Adjustable wire cutting machining device and data processing method thereof

Abstract

The application provides an adjustable wire cutting machining device and a data processing method thereof, and relates to the field of wire cutting machining.The adjustable wire cutting machining device comprises a base, a machining table is fixed at the top of the front end of the base, a protection frame is arranged at the outer circle of the machining table and is fixed at the top of the base, and a power control cabinet is arranged at the right side of the protection frame.Through multi-stage data collection of primary, secondary, tertiary and standby collection modules and in combination with data processing of a 3D modeling end, omnibearing monitoring of the machining process is realized, the "difference degree" of each stage is calculated by comparing the actually measured data with the theoretical model data, and iterative and forward-looking process parameter compensation is carried out by using a correction module.This closed-loop self-adaptive control can dynamically correct the cutting path, energy input and cooling condition, effectively compensate the size deviation and thermal deformation generated in the machining process, and thus significantly improve the size precision, geometric shape precision and surface integrity of the final workpiece.

Description

Technical Field

[0001] This invention relates to the field of wire EDM technology, specifically to an adjustable wire EDM apparatus and its data processing method. Background Technology

[0002] Wire EDM (Electrical Discharge Machining) utilizes the principle of non-contact electrical discharge erosion to process ultra-hard, complex-shaped, and thin-walled parts, finding wide application in aerospace, precision molds, medical devices, and electronics. Its core mechanism involves pulsed spark discharge between a continuously moving electrode wire (using molybdenum wire for fast wire EDM and brass wire for slow wire EDM) and the workpiece to achieve cutting. This is complemented by a CNC system, wire feeding mechanism, tensioning mechanism, and cooling / filtration system to complete the machining process. However, as manufacturing demands for machining accuracy, efficiency, and flexibility, the limitations of traditional wire EDM equipment in terms of adjustability are becoming increasingly apparent.

[0003] Dimensional deviations caused by electrode wire wear, tension fluctuations, and table feed errors during processing, as well as geometric distortions caused by uneven discharge energy and wire feed speed variations (such as cone angle deviation during tapered cutting and contour distortion during complex surface machining), cannot be detected and corrected in a timely manner. The final workpiece will exhibit problems such as out-of-tolerance dimensions and non-compliance with geometric tolerances. This is particularly problematic in high-precision mold and aerospace precision parts machining scenarios, where the workpiece pass rate drops significantly, and workpieces may even be scrapped outright. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an adjustable wire EDM processing device that solves the problems of undetectable and uncompensated processing deviations and loss of control over dimensional and geometric accuracy.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solution: an adjustable wire cutting processing device, including a base, a processing table fixed at the top front end of the base, a protective frame provided on the outer ring of the processing table and fixed to the top of the base, a power control cabinet provided on the right side of the protective frame and fixed to the right side of the top of the base, an operating frame fixedly connected to the rear end of the top of the base, and a control component rotating on the outer wall of the right side of the operating frame;

[0006] The control components include a primary acquisition module, a secondary acquisition module, a tertiary acquisition module, a backup acquisition module, a 3D modeling terminal, a data processing module, a correction module, and a control panel.

[0007] Preferably, the primary acquisition module includes a primary dimension acquisition unit, a primary image acquisition unit, and a primary temperature acquisition unit. The primary dimension acquisition unit is used to acquire three-dimensional data of the raw material and three-dimensional data after the first cut. The primary image acquisition unit is used to acquire image data after the first cut. The primary temperature acquisition unit is used to acquire temperature data.

[0008] Preferably, the secondary acquisition module includes a secondary size acquisition unit, a secondary image acquisition unit, and a secondary temperature acquisition unit. The secondary size acquisition unit is used to acquire three-dimensional data after secondary cutting, the secondary image acquisition unit is used to acquire image data after secondary cutting, and the secondary temperature acquisition unit is used to acquire temperature data.

[0009] Preferably, the control logic of the third acquisition module and the backup acquisition module is the same as that of the second acquisition module.

[0010] Preferably, the 3D modeling terminal is used to construct 3D models of image data acquired by the primary acquisition module, secondary acquisition module, tertiary acquisition module, and backup acquisition module, as well as 3D models of theoretically processed parts in the prior art, annotating the coordinates of key feature points, feature vector parameters, and coordinates of defect distribution areas.

[0011] Preferably, the data processing module includes a primary processing unit, a secondary processing unit, a tertiary processing unit, and a backup processing unit. The primary processing unit is used to obtain the size deviation, defect deviation, and temperature deviation through two progressive calculations, and then obtain the difference degree through two fusion calculations. The processing logic of the secondary processing unit, the tertiary processing unit, and the backup processing unit is the same as above, and each processing is based on the deviation trend data of the previous processing to optimize the calculation parameters.

[0012] Preferably, step 1: Dimensional deviation calculation

[0013] The "dual-dimensional quantization + confidence weighting" model is adopted, with two calculations quantifying from the perspectives of "point deviation" and "contour similarity" respectively, and the final fusion yields the size deviation:

[0014] Step 1-1: First Dimensional Deviation Calculation

[0015] The 3D Euclidean distance deviation of a single feature point i: dᵢ 1-1 =√[(Xᵢ 实1 -Xᵢ 理1 ) 2 +(Yᵢ 实1 -Yᵢ 理1 ) 2 +(Zᵢ 实1 -Zᵢ 理1 ) 2(i=1,2,...,k, k is the total number of key feature points, and the Euclidean distance directly reflects the magnitude of spatial point deviation); the average size deviation of all feature points: ΔS 1-1 =(1 / k)×Σ(i=1 to k)dᵢ 1-1 Simultaneously calculate the standard deviation of the Euclidean distance σS 1-1 Used to evaluate the deviation dispersion, σS 1-1 ≤σ S允 , σ S允 This refers to the allowable value for dimensional deviation dispersion.

[0016] Step 1-2: Second Dimensional Deviation Calculation

[0017] Extracting Fourier descriptor feature vectors (F_i) from actual and theoretical contours 实1 =[f 1实 f 2实 , ...,f n实 ], F 理1 =[f 1理 f 2理 , ...,f n理 ]), calculate the cosine similarity of the feature vectors: cosθ1=(F 实1 •F 理1 ) / (|F 实1 |×|F 理1 |); Convert similarity to deviation value: ΔS 1-2 =1-cosθ1; simultaneously calculate the feature vector matching rate η. S1-2 η is used to evaluate the confidence level of the calculation. S1-2 ≥η S允 η S允 This represents the allowable value for the feature vector matching rate.

[0018] Steps 1-3: Dimensional Deviation Fusion

[0019] Based on the weighted confidence scores calculated twice, the confidence score is calculated as follows: Confidence Score = (Standard Deviation / Maximum Allowable Standard Deviation) × 0.5 + (Match Rate / 1) × 0.5: Confidence Score C 1-1 =1-(σS 1-1 / σS max )×0.5+η S1-1 ×0.5, η S1-1 =1-σS 1-1 / σS max , σS max Maximum allowable standard deviation; confidence level C 1-2 =1-(1-η S1-2 )×0.5+(1-ΔS 1-2 )×0.5; weight ω S1-1 =C 1-1 / (C1-1 +C 1-2 ), ω S1-2 =C 1-2 / (C 1-1 +C 1-2 Final dimensional deviation: ΔS1=ω S1-1 ×ΔS 1-1 +ωS 1-2 ×ΔS 1-2 ×S max S max This represents the maximum permissible dimensional deviation.

[0020] Step 2: Defect Deviation Calculation

[0021] Step 2-1: First Defect Deviation Calculation

[0022] ΔD 1-1 =α×|n1-n 1理 |+β×|m1-m 1理 |;Where: n1 is the actual number of defects, n1theoretical is the theoretically allowed number of defects; m1 is the actual defect index, calculated based on the area and depth of pit / convex defects, m1theoretical is the theoretically allowed defect index; α and β are weighting coefficients;

[0023] Step 2-2: Second Defect Deviation Calculation

[0024] Calculate the percentage of defect area: S 缺占1 =S 实缺1 / S 总1 S 实缺1 S represents the total area of ​​the actual defects. 总1 Given the total area of ​​the machined surface; calculate the defect depth gradient deviation: ΔG1=|G 实1 -G 理1 |,G 实1 G represents the average gradient of the actual defect depth. 理1 Theoretically, the mean depth gradient is allowed; ΔD 1-2 =γ×S 缺占1 +δ×ΔG1; where: γ and δ are weighting coefficients;

[0025] Steps 2-3: Defect Deviation Fusion

[0026] Defect type-based weight correction: ΔD1 = (ωD 1-1 ×ΔD 1-1 +ωD 1-2 ×ΔD 1-2 )×K 缺1 Where: ωD 1-1 ,ωD 1-2 K represents the default weight. 缺1 This is a correction factor for the defect type;

[0027] Step 3: Temperature Deviation Calculation

[0028] The thermal impact is quantified from two dimensions: "instantaneous peak value" and "process integral," with the two calculations being complementary.

[0029] Step 3-1: First Temperature Deviation Calculation

[0030] Extracting the highest instantaneous temperature T during the cutting process 峰实1 With respect to the theoretical peak temperature T 峰理1 Comparison: ΔT 1-1 =|T 峰实1 -T 峰理1 |;Simultaneously calculate the peak occurrence time deviation Δt1=|t 峰实1 -t 峰理1 | is used to adjust the confidence level;

[0031] Step 3-2: Second temperature deviation calculation

[0032] Integrate the temperature change curve over the processing time t1 and calculate the actual integral value ∫T. 实1 (t)dt (t∈[0,t1]) and the theoretical integral value ∫T 理1 (t)dt:ΔT 1-2 =|∫T 实1 (t)dt-∫T 理1 (t)dt| / t1;

[0033] Step 3-3: Temperature Deviation Fusion

[0034] Weighting based on peak time deviation: If Δt1≤t 允 (t) 允 (where ω is the allowable value for peak time deviation, preset according to the processing time ratio), then ω T1 -1=ω T标 ω T1-2 =ω T标 (ω) T标 (for standard weights); if Δt1>t 允 , then ω T1-1 =ω T降 ω T1-2 =ω T升 (ω) T降 To reduce the weight, ω T升 To increase the post-weighting, weaken the peak bias, and strengthen the process bias); final temperature deviation: ΔT1=ω T1-1 ×ΔT 1-1 +ω T1-2 ×ΔT 1-2 ;

[0035] Step 4: Difference Fusion. Through two standardization steps to eliminate the influence of dimensions, and combined with dynamic weight adjustments during the processing stage, the final difference score is obtained:

[0036] Step 4-1: First Standardization and Intermediate Difference

[0037] Standardize ΔS1, ΔD1, and ΔT1 to the [0,1] interval: Dimensional deviation standardization: ΔS1' = ΔS1 / S max If ΔS1>S max Then ΔS1'=1; Defect deviation standardization: ΔD1'=ΔD1 / D max If ΔD1>D max Then ΔD1'=1; Temperature deviation standardization: ΔT1'=ΔT1 / T max If ΔT1>T max Then ΔT1'=1; the intermediate difference is calculated using fixed initial weights: D 1-1 =ω1×ΔS1'+ω2×ΔD1'+ω3×ΔT1'; where: ω1, ω2, and ω3 are the initial weights;

[0038] Step 4-2: Second dynamic weight adjustment and final difference

[0039] Adjusting weights based on preceding processed data: Calculating the proportion of each deviation: S 占1 =ΔS1' / (ΔS1'+ΔD1'+ΔT1'), D 占1 =ΔD1' / (ΔS1'+ΔD1'+ΔT1'), T 占1 =ΔT1' / (ΔS1'+ΔD1'+ΔT1'); dynamic weight: ω1'=ω 1基 +ω 1调 ×S 占1 , ω2'=ω 2基 +ω 2调 ×D 占1 ,ω3'=ω 3基 +ω 3调 ×T 占1 (ω) 1基 ω 2基 ω 3基 Based on the weights, ω 1调 ω 2调 ω 3调 To adjust the coefficients, the total weights are kept at 1.0); final variance: D1 = γ × D 1-1 +(1-γ)×(ω1'×ΔS1'+ω2'×ΔD1'+ω3'×ΔT1'); where: γ is the confidence coefficient of the intermediate degree of difference.

[0040] Preferably, the correction module includes a primary correction unit, a secondary correction unit, a tertiary correction unit, and a backup correction unit. Each unit works in accordance with the analysis results of the primary processing unit, the secondary processing unit, the tertiary processing unit, and the backup processing unit, and their core control logic is the same.

[0041] This invention provides an adjustable wire EDM (Electrical Discharge Machining) device. It has the following advantages:

[0042] This invention achieves comprehensive monitoring of the processing process through multi-stage data acquisition via primary, secondary, tertiary, and backup acquisition modules, combined with data processing at the 3D modeling end. By comparing measured data with theoretical model data, the "difference degree" of each stage is calculated, and iterative and forward-looking process parameter compensation (such as dimensional compensation and thermal effect compensation) is performed using a correction module. This closed-loop adaptive control can dynamically correct the cutting path, energy input, and cooling conditions, effectively compensating for dimensional deviations and thermal deformations generated during processing, thereby significantly improving the dimensional accuracy, geometric accuracy, and surface integrity of the final workpiece.

[0043] This invention uses an image acquisition unit to perform quantitative analysis of defects (quantity, index, color intensity), a temperature acquisition unit to monitor thermal effects, and a "thermal effect correction subunit" in the correction module to intelligently adjust cooling strategies and power parameters based on these data (such as defect index m1 and color intensity). This surpasses the traditional passive mode that relies on experience or post-event detection, and reflects technological progress.

[0044] This invention achieves a high degree of automation throughout the entire "acquisition-modeling-processing-correction" process. The system can automatically detect, analyze, and generate optimization instructions for the next stage after each processing stage, reducing the time spent on manual measurement, debugging, and intervention. By preventing defects and reducing rework, the overall processing flow is smoother and more reliable, thereby improving production efficiency and equipment utilization. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the present invention;

[0046] Figure 2 This is a schematic diagram of the structure of the worktable of the present invention;

[0047] Figure 3 This is a schematic diagram of the power control cabinet of the present invention;

[0048] Figure 4 This is a schematic diagram of the first cut of the entire invention;

[0049] Figure 5 This is a schematic diagram of the overall control component controlling the cutting process of the present invention.

[0050] The components include: 1. base; 2. control components; 3. protective frame; 4. operating frame; 5. processing table; and 6. power control cabinet. Detailed Implementation

[0051] 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.

[0052] like Figure 1-5 As shown, this embodiment of the invention provides an adjustable wire cutting processing device, including a base 1, a processing table 5 fixed to the top front end of the base 1, a protective frame 3 arranged around the outer ring of the processing table 5, and the protective frame 3 fixed to the top of the base 1, a power control cabinet 6 arranged on the right side of the protective frame 3, and the power control cabinet 6 fixed to the right side of the top of the base 1, an operating frame 4 fixedly connected to the rear end of the top of the base 1, and a control component 2 rotatably mounted on the outer wall of the right side of the operating frame 4.

[0053] Control unit 2 includes a primary acquisition module, a secondary acquisition module, a tertiary acquisition module, a backup acquisition module, a 3D modeling terminal, a data processing module, a correction module, and a control panel;

[0054] In this embodiment, it should be specifically noted that the power control cabinet 6 is electrically connected to the operating frame 4, the processing table 5, and the actuators thereon via cables, providing them with power and transmitting control signals to ensure the normal operation of the device. The processing table 5 is used to fix the workpiece to be processed, and it can integrate a clamp or positioning mechanism (not shown in the figure), which is a conventional technical means in this field. The protective frame 3 is preferably made of a transparent high-strength material, which can prevent coolant or debris from splashing during the cutting process and facilitate observation of the processing status. This embodiment does not limit this.

[0055] It should be further explained that the power control cabinet 6 adopts conventional power control. The processing table 5 and the operating frame 4 are equipped with control components 2 on their exterior. The control components 2 are used to collect and control the operation data of the processing table 5 and the operating frame 4. This embodiment does not make specific limitations.

[0056] The primary acquisition module includes a primary dimension acquisition unit, a primary image acquisition unit, and a primary temperature acquisition unit. The primary dimension acquisition unit is used to acquire three-dimensional data of the raw material and three-dimensional data after the first cut. The primary image acquisition unit is used to acquire image data after the first cut. The primary temperature acquisition unit is used to acquire temperature data.

[0057] In this embodiment, it should be further explained that the three-dimensional data of the raw material and the three-dimensional data after one cutting are based on the construction of a 3D model, and are respectively denoted as (X...原 ,Y 原 Z 原 ) and (X i实1 ,Y i实1 Z i实1 );

[0058] The image data includes the number of defects, the defect index, the defect area, the defect depth gradient, and the surface color distribution; the temperature data includes instantaneous temperature data and temperature change curve data during the cutting process.

[0059] It should be further explained that the single defect index is derived based on the area and depth of pit-type defects of a single defect (the data is negative if it is a protruding defect). The calculation process is a conventional technical means, and in this embodiment it is denoted as m1, without specific limitation.

[0060] The secondary acquisition module includes a secondary size acquisition unit, a secondary image acquisition unit, and a secondary temperature acquisition unit. The secondary size acquisition unit is used to acquire three-dimensional data after secondary cutting, the secondary image acquisition unit is used to acquire image data after secondary cutting, and the secondary temperature acquisition unit is used to acquire temperature data.

[0061] In this embodiment, it should be further explained that the three-dimensional data after secondary cutting is based on the construction of a 3D model, and is denoted as (X... i实2 ,Y i实2 Z i实2 Secondary image data includes the number of secondary defect data, secondary defect index, and secondary color intensity; secondary temperature data are the temperature change parameters during the cutting process.

[0062] It should be further explained that the secondary defect index is derived based on the area and depth of pit-type defects of a single defect (the data is negative if it is a protruding defect). Its calculation process is a conventional technical means. In this embodiment, it is denoted as m2 and is not specifically limited.

[0063] The control logic of the tertiary acquisition module and the backup acquisition module is the same as that of the secondary acquisition module, and will not be described in detail in this embodiment.

[0064] The 3D modeling terminal is used to construct 3D models of image data acquired by the primary acquisition module, secondary acquisition module, tertiary acquisition module, and backup acquisition module, as well as 3D models of theoretically processed parts based on existing technologies;

[0065] The data processing module includes a primary processing unit, a secondary processing unit, a tertiary processing unit, and a backup processing unit. The primary processing unit is used to compare and analyze the comprehensive data collected by the primary acquisition module with the theoretical model data after primary segmentation, and finally obtain the primary difference degree. The processing logic of the secondary processing unit, the tertiary processing unit, and the backup processing unit is the same as above, and will not be specifically described in this embodiment.

[0066] This embodiment provides an example, specifically illustrating the processing procedure of a single processing unit, as follows:

[0067] Step 1: Dimensional Deviation Calculation (ΔS1) – Two Calculations + Fusion

[0068] The "dual-dimensional quantization + confidence weighting" model is adopted, with two calculations quantifying from the perspectives of "point deviation" and "contour similarity" respectively, and the final fusion yields the size deviation:

[0069] Step 1-1: First Dimensional Deviation Calculation (ΔS) 1-1 (Mean Euclidean distance of key feature points)

[0070] The 3D Euclidean distance deviation of a single feature point i: dᵢ 1-1 =√[(Xᵢ 实1 -Xᵢ 理1 ) 2 +(Yᵢ 实1 -Yᵢ 理1 ) 2 +(Zᵢ 实1 -Zᵢ 理1 ) 2 (i=1,2,...,k, k is the total number of key feature points, and the Euclidean distance directly reflects the magnitude of spatial point deviation); the average size deviation of all feature points: ΔS 1-1 =(1 / k)×Σ(i=1 to k)dᵢ 1-1 Simultaneously calculate the standard deviation of the Euclidean distance σS 1-1 Used to evaluate the deviation dispersion (σS) 1-1 ≤σ S允 , σ S允 (This refers to the allowable value for dimensional deviation dispersion).

[0071] Step 1-2: Second Dimensional Deviation Calculation (ΔS) 1-2 (Feature vector similarity deviation)

[0072] Extracting Fourier descriptor feature vectors (F_i) from actual and theoretical contours 实1 =[f 1实 f 2实 , ...,f n实 ], F 理1 =[f 1理 f 2理 , ...,f n理 ]), calculate the cosine similarity of the feature vectors: cosθ1=(F 实1 •F 理1 ) / (|F 实1 |×|F 理1|); Convert similarity to deviation (the higher the similarity, the smaller the deviation): ΔS 1-2 =1-cosθ1; simultaneously calculate the feature vector matching rate η. S1-2 (Number of matching dimensions / Total number of dimensions), used to evaluate and calculate confidence (η) S1-2 ≥η S允 η S允 (This refers to the allowable value for the feature vector matching rate).

[0073] Steps 1-3: Size deviation fusion (ΔS1)

[0074] Weighted confidence levels based on two calculations (Confidence level = (Standard deviation / Maximum allowable standard deviation) × 0.5 + (Match rate / 1) × 0.5): Confidence level C 1-1 =1-(σS 1-1 / σS max )×0.5+η S1-1 ×0.5 (η) S1-1 =1-σS 1-1 / σS max , σS max (Maximum allowable standard deviation); confidence level C 1-2 =1-(1-η S1-2 )×0.5+(1-ΔS 1-2 )×0.5; weight ω S1-1 =C 1-1 / (C 1-1 +C 1-2 ), ω S1-2 =C 1-2 / (C 1-1 +C 1-2 Final dimensional deviation: ΔS1=ω S1-1 ×ΔS 1-1 +ωS 1-2 ×ΔS 1-2 ×S max (S) max (Using the maximum permissible dimensional deviation, with standardized dimensions).

[0075] Step 2: Defect Deviation Calculation (ΔD1) – Two Calculations + Fusion

[0076] By integrating "basic defect parameters" and "depth feature parameters," the two calculations focus on "quantity-exponential" and "area-gradient" respectively, and finally perform a weighted fusion:

[0077] Step 2-1: First Defect Deviation Calculation (ΔD) 1-1 (Quantitative-exponential linear fusion)

[0078] ΔD 1-1 =α×|n1-n 1理|+β×|m1-m 1理 |; where: n1 is the actual number of defects, n1theoretical is the theoretically allowed number of defects; m1 is the actual defect index (calculated based on the area and depth of pit / convex defects), m1theoretical is the theoretically allowed defect index; α and β are weighting coefficients (which can be adjusted according to material properties).

[0079] Step 2-2: Second Defect Deviation Calculation (ΔD) 1-2 (Area-gradient fusion)

[0080] Calculate the percentage of defect area: S 缺占1 =S 实缺1 / S 总1 (S) 实缺1 S represents the total area of ​​the actual defects. 总1 (Total area of ​​the machined surface); Calculate the defect depth gradient deviation: ΔG1=|G 实1 -G 理1 |(G 实1 G represents the average gradient of the actual defect depth. 理1 (for theoretically permissible depth gradient mean); ΔD 1-2 =γ×S 缺占1 +δ×ΔG1; where: γ and δ are weighting coefficients (which can be adjusted according to the machining accuracy requirements).

[0081] Step 2-3: Defect Deviation Fusion (ΔD1)

[0082] Based on defect type weight adjustment (e.g., the weight of crack-type defects is increased by a preset ratio, while the weight of dent-type defects remains unchanged): ΔD1=(ωD 1-1 ×ΔD 1-1 +ωD 1-2 ×ΔD 1-2 )×K 缺1 Where: ωD 1-1 ,ωD 1-2 The default weights (can be dynamically adjusted based on preceding processed data); K 缺1 This is a defect type correction factor (with a preset value based on the defect type).

[0083] Step 3: Temperature Deviation Calculation (ΔT1) – Two Calculations + Fusion

[0084] The thermal impact is quantified from two dimensions: "instantaneous peak value" and "process integral," with the two calculations being complementary.

[0085] Step 3-1: First temperature deviation calculation (ΔT) 1-1 (Instantaneous peak absolute deviation)

[0086] Extracting the highest instantaneous temperature T during the cutting process 峰实1 With respect to the theoretical peak temperature T 峰理1Comparison: ΔT 1-1 =|T 峰实1 -T 峰理1 |;Simultaneously calculate the peak occurrence time deviation Δt1=|t 峰实1 -t 峰理1 | is used to adjust the confidence level.

[0087] Step 3-2: Second temperature deviation calculation (ΔT) 1-2 (Integral deviation of temperature curve)

[0088] Integrate the temperature change curve over the processing time t1 and calculate the actual integral value ∫T. 实1 (t)dt (t∈[0,t1]) and the theoretical integral value ∫T 理1 (t)dt:ΔT 1-2 =|∫T 实1 (t)dt-∫T 理1 (t)dt| / t1; (Normalize to unit time, unify dimensions)

[0089] Step 3-3: Temperature Deviation Fusion (ΔT1)

[0090] Weighting based on peak time deviation: If Δt1≤t 允 (t) 允 (where ω is the allowable value for peak time deviation, preset according to the processing time ratio), then ω T1 -1=ω T标 ω T1-2 =ω T标 (ω) T标 (for standard weights); if Δt1>t 允 , then ω T1-1 =ω T降 ω T1-2 =ω T升 (ω) T降 To reduce the weight, ω T升 To increase the post-weighting, weaken the peak bias, and strengthen the process bias); final temperature deviation: ΔT1=ω T1-1 ×ΔT 1-1 +ω T1-2 ×ΔT 1-2 .

[0091] Step 4: Difference Fusion (D1) – Two Standardizations + Dynamic Weighting

[0092] By eliminating the influence of dimensions through two standardizations and dynamically adjusting weights during the processing stage, the final difference is obtained:

[0093] Step 4-1: First Standardization and Intermediate Difference (D) 1-1 )

[0094] Standardize ΔS1, ΔD1, and ΔT1 to the [0,1] interval: Dimensional deviation standardization: ΔS1' = ΔS1 / S max (If ΔS1>S) max Then ΔS1'=1); Defect deviation standardization: ΔD1'=ΔD1 / D max (If ΔD1>D) max Then ΔD1'=1); Temperature deviation standardization: ΔT1'=ΔT1 / T max (If ΔT1>T) max Then ΔT1'=1); the intermediate difference is calculated using fixed initial weights: D 1-1 =ω1×ΔS1'+ω2×ΔD1'+ω3×ΔT1'; where: ω1, ω2, and ω3 are the initial weights (set based on processing priority).

[0095] Step 4-2: Second dynamic weight adjustment and final difference (D1)

[0096] Weights are adjusted based on previous processing data (theoretical deviation trend is used when there is no historical data before the first cut, and the deviation percentage of the previous processing is used for the second and subsequent cuts): Calculate the percentage of each deviation: S 占1 =ΔS1' / (ΔS1'+ΔD1'+ΔT1'), D 占1 =ΔD1' / (ΔS1'+ΔD1'+ΔT1'), T 占1 =ΔT1' / (ΔS1'+ΔD1'+ΔT1'); dynamic weight: ω1'=ω 1基 +ω 1调 ×S 占1 , ω2'=ω 2基 +ω 2调 ×D 占1 ,ω3'=ω 3基 +ω 3调 ×T 占1 (ω) 1基 ω 2基 ω 3基 Based on the weights, ω 1调 ω 2调 ω 3调 To adjust the coefficients, the total weights are kept at 1.0); final variance: D1 = γ × D 1-1 +(1-γ)×(ω1'×ΔS1'+ω2'×ΔD1'+ω3'×ΔT1'); where: γ is the confidence coefficient of the intermediate difference (which can be adjusted according to the processing stability).

[0097] The correction module includes a primary correction unit, a secondary correction unit, a tertiary correction unit, and a backup correction unit. Each unit works based on the analysis results of the primary, secondary, tertiary, and backup processing units, respectively. The two correction calculations plus fusion optimization have the same core control logic, and a complete correction calculation is performed after each cut in the three-stage cutting process to provide parameter compensation for the next cut.

[0098] The specific calculation process of the correction unit (taking the first correction unit as an example, the second and third correction units are similar).

[0099] The first correction unit generates the final correction instruction set through two correction calculations based on the first difference degree D1, the deviations of each dimension (ΔS1, ΔD1, ΔT1), and the preceding trend data.

[0100] Step A: First correction calculation (ΔC) 1-1 (Proportional-threshold correction)

[0101] Based on the ratio of the difference to the allowable threshold, the basic correction amount is calculated: Size correction amount ΔC1-S1=k S ×(D1 / D 允 )×ΔS1;Defect correction amount ΔC1-D1=k D ×(D1 / D 允 )×ΔD1;Temperature correction ΔC1-T1=k T ×(D1 / D 允 )×ΔT1; where: k S k D k T To correct the scaling factor (preset according to the accuracy priority of each dimension); D 允 The allowable threshold for difference (set according to the machining accuracy level).

[0102] Step B: Second correction calculation (ΔC) 1-2 (Trend - Forecast Correction)

[0103] Analyze the deviation change trend (first cut is the initial processing, the trend is based on the theoretical deviation gradient; second cut is based on the deviation change rate of the first cut; third cut is based on the deviation change rates of the first and second cuts): Calculate the deviation change rate: r S1 =(ΔS1-ΔS 理1 ) / ΔS 理1 (ΔS) 理1 (for theoretical dimensional deviation); r D1 =(ΔD1-ΔD 理1 ) / ΔD 理1 ;r T1 =(ΔT1-ΔT 理1 ) / ΔT 理1Prediction correction amount: ΔC1-S2=k S '×r S1 ×ΔS1(k S (where k is the size trend correction factor); ΔC1-D2=k D '×r D1 ×Δ D1 (k) D (where ' is the defect trend correction coefficient); ΔC1-T2=k T '×r T1 ×ΔT1(k T (This is a temperature trend correction factor).

[0104] Step C: Correction Quantity Fusion and Instruction Generation

[0105] Weights are assigned based on a priority system (dimensional accuracy has the highest priority, followed by defects, and finally temperature): ω C-S ω C-D ω C-T To adjust the priority weights (summing up to 1.0); the final adjustment for each dimension: ΔC 1-S =ω C-S ×(ΔC1-S1+ΔC1-S2); ΔC 1-D =ω C-D ×(ΔC1-D1+ΔC1-D2); ΔC 1-T =ω C-T ×(ΔC1-T1+ΔC1-T2); Generate a correction instruction set: including cutting path compensation parameters (based on ΔC). 1-S ), discharge power adjustment parameters (based on ΔC) 1-T Cooling flow / temperature adjustment parameters (based on ΔC) 1-T +ΔC 1-D ), Wire feed speed optimization parameters (based on ΔC) 1-S +ΔC 1-D ).

[0106] In this embodiment, it should be further explained that the primary correction unit is used to generate a pre-correction instruction set for the secondary cutting process based on the primary difference, primary image data (including the primary defect data quantity, primary defect index m1, and primary color intensity), and primary temperature change parameters. The pre-correction instructions are used to optimize the cutting path, power input, and cooling to compensate for identified deviations and prevent defects.

[0107] It should be further explained that the primary correction unit includes a dimensional correction subunit, a heat-affected zone correction subunit, and a process parameter integration subunit;

[0108] Size correction subunit based on first-order difference with respect to (X) 原 ,Y 原 Z原 ) and (X i实1 ,Y i实1 Z i实1 The deviation vector of ) is used to calculate and generate compensation parameters for the theoretical path of the secondary cutting. The compensation parameters are adjusted to adjust the start and end coordinates and compensation points of the secondary cutting, denoted as (X original 1, Y original 1, Z original 1). The spatial geometric transformation and error compensation algorithm of its calculation process are conventional technical means in this field, and this embodiment does not make specific limitations.

[0109] The thermal effect correction subunit includes a primary temperature parameter unit, a primary defect index, and a primary color degree. If the index shows that the temperature is too high or the local color is darkened, the cooling command and power input coefficient will change. If the value of the primary defect index m1 changes, the relevant temperature data will increase and slag will accumulate. Conversely, it will decrease and the cutting head will be adjusted.

[0110] The process parameter integration subunit integrates dimensional compensation parameters (ΔX1, ΔY1, ΔZ1) and heat-affected zone adjustment instructions, and combines them with the original three-dimensional model of the raw material to generate an optimized process parameter file that can be directly sent to the secondary cutting equipment. This file synchronously optimizes core parameters such as cutting speed V, laser power W (or other energy source power), and focal point position F.

[0111] The control logic of the secondary correction unit, tertiary correction unit, and backup correction unit is the same as that of the primary correction unit. Based on the processing results of the corresponding stage, they generate correction instructions for the next stage (or backup plan) of the cutting process. Through iterative optimization, the correction module achieves closed-loop adaptive control of the entire multi-step cutting process to improve the dimensional accuracy and surface quality of the final processed part.

[0112] Working principle:

[0113] Preparation steps:

[0114] Construct the three-dimensional dimensional data of the theoretical product, denoted as theoretical three-dimensional data (X). i理 ,Y i理 Z i理The key feature point coordinates are marked (the number of key feature points k is set according to the complexity of the workpiece; for conventional workpieces, k ≥ k_constant; for complex curved surface workpieces, k ≥ k_complex; k_constant and k_complex are preset threshold values ​​for the number of feature points). Based on the theoretical 3D data, the product cutting process is divided into three cutting processes. The theoretical data for each cutting process is constructed, and the cutting threshold for each cutting process is set. The product threshold is set the same for all three cutting processes because they are made of the same material. The cutting threshold for the size part can be adjusted accordingly. This embodiment does not impose specific limitations. The acquisition module is preprocessed and calibrated: the size acquisition unit is calibrated using standard gauge blocks to ensure that the 3D data acquisition error is ≤ ±0.001mm; the image acquisition unit is calibrated using standard defect templates (including pits and cracks of known size and depth).

[0115] Cutting steps:

[0116] Raw materials are conveyed into the processing area of ​​the processing table (5) via a conveyor belt. The primary acquisition module performs a three-dimensional scan of the raw materials to construct a 3D model of the raw materials. The initial single-point three-dimensional data is recorded as (X). i ,Y i Z i Based on the deviation between the theoretical 3D data to be processed and the initial single-point 3D data and the theoretical 3D data, the actual 3D data to be cut is output, denoted as (X). i实 ,Y i实 Z i实 Based on actual 3D data, the initial process parameters for the first cut were set (cutting speed V1=V 粗 Discharge power P1 = P 粗 Cooling flow rate Q1=Q 粗 Wire speed V 丝1 =V 丝粗 Then, a cutting operation is performed. After each cutting operation, the data acquisition module collects the cut data text once. The cutting time is t1 = t 粗 The text of a single cut data includes the dimensions of the single cut data, the defects of the single cut surface, the color of the single cut surface, and the temperature of the single cut.

[0117] After one cut is completed, the acquisition module collects the cut data text:

[0118] A single size acquisition unit acquires the actual three-dimensional coordinates (X, Y, Z) of k key feature points. i实 ,Y i实 Z i实 );

[0119] The number of defects n1 identified by a single image acquisition unit, and the total defect area S 实缺1 Defect depth gradient mean G 实1 Surface color distribution data;

[0120] The highest instantaneous temperature T was recorded by a single temperature acquisition unit. 峰实1 The integral value of the temperature change curve ∫T 实1 (t)dt, peak occurrence time t 峰实1 ;

[0121] The 3D modeling end constructs an "actual model after one cut" based on the collected data and marks the coordinates of the defect distribution area;

[0122] One processing unit performs a deep calculation step to obtain a first difference degree D1:

[0123] Step 1: Calculation of Dimensional Deviation

[0124] First calculation (ΔS) 1-1 Substitute the coordinates of k feature points to calculate the Euclidean distance dᵢ of a single point. 1-1 Summing yields ΔS 1-1 =(1 / k)×Σdᵢ 1-1 Standard deviation σS 1-1 (≤σS) 允 (Confidence level qualified)

[0125] Second calculation (ΔS) 1-2 ): Extract the Fourier descriptor feature vector (n-dimensional), calculate the cosine similarity cosθ1, ΔS 1-2 =1-cosθ1, feature vector matching rate η S1-2 (≥η) S允 (Confidence level qualified)

[0126] Fusion calculation: C is calculated according to the confidence formula. 1-1 C 1-2 Assign weights ω S1-1 ω S1-2 Substituting into the formula, we get ΔS1;

[0127] Step 2: Defect Deviation Calculation

[0128] First calculation (ΔD) 1-1 Substitute n1 and n 1理 m1, m 1理 And α, β, calculate ΔD according to the formula. 1-1 ;

[0129] Second calculation (ΔD) 1-2 ): Calculate S 缺占1 =S 实缺1 / S 总1 (S) 总1 (where G is the machined surface area of ​​the workpiece), ΔG1=|G 实1 -G 理1 Substituting γ and δ, we get ΔD.1-2 ;

[0130] Fusion calculation: Determine K according to defect type 缺1 Substituting into the formula, we get ΔD1;

[0131] Step 3: Temperature Deviation Calculation

[0132] First calculation (ΔT) 1-1 Substitute T 峰实1 T 峰理1 ΔT 1-1 Calculate Δt1=|t 峰实1 -t 峰理1 |;

[0133] Second calculation (ΔT) 1-2 Substitute ∫T 实1 (t)dt、∫T 理1 ΔT is obtained from dt and t1 using the formula. 1-2 ;

[0134] Fusion calculation: based on Δt1 and t 允 The weights of the relationships are assigned and substituted into the formula to obtain ΔT1;

[0135] Step 4: Difference Fusion

[0136] First standardization: Calculate ΔS1', ΔD1', and ΔT1' respectively, and substitute them into the initial weights to obtain D. 1-1 4-2 Second Dynamic Weighting: Calculating S 占1 D 占1 T 占1 Determine the dynamic weights ω1', ω2', and ω3', and substitute them into the formula to obtain D1;

[0137] One correction unit performs two correction calculations to generate the second cutting optimization parameters:

[0138] Step A: Proportional-threshold correction (If D1 ≤ D_allow, proceed with subsequent calculations; if D1 > D_allow, proceed with the next step). 允 (Based on the correction amount, recut) Substitute D1 and D 允 And the deviations in each dimension, calculate ΔC1-S1, ΔC1-D1, and ΔC1-T1 according to the formula;

[0139] Step B: Calculate the rate of change of deviation r for trend-forecast correction. S1 r D1 r T1 Substituting the trend correction coefficients, we obtain ΔC1-S2, ΔC1-D2, and ΔC1-T2;

[0140] Step C: Correction amount fusion and instruction generation according to correction priority weight ω C-S ωC-D ω C-T ΔC is obtained by fusion 1-S ΔC 1-D ΔC 1-T Generate the second cutting optimization parameters (V2=V 粗优 P2=P 粗优 Q2=Q 粗优 V 丝 2=V 丝粗优 Path compensation amount = Δ 补1 );

[0141] Determine D1≤D 允 Then, proceed to the second cutting.

[0142] Based on the optimized parameters after the first correction, a second cutting is performed, with a cutting time t2=t. 半 ;

[0143] The secondary acquisition module collects secondary cutting data text (coordinates of key feature points, defect parameters, and temperature data).

[0144] The 3D modeling unit constructs the "actual model after secondary cutting," and the secondary processing unit performs the detail calculation step to obtain the secondary difference degree D2.

[0145] Dimensions, defects, and temperature deviations are all obtained through two calculations and fusion (ΔS2, ΔD2, ΔT2). The deviation change rate of the first cut is introduced into the calculation process to optimize the confidence weight.

[0146] The difference is obtained by two fusion calculations to obtain D2 (the dynamic weight is adjusted based on the deviation ratio of the first cut);

[0147] The secondary correction unit performs two correction calculations:

[0148] The basic correction is based on D2 and D 阶允 (D) 阶允 =λ1×D 允 The ratio of λ1 to the threshold coefficient of the semi-precision cutting stage is calculated.

[0149] The predicted correction is based on the rate of change of deviation from the first to the second (r). S2 r D2 r T2 )calculate;

[0150] The fusion yields the correction amount, generating the third cutting optimization parameters (V3=V 半优 P3=P 半优 Q3=Q 半优 V 丝3 =V 丝半优 Path compensation amount = Δ 补2 );

[0151] Determine D2≤D 阶允 Then, the third cutting begins.

[0152] Based on the optimized parameters after the second correction, the third cutting is performed, with a cutting time t3=t 精 ;

[0153] The third acquisition module collects the third segmentation data text (coordinates of key feature points, defect parameters, and temperature data).

[0154] The 3D modeling end constructs the "final actual model," and three processing units perform a refinement calculation step to obtain the final difference D3:

[0155] Deviation calculation incorporates deviation trend data from the first and second cuts to optimize the weight allocation between the two calculations;

[0156] The difference is obtained by two fusion calculations to get D3 (dynamic weight is adjusted based on the deviation ratio of the first and second cuts).

[0157] The triple correction unit performs two correction calculations (final accuracy calibration):

[0158] The basic correction is based on D3 and D 精允 (D) 精允 =λ2×D 允 λ2 is the threshold coefficient for the fine cutting stage, and the ratio of λ2 < λ1 is calculated.

[0159] The predicted correction amount is calculated based on the deviation change trend from the first and second cuts to the third cut;

[0160] The final calibration parameters are obtained by fusion, and the cutting path is then fine-tuned.

[0161] After cutting is completed, the final workpiece data is collected, compared with the theoretical product to obtain the degree of completion, the product qualification level is calculated, and the data is stored according to the accuracy level.

[0162] When the difference in any single cut exceeds λ 允 ×D 允 (λ) 允 When the deviation exceeds the threshold proportional coefficient, or when the deviation change rate of two consecutive cuts is positive (the deviation continues to expand), the backup acquisition module will be automatically activated.

[0163] The acquisition module synchronously acquires size, image, and temperature data for the current cutting stage, with an acquisition frequency of λfrequency × main module acquisition frequency (λ). 频 >1);

[0164] The standby processing unit independently performs the in-depth calculation steps to obtain the standby difference degree D. 备 ;

[0165] Compare the confidence levels of the primary and backup differences (based on parameters such as standard deviation and matching rate), and take the result with the higher confidence level as the final difference.

[0166] The correction module recalculates the correction amount based on the preferred degree of difference, generates emergency processing parameters, and initiates a re-cutting or supplementary cutting process until the deviation meets the threshold requirement.

[0167] In addition, the power control cabinet 6 supplies power to the entire system and controls the current, voltage, and motion parameters during the wire cutting process. The cutting path, speed, and pulse parameters can be set via the control unit 2 to achieve high-precision cutting. The rotatable design of the control unit 2 and the programmable control of processing parameters adapt to different workpieces and process requirements, improving operational flexibility and processing efficiency. The processing table 5 is located at the front of the base 1 and is used to place the workpiece to be processed. The protective frame 3 surrounds the processing table 5, providing protection against coolant or debris splashing during cutting. The operating frame 4 is located at the rear of the base 1 and is equipped with the rotatable control unit 2, allowing the operator to observe and adjust processing parameters from different angles.

[0168] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An adjustable wire cutting processing device, comprising a base (1), characterized in that: A processing table (5) is fixed to the top front end of the base (1). A protective frame (3) is provided on the outer ring of the processing table (5), and the protective frame (3) is fixed to the top of the base (1). A power control cabinet (6) is provided on the right side of the protective frame (3), and the power control cabinet (6) is fixed to the right side of the top of the base (1). An operating frame (4) is fixedly connected to the rear end of the top of the base (1). A control component (2) is rotatably provided on the right outer wall of the operating frame (4). The control unit (2) includes a primary acquisition module, a secondary acquisition module, a tertiary acquisition module, a backup acquisition module, a 3D modeling terminal, a data processing module, a correction module, and a control panel; The 3D modeling terminal is used to construct 3D models of image data acquired by the primary acquisition module, secondary acquisition module, tertiary acquisition module, and backup acquisition module, as well as 3D models of theoretically processed parts in the prior art, annotating the coordinates of key feature points, feature vector parameters, and coordinates of defect distribution areas; The data processing module includes a primary processing unit, a secondary processing unit, a tertiary processing unit, and a backup processing unit. The primary processing unit is used to obtain the size deviation, defect deviation, and temperature deviation through two progressive calculations, and then obtain the difference degree through two fusion calculations. The secondary processing unit, tertiary processing unit, and backup processing unit have the same processing method as the primary processing unit, and each processing is based on the deviation trend data of the previous processing to optimize the calculation parameters.

2. The adjustable wire EDM processing device according to claim 1, characterized in that: The primary acquisition module includes a primary dimension acquisition unit, a primary image acquisition unit, and a primary temperature acquisition unit. The primary dimension acquisition unit is used to acquire three-dimensional data of the raw material and three-dimensional data after the first cut. The primary image acquisition unit is used to acquire image data after the first cut. The primary temperature acquisition unit is used to acquire temperature data.

3. The adjustable wire EDM processing device according to claim 1, characterized in that: The secondary acquisition module includes a secondary size acquisition unit, a secondary image acquisition unit, and a secondary temperature acquisition unit. The secondary size acquisition unit is used to acquire three-dimensional data after secondary cutting, the secondary image acquisition unit is used to acquire image data after secondary cutting, and the secondary temperature acquisition unit is used to acquire temperature data.

4. The adjustable wire EDM processing device according to claim 1, characterized in that: The processing methods of the tertiary acquisition module and the backup acquisition module are the same as those of the secondary acquisition module.

5. An adjustable wire EDM machining data processing method, characterized in that, include: Step 1: Calculation of Dimensional Deviation A "dual-dimensional quantization + confidence weighting" model is adopted, with two calculations quantifying from the perspectives of "point deviation" and "contour similarity" respectively, and the final fusion yields the size deviation: Step 1-1: First Dimensional Deviation Calculation The 3D Euclidean distance deviation of a single feature point i: dᵢ 1-1 =√[(Xᵢ 实1 -Xᵢ 理1 ) 2 +(Yᵢ 实1 -Yᵢ 理1 ) 2 +(Zᵢ 实1 -Zᵢ 理1 ) 2 (i=1,2,...,k, k is the total number of key feature points, and the Euclidean distance directly reflects the magnitude of spatial point deviation); the average size deviation of all feature points: ΔS 1-1 =(1 / k)×Σ(i=1 to k)dᵢ 1-1 Simultaneously calculate the standard deviation of the Euclidean distance σS 1-1 Used to evaluate the deviation dispersion, σS 1-1 ≤σ S允 , σ S允 This refers to the allowable value for dimensional deviation dispersion. Step 1-2: Second Dimensional Deviation Calculation Extracting Fourier descriptor feature vectors (F_i) from actual and theoretical contours 实1 =[f 1实 f 2实 , ...,f n实 ], F 理1 =[f 1理 f 2理 , ...,f n理 ]), calculate the cosine similarity of the feature vectors: cosθ1=(F 实1 •F 理1 ) / (|F 实1 |×|F 理1 |); Convert similarity to deviation value: ΔS 1-2 =1-cosθ1; simultaneously calculate the feature vector matching rate η. S1-2 η is used to evaluate the confidence level of the calculation. S1-2 ≥η S允 η S允 This represents the allowable value for the feature vector matching rate. Steps 1-3: Dimensional Deviation Fusion Based on the weighted confidence scores calculated twice, the confidence score is calculated as follows: Confidence Score = (Standard Deviation / Maximum Allowable Standard Deviation) × 0.5 + (Match Rate / 1) × 0.5: Confidence Score C 1-1 =1-(σS 1-1 / σS max )×0.5+η S1-1 ×0.5, η S1-1 =1-σS 1-1 / σS max , σS max Maximum allowable standard deviation; confidence level C 1-2 =1-(1-η S1-2 )×0.5+(1-ΔS 1-2 )×0.5; weight ω S1-1 =C 1-1 / (C 1-1 +C 1-2 ), ω S1-2 =C 1-2 / (C 1-1 +C 1-2 Final dimensional deviation: ΔS1=ω S1-1 ×ΔS 1-1 +ωS 1-2 ×ΔS 1-2 ×S max S max This represents the maximum permissible dimensional deviation. Step 2: Defect Deviation Calculation Step 2-1: First Defect Deviation Calculation ΔD 1-1 =α×|n1-n 1理 |+β×|m1-m 1理 |;Where: n1 is the actual number of defects, n1theoretical is the theoretically allowed number of defects; m1 is the actual defect index, calculated based on the area and depth of pit / convex defects, m1theoretical is the theoretically allowed defect index; α and β are weighting coefficients; Step 2-2: Second Defect Deviation Calculation Calculate the percentage of defect area: S 缺占1 =S 实缺1 / S 总1 S 实缺1 S represents the total area of ​​the actual defects. 总1 Given the total area of ​​the machined surface; calculate the defect depth gradient deviation: ΔG1=|G 实1 -G 理1 |,G 实1 G represents the average gradient of the actual defect depth. 理1 Theoretically, the mean depth gradient is allowed; ΔD 1-2 =γ×S 缺占1 +δ×ΔG1; where: γ and δ are weighting coefficients; Steps 2-3: Defect Deviation Fusion Defect type-based weight correction: ΔD1 = (ωD 1-1 ×ΔD 1-1 +ωD 1-2 ×ΔD 1-2 )×K 缺1 Where: ωD 1-1 ,ωD 1-2 K represents the default weight. 缺1 This is a correction factor for the defect type. Step 3: Temperature Deviation Calculation The thermal impact is quantified from two dimensions: "instantaneous peak value" and "process integral," with the two calculations being complementary. Step 3-1: First Temperature Deviation Calculation Extracting the highest instantaneous temperature T during the cutting process 峰实1 With respect to the theoretical peak temperature T 峰理1 Comparison: ΔT 1-1 =|T 峰实1 -T 峰理1 |;Simultaneously calculate the peak occurrence time deviation Δt1=|t 峰实1 -t 峰理1 | is used to adjust the confidence level; Step 3-2: Second temperature deviation calculation Integrate the temperature change curve over the processing time t1 and calculate the actual integral value ∫T. 实1 (t)dt (t∈[0,t1]) and the theoretical integral value ∫T 理1 (t)dt:ΔT 1-2 =|∫T 实1 (t)dt-∫T 理1 (t)dt| / t1; Step 3-3: Temperature Deviation Fusion Weighting based on peak time deviation: If Δt1≤t 允 (t) 允 (where ω is the allowable value for peak time deviation, preset according to the processing time ratio), then ω T1 -1=ω T标 ω T1-2 =ω T标 (ω) T标 (for standard weights); if Δt1>t 允 , then ω T1-1 =ω T降 ω T1-2 =ω T升 (ω) T降 To reduce the weight, ω T升 To increase the post-weighting, weaken the peak bias, and strengthen the process bias); final temperature deviation: ΔT1=ω T1-1 ×ΔT 1-1 +ω T1-2 ×ΔT 1-2 ; Step 4: Difference Fusion. Through two standardization steps to eliminate the influence of dimensions, and combined with dynamic weight adjustments during the processing stage, the final difference score is obtained: Step 4-1: First Standardization and Intermediate Difference Standardize ΔS1, ΔD1, and ΔT1 to the [0,1] interval: Dimensional deviation standardization: ΔS1' = ΔS1 / S max If ΔS1>S max Then ΔS1'=1; Defect deviation standardization: ΔD1'=ΔD1 / D max If ΔD1>D max Then ΔD1'=1; Temperature deviation standardization: ΔT1'=ΔT1 / T max If ΔT1>T max Then ΔT1'=1; the intermediate difference is calculated using fixed initial weights: D 1-1 =ω1×ΔS1'+ω2×ΔD1'+ω3×ΔT1'; where: ω1, ω2, and ω3 are the initial weights; Step 4-2: Second dynamic weight adjustment and final difference Adjusting weights based on preceding processed data: Calculating the proportion of each deviation: S 占1 =ΔS1' / (ΔS1'+ΔD1'+ΔT1'), D 占1 =ΔD1' / (ΔS1'+ΔD1'+ΔT1'), T 占1 =ΔT1' / (ΔS1'+ΔD1'+ΔT1'); dynamic weight: ω1'=ω 1基 +ω 1调 ×S 占1 , ω2'=ω 2基 +ω 2调 ×D 占1 ,ω3'=ω 3基 +ω 3调 ×T 占1 (ω) 1基 ω 2基 ω 3基 Based on the weights, ω 1调 ω 2调 ω 3调 To adjust the coefficients, the total weights are kept at 1.0); final variance: D1 = γ × D 1-1 +(1-γ)×(ω1'×ΔS1'+ω2'×ΔD1'+ω3'×ΔT1'); where: γ is the confidence coefficient of the intermediate degree of difference.

6. The adjustable wire EDM processing device according to claim 1, characterized in that: The correction module includes a primary correction unit, a secondary correction unit, a tertiary correction unit, and a backup correction unit. Each unit works according to the analysis results of the primary processing unit, the secondary processing unit, the tertiary processing unit, and the backup processing unit, and their core control logic is the same.

Images