring initial data and real-time operational status data of an EDM machine tool, and using initial offset data from the initial data to initially correct the theoretical machining path, obtains an initial machining path. This specifically compensates for inherent errors in the machine tool's mechanical manufacturing process, such as the offset error between the electrode fixture and the axis control center, and the deviation between the electrode's own reference point and the actual machining surface. Real-time change data is obtained by subtracting the operational status data from the initial status data. Combined with preset error types, data matching and fusion are completed to accurately identify the current error type and magnitude introduced during machining due to dynamic factors such as electrode wear and temperature changes, clarifying the priority compensation targets. Subsequently, based on the identified current error... The error type and magnitude are obtained through fuzzy inference to obtain path compensation parameters. The corresponding parameters of the initial machining path are adjusted to realize real-time compensation for the amplification effect of inherent errors and dynamic cumulative errors, forming a target adaptive machining path. Finally, the target adaptive machining path is integrated with the preset NC program template file to obtain the NC program adapted to the machine tool, which drives the machine tool to process according to the target adaptive machining path. This realizes the real-time identification, compensation and execution of dynamic errors from the correction of initial inherent errors to the correction of inherent errors. It effectively solves the technical problem of insufficient machining accuracy caused by machining paths based only on theoretical parameters without considering inherent errors and dynamic cumulative errors. Ultimately, it ensures the machining accuracy of the EDM machine tool and meets the requirements of high precision and high quality machining.

[0099] Please see Figure 3 , Figure 3 An embodiment diagram of an electronic device provided in accordance with the present invention. For example... Figure 3 As shown, this embodiment of the invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, it performs the following steps: The initial data of the electrical discharge machining machine tool and the real-time operating status data during the machining process are acquired. Based on the initial offset data in the initial data, the preset theoretical machining path is initially corrected to obtain the initial machining path. The initial data also includes the initial status data corresponding to the operating status data. Based on the running status data and the status data in the initial processing path, the real-time dynamic change data is determined, and the real-time dynamic change data is matched with the preset error type library to obtain the current error type and the corresponding error magnitude; the current error type includes the current dominant error type and the current auxiliary error type. Based on the current error type and magnitude, fuzzy reasoning is performed to obtain the target path compensation parameters. Based on the path compensation parameters, the corresponding parameters of the initial processing path are adjusted to obtain the target processing path. The target machining path is merged with the preset NC program template file to obtain the NC program adapted to the EDM machine tool, and the workpiece machining operation is performed according to the target machining path based on the NC program.

[0100] Please see Figure 4 , Figure 4 An embodiment diagram of a computer-readable storage medium provided in accordance with an embodiment of the present invention is shown. Figure 4 As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 311 is stored. When the computer program 311 is executed by a processor, it performs the following steps: The initial data of the electrical discharge machining machine tool and the real-time operating status data during the machining process are acquired. Based on the initial offset data in the initial data, the preset theoretical machining path is initially corrected to obtain the initial machining path. The initial data also includes the initial status data corresponding to the operating status data. Based on the running status data and the status data in the initial processing path, the real-time dynamic change data is determined, and the real-time dynamic change data is matched with the preset error type library to obtain the current error type and the corresponding error magnitude; the current error type includes the current dominant error type and the current auxiliary error type. Based on the current error type and magnitude, fuzzy reasoning is performed to obtain the target path compensation parameters. Based on the path compensation parameters, the corresponding parameters of the initial processing path are adjusted to obtain the target processing path. The target machining path is merged with the preset NC program template file to obtain the NC program adapted to the EDM machine tool, and the workpiece machining operation is performed according to the target machining path based on the NC program.

[0101] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the adaptive control method for the machining path of the electrical discharge machining machine tool provided by the above methods, the method including: The initial data of the electrical discharge machining machine tool and the real-time operating status data during the machining process are acquired. Based on the initial offset data in the initial data, the preset theoretical machining path is initially corrected to obtain the initial machining path. The initial data also includes the initial status data corresponding to the operating status data. Based on the running status data and the status data in the initial processing path, the real-time dynamic change data is determined, and the real-time dynamic change data is matched with the preset error type library to obtain the current error type and the corresponding error magnitude; the current error type includes the current dominant error type and the current auxiliary error type. Based on the current error type and magnitude, fuzzy reasoning is performed to obtain the target path compensation parameters. Based on the path compensation parameters, the corresponding parameters of the initial processing path are adjusted to obtain the target processing path. The target machining path is merged with the preset NC program template file to obtain the NC program adapted to the EDM machine tool, and the workpiece machining operation is performed according to the target machining path based on the NC program.

[0102] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0103] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for adaptive control of a machining path of an electrical discharge machine, characterized in that, The method comprises the following steps: acquiring initial data of the electric spark cutting machine tool and running state data in the real-time machining process, and based on initial bias data in the initial data, performing initial correction on a preset theoretical machining path to obtain an initial machining path; the initial data further comprises initial state data corresponding to the running state data; based on the running state data and the state data in the initial machining path, determining real-time dynamic change data, and based on the real-time dynamic change data, matching a preset error type library to obtain a current error type and a corresponding error level; the current error type comprises a current dominant error type and a current auxiliary error type; based on the current error type and the error level, performing fuzzy reasoning to obtain target path compensation parameters, and based on the path compensation parameters, adjusting corresponding parameters of the initial machining path to obtain a target machining path; fusing the target machining path with a preset NC program template file to obtain an execution NC program suitable for the electric spark cutting machine tool, and based on the execution NC program, performing workpiece machining operations according to the target machining path.

2. The adaptive control method of the machining path of an electric spark machine tool according to claim 1, characterized in that, The method comprises the following steps: based on the current dominant error type, the current auxiliary error type and the corresponding error level, determining a physical carrier shared by different error types, and based on the characteristics of the shared physical carrier and the error levels of the error types, constructing an error direct coupling correlation factor; the error direct coupling correlation factor is used to quantify the direct coupling influence degree between different error types caused by sharing the physical carrier; the direct coupling influence refers to that multiple error types share the same physical carrier, resulting in error level superposition or amplification; based on the error direct coupling correlation factor, combining the current dominant error type, the current auxiliary error type and the corresponding error level, analyzing the comprehensive influence trend of different error types on machining path deviation to obtain an initial path compensation direction; based on the initial path compensation direction, combining the characteristics of the current dominant error type and the current auxiliary error type, judging to obtain potential risk points of indirect coupling influence and corresponding potential risk values; the indirect coupling influence refers to that the compensation operation of a certain type of error indirectly triggers a new error which has no direct correlation with the original error; based on the potential risk points of indirect coupling influence and the corresponding potential risk values, sorting different potential risk points to obtain a risk avoidance priority; based on the error direct coupling correlation factor, the initial path compensation direction, the potential risk points of indirect coupling influence and the risk avoidance priority, adjusting the compensation parameters to obtain the target path compensation parameters.

3. The adaptive control method of the machining path of an electrical discharge machine according to claim 2, characterized in that, The method comprises the following steps: based on the error direct coupling correlation factor, the initial path compensation direction, the potential risk points of indirect coupling influence and the risk avoidance priority, adjusting the compensation parameters to obtain the target path compensation parameters, which comprises: based on the error direct coupling correlation factor, the initial path compensation direction and the risk avoidance priority, combining the machining process limitations of the electric spark cutting machine tool to determine the parameter adjustment range; determine a candidate path compensation parameter group based on the parameter adjustment range in combination with a variation trend of an error magnitude of the current dominant error type and the current auxiliary error type; analyze, for each candidate path compensation parameter, an influence degree of each potential risk point corresponding to an error type based on the candidate path compensation parameter group in combination with the potential risk points of the indirect coupling effect, to obtain a comprehensive risk influence degree of each candidate path compensation parameter; evaluate each candidate parameter based on the comprehensive risk influence degree of each candidate path compensation parameter in combination with an error compensation requirement of the electric spark cutting machine tool, and take a candidate path compensation parameter with the lowest comprehensive risk influence degree and capable of meeting the error compensation requirement as the target path compensation parameter.

4. The adaptive control method of the machining path of the electric discharge machine according to claim 2, wherein The determining of the potential risk points and the corresponding potential risk values of the indirect coupling effect based on the initial path compensation direction in combination with the characteristics of the current dominant error type and the current auxiliary error type comprises: identify all feasible compensation operations capable of achieving the initial path compensation direction based on the initial path compensation direction in combination with a function of the electric spark cutting machine tool, to obtain a compensation operation set; analyze, for each compensation operation, an influence on other error types in an execution process based on the compensation operation set in combination with the characteristics of the current dominant error type and the current auxiliary error type, to obtain an operation error association table; determine and filter whether an influence range of an induced error type overlaps with a current machining area based on the operation error association table in combination with the influence range of each induced error type, to obtain a high-correlation risk operation subset; the high-correlation risk operation refers to a compensation operation with an overlap; determine a risk contribution value of each operation based on the high-correlation risk operation subset in combination with an execution frequency of each high-correlation risk operation, and determine a combination of a compensation operation corresponding to an operation with a risk contribution value greater than a preset contribution threshold value and an induced error type as the potential risk points of the indirect coupling effect, and determine a potential risk value corresponding to the potential risk points based on the potential risk points and the risk contribution value.

5. The adaptive control method of the machining path of the electrospark machine tool according to claim 2, characterized in that, The obtaining of the initial path compensation direction comprises: if the error direct coupling association factor is greater than or equal to a preset association factor threshold value, and a machining path deviation direction caused by the current dominant error type and the current auxiliary error type is the same, then a direction opposite to the deviation direction is determined as the initial path compensation direction; if the error direct coupling association factor is greater than or equal to the association factor threshold value, and the machining path deviation direction caused by the current dominant error type and the current auxiliary error type is opposite, then a direction opposite to a deviation direction corresponding to an error type with a greater error magnitude is determined as the initial path compensation direction; if the error direct coupling association factor is less than the association factor threshold value, then a direction opposite to a machining path deviation direction caused by the current dominant error type is directly determined as the initial path compensation direction.

6. The adaptive control method of the machining path of an electrical discharge machine according to claim 3, characterized in that, The determining of the candidate path compensation parameter group based on the parameter adjustment range in combination with the variation trend of the error magnitude of the current dominant error type and the current auxiliary error type comprises: Based on the current dominant error type, the current auxiliary error type, and the error magnitude time series data of the two error types in the adjacent preset time period, the difference value mean of the error magnitude at the adjacent time is calculated to obtain the change trend of the error magnitude; Based on the change trend of the error magnitude and the difference value mean of the two error types, the error magnitude change trend coefficient of the current dominant error type and the error magnitude change trend coefficient of the current auxiliary error type are determined; Based on the error magnitude change trend coefficient of the current dominant error type and the error magnitude change trend coefficient of the current auxiliary error type, and combined with the median of the parameter adjustment range, the candidate parameter reference value is determined; Based on the candidate parameter reference value, the parameter adjustment range is divided to obtain a plurality of equidistant subintervals, and a preset number of parameter values are uniformly selected in the subinterval containing the candidate parameter reference value and the adjacent two subintervals to obtain the candidate path compensation parameter group.

7. The adaptive control method of the machining path of an electrical discharge machine according to claim 1, characterized in that, The matching based on the real-time dynamic change data and the preset error type library to obtain the current error type and the corresponding error magnitude includes: Based on the real-time dynamic change data and the error type library, the error types that have a matching relationship with each real-time dynamic change data in the feature dimension are screened out to obtain a preliminary error type association group; Based on the consistency judgment of the feature change trend of each error type in the preliminary error type association group and the actual change trend of each real-time dynamic change data, an error type trend matching group is obtained; Based on each error type in the error type trend matching group, the influence association relationship of each real-time dynamic change data on the error type is analyzed, and based on the analysis result, the change amplitude of the influence real-time dynamic change data corresponding to each error type is counted. According to the change amplitude, the error types in the error type trend matching group are sorted to obtain an error type influence sorting result; Based on the error type influence sorting result and the preset real-time machining precision requirement, the adaptability of each error type after sorting is evaluated to obtain an error type adaptability classification result; The error type adaptability classification result includes adaptive error types and non-adaptive error types; Based on the error type adaptability classification result, the interaction relationship between the adaptive error types is analyzed, and according to the analysis result, the adaptive error type with the highest influence on the machining precision in the interaction relationship is determined as the current dominant error type, and the remaining adaptive error types are determined as the current auxiliary error type; Based on the change amplitude of the influence real-time dynamic change data corresponding to each error type and the preset error magnitude rule, the error magnitude of the current dominant error type and the error magnitude of the current auxiliary error type are calculated.

8. A machining path adaptive control system of an electric discharge machine, characterized by, The machining path adaptive control method applied to the electric spark cutting machine tool of any one of claims 1 to 7; The machining path adaptive control system of the electric spark cutting machine tool comprises: The path static correction module is configured to acquire initial data of the electric spark cutting machine tool and running state data in a real-time machining process, and correct a preset theoretical machining path based on initial bias data in the initial data to obtain an initial machining path. The initial data further includes initial state data corresponding to the running state data. The error type determination module is configured to determine real-time dynamic change data based on the running state data and the state data in the initial machining path, and match the real-time dynamic change data with a preset error type library to obtain a current error type and a corresponding error magnitude. The current error type includes a current dominant error type and a current auxiliary error type. The path dynamic adjustment module is configured to perform fuzzy reasoning based on the current error type and the error magnitude to obtain target path compensation parameters, and adjust corresponding parameters of the initial machining path based on the path compensation parameters to obtain a target machining path. The execution output module is configured to fuse the target machining path with a preset NC program template file to obtain an execution NC program adapted to the electric spark cutting machine tool, and perform a workpiece machining operation according to the target machining path based on the execution NC program.

9. An electronic device, comprising: The memory is configured to store a computer software program. The processor is configured to read and execute the computer software program. When the processor executes the computer software program, the machining path adaptive control method of the electric spark cutting machine tool according to any one of claims 1 to 7 is implemented. The storage medium stores a computer software program. When the computer software program is executed by the processor, the machining path adaptive control method of the electric spark cutting machine tool according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium, comprising: ​

Images