A method and system for optimizing a wire drawing machine based on big data analysis

Abstract

The present application belongs to the technical field of wire drawing machine, and specifically relates to a wire drawing machine optimization method and system based on big data analysis, which realizes dynamic monitoring and automatic adjustment by collecting real-time image data of the micro-wear morphology of the mold surface and process parameters. First, the image data is subjected to morphology feature extraction to generate wear quantization parameters and calculate three-dimensional topography change values. Next, a dynamic correlation relationship between the process parameters and the topography change values is established, based on which a real-time drift compensation amount is calculated, the compensation amount is added to the process parameter reference value to form a correction instruction sent to the control system. In addition, by comparing the difference between the wire drawing finished product diameter and the standard value, the parameter weight in the correlation relationship is corrected in reverse, the reference value is updated, and the production process is ensured to run continuously and stably. The method improves the consistency and quality stability of the finished product, reduces downtime and material waste, and improves production efficiency and economic benefits.

Description

Technical Field

[0001] This invention belongs to the field of wire drawing machine technology, specifically relating to an optimization method and system for wire drawing machines based on big data analysis. Background Technology

[0002] In the wire drawing process, microscopic wear on the die surface is a significant factor affecting product quality and production efficiency. Current technologies typically address the impact of die wear by periodically inspecting and replacing the die. While this method can ensure production continuity and product quality to some extent, it cannot respond in real-time to changes in the microscopic wear of the die surface. Because the microscopic wear morphology of the wire drawing die surface exhibits random variations, this leads to drift in process parameters (such as speed and pressure), thereby affecting the quality stability of the finished wire drawing product.

[0003] For example, in actual production, operators may need to manually adjust process parameters based on experience or periodic test results. This method is not only inefficient but also difficult to control precisely, which can easily lead to waste of resources and increased costs. Summary of the Invention

[0004] The purpose of this invention is to provide an optimization method and system for wire drawing machines based on big data analysis, which effectively solves the problem of process parameter drift caused by the random changes in the microscopic wear morphology of the wire drawing die surface. By monitoring the wear condition of the die surface in real time and automatically adjusting relevant process parameters, not only can the consistency and quality stability of the finished wire drawing products be significantly improved, but also unnecessary downtime and material waste can be reduced.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an optimization method for a wire drawing machine based on big data analysis, comprising the following steps: Real-time image data of the microscopic wear morphology on the die surface and corresponding process parameters were collected during the wire drawing process; The morphological features of the real-time image data are extracted to generate quantitative parameters of wear morphology, and the three-dimensional morphological change value of the wear on the mold surface is calculated based on the quantitative parameters. Based on the three-dimensional morphology change value, a dynamic correlation between the process parameters and the three-dimensional morphology change value is established, and the real-time drift compensation amount of the process parameters is calculated based on the dynamic correlation. The real-time drift compensation is superimposed on the baseline value of the process parameters to generate a corrected process parameter instruction. The corrected process parameter instruction is then sent to the wire drawing machine control terminal to adjust the wire drawing process parameters of the die. By adjusting the difference between the finished wire diameter and the preset standard value, the parameter weights in the dynamic correlation are corrected in reverse, and the baseline values ​​of the process parameters are updated.

[0006] Preferably, the real-time image data of the micro-wear morphology of the die surface and the corresponding process parameters during the wire drawing process includes: installing a camera above the die during the operation of the wire drawing machine to capture the micro-wear morphology of the die surface; preprocessing the images acquired by the camera, including grayscale conversion and noise filtering, to enhance the contrast of wear features; determining the boundary of the wear area on the die surface using an edge detection method based on the preprocessed image data; and synchronously recording the coordinate information of the wear area boundary and the corresponding process parameters to form a related dataset.

[0007] Preferably, the step of extracting morphological features from the real-time image data to generate quantitative parameters of the wear pattern includes: calculating the area of ​​the wear region based on the coordinate information of the boundary of the wear region; calculating the average brightness using the area of ​​the wear region and the image brightness value obtained by the camera to reflect the overall brightness characteristics of the wear pattern; and defining quantitative parameters based on the area of ​​the wear region and the average brightness to describe the degree of wear on the mold surface.

[0008] Preferably, calculating the three-dimensional morphological change value of mold surface wear based on the quantification parameters includes: firstly, calculating the average depth of the wear area on the mold surface according to the quantification parameters; calculating the wear volume using the average depth and the area of ​​the wear area to describe the overall three-dimensional morphological change of the wear area; calculating the three-dimensional morphological change rate of wear based on the wear volume and the quantification parameters, the change rate being used to measure the change of wear morphology over time or usage; and correlating the three-dimensional morphological change rate with the process parameters recorded during the wire drawing process to form a dataset.

[0009] Preferably, establishing a dynamic correlation between the process parameters and the three-dimensional morphology change values ​​based on the three-dimensional morphology change values ​​includes: firstly, determining the degree of influence of speed and pressure in the process parameters on the wear three-dimensional morphology change based on the three-dimensional morphology change rate, defining an influence factor to quantify the effect of the process parameters on the wear morphology change; calculating an adjustment coefficient using the influence factor and the three-dimensional morphology change rate; establishing a dynamic correlation between the process parameters and the three-dimensional morphology change values ​​based on the adjustment coefficient and combined with the speed and pressure values ​​recorded in the historical dataset; and applying the dynamic correlation to the real-time monitoring system, continuously updating the three-dimensional morphology change rate and the corresponding process parameters to form a dataset for continuously optimizing the working status of the wire drawing machine.

[0010] Preferably, calculating the real-time drift compensation amount of the process parameters based on the dynamic correlation includes: firstly, calculating the deviation value between the current process parameters and the ideal state based on the dynamic correlation; using the deviation value and the three-dimensional morphology change rate to calculate the basic amount of real-time drift compensation, which is used to quantify the degree of process parameter drift caused by wear; based on the basic amount, combined with the speed and pressure in actual operation, determining the specific compensation amount for each process parameter; applying the compensation amount to adjust the actual value of the process parameters to form a new adjustment instruction, and sending the adjustment instruction to the wire drawing machine control system to achieve real-time correction of the process parameters.

[0011] Preferably, the real-time drift compensation is superimposed on the baseline value of the process parameters to generate a corrected process parameter instruction, including: first determining the baseline value of the current process parameters based on the compensation amount; calculating the corrected speed and pressure values ​​using the compensation amount and the baseline value; generating a new process parameter combination instruction based on the corrected speed and pressure values ​​to guide the wire drawing machine in parameter adjustment; sending the process parameter combination instruction to the wire drawing machine control system and recording the adjustment results.

[0012] Preferably, the step of sending the corrected process parameter command to the wire drawing machine control terminal to adjust the wire drawing process parameters of the die includes: firstly, parsing the corrected speed and pressure values ​​according to the process parameter combination command; converting the parsed speed and pressure values ​​into control signals to match the input signal format required by the wire drawing machine control system; using the control signals, sending these signals to the wire drawing machine control terminal through a communication interface to form an adjustment command; after receiving the adjustment command, the wire drawing machine control system updates the current working parameters according to the adjustment command to realize real-time adjustment of the die wire drawing process parameters and record the actual adjustment results.

[0013] Preferably, the parameter weights in the dynamic correlation are reversed by adjusting the difference between the adjusted diameter of the drawn wire and the preset standard value, and the baseline value of the process parameters is updated. This includes: calculating the difference value based on the difference between the actual diameter of the drawn wire and the preset standard value to quantify the deviation between the actual production result and the expected target; adjusting the parameter weights in the dynamic correlation using the difference value and the dynamic correlation; recalculating the baseline value based on the updated parameter weights and the speed and pressure values ​​in the corrected process parameter instructions; and applying the newly calculated baseline value to subsequent process parameter adjustments to form an updated dataset for continuous monitoring and optimization of the drawing machine's operating status.

[0014] On the other hand, this invention proposes an optimization system for wire drawing machines based on big data analysis, comprising: The real-time data acquisition module is used to acquire real-time image data of the micro-wear morphology on the die surface and the corresponding process parameters during the wire drawing process; The wear morphology quantification analysis module is used to extract morphological features from the real-time image data, generate quantification parameters of wear morphology, and calculate the three-dimensional morphological change value of the wear on the mold surface based on the quantification parameters. The dynamic correlation and compensation calculation module is used to establish a dynamic correlation between the process parameters and the three-dimensional morphology change values ​​based on the three-dimensional morphology change values, and to calculate the real-time drift compensation amount of the process parameters based on the dynamic correlation. The control command sending module is used to execute the superposition of the real-time drift compensation amount to the reference value of the process parameters, generate the corrected process parameter command, and send the corrected process parameter command to the wire drawing machine control terminal to adjust the wire drawing process parameters of the die. The parameter update module is used to reverse-correct the parameter weights in the dynamic correlation by adjusting the difference between the adjusted wire drawing diameter and the preset standard value, and update the baseline value of the process parameters.

[0015] Technical effects and advantages of the present invention: The optimization method and system for wire drawing machines based on big data analysis proposed in this invention have the following advantages compared with the prior art: This invention achieves dynamic monitoring of die surface wear and automatic adjustment of process parameters by collecting real-time image data of the microscopic wear morphology on the die surface during the wire drawing process and processing and analyzing the data.

[0016] By extracting morphological features from real-time image data, quantitative parameters of wear morphology are generated, and the three-dimensional morphological change value of mold surface wear is calculated based on these parameters. Subsequently, a dynamic correlation between process parameters and the three-dimensional morphological change value is established, and the real-time drift compensation amount of process parameters is calculated using this relationship. Finally, these compensation amounts are superimposed on the baseline value of process parameters to form a corrected process parameter instruction, which is then sent to the wire drawing machine control system.

[0017] In addition, this method also corrects the parameter weights in the dynamic correlation by comparing the difference between the adjusted wire drawing diameter and the preset standard value, thereby updating the benchmark value of the process parameters and ensuring that the production process can operate continuously and stably. Attached Figure Description

[0018] Figure 1 This is a flowchart of the optimization method for a wire drawing machine based on big data analysis according to the present invention; Figure 2 This is a block diagram of the optimization system for a wire drawing machine based on big data analysis according to the present invention. Detailed Implementation

[0019] 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. The specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention. 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.

[0020] This invention provides, for example Figure 1 The method described here, based on big data analysis, optimizes a wire drawing machine and effectively solves the problem of process parameter drift caused by the random variation of microscopic wear patterns on the surface of the wire drawing die. By monitoring the wear condition of the die surface in real time and automatically adjusting relevant process parameters, not only can the consistency and quality stability of the finished wire drawing be significantly improved, but also unnecessary downtime and material waste can be reduced.

[0021] Preferably, an optimization method for a wire drawing machine based on big data analysis includes the following steps: Step 1: Collect real-time image data of the microscopic wear morphology on the die surface during the wire drawing process, along with the corresponding process parameters, specifically including: During the operation of the wire drawing machine, a camera is installed directly above the mold to ensure that the microscopic wear patterns on the surface of the mold can be captured; Using optical imaging principles, the image acquired by the camera is preprocessed, including grayscale conversion and noise filtering. The formula is: I_new = (I_old - N) / G, where I_old represents the original image brightness value, N represents the noise value, and G is the conversion coefficient for converting the image from color to grayscale to enhance the contrast of the wear features. The formula aims to remove noise from the image and convert the color image to grayscale. By subtracting the noise and dividing by G, a clearer, higher-contrast grayscale image can be obtained, which helps to highlight the wear features.

[0022] Based on the preprocessed image data, the boundary of the wear area on the mold surface is determined by the edge detection method. The calculation formula is: E=sqrt((dI / dx)^2+(dI / dy)^2), where E represents the edge intensity, dI / dx and dI / dy represent the brightness change rate along the x-axis and y-axis directions, respectively. A larger E value indicates that there is a clear edge or boundary here.

[0023] Finally, the coordinate information of the boundary of the wear area and the corresponding process parameters (such as speed V and pressure P) are recorded synchronously to form a related dataset D={E,V,P}, so as to analyze the relationship between the wear of the mold surface and the process parameters in subsequent analysis.

[0024] Step 2: Extract morphological features from the real-time image data to generate quantitative parameters of the wear pattern, specifically including: Based on the coordinate information E of the wear area boundary, the area S of the wear area is calculated using the formula S=∑(x_i-x_(i-1))*(y_i-y_(i-1)), where x_i and y_i represent the coordinates of the i-th point on the boundary. This formula is used to calculate the area of ​​the polygon (i.e., the wear area). The area is calculated by iterating through each point (x_i, y_i) on the boundary and calculating the area of ​​a small rectangle based on the coordinate difference between adjacent points. The areas of all these small rectangles are then summed to obtain the total area S.

[0025] Using the area S of the wear region and the image brightness value I_new obtained by the camera, the average brightness L_avg=(∑I_new) / A is calculated, where A represents the total area of ​​the image, thus reflecting the overall brightness characteristics of the wear pattern. By summing the brightness values ​​of all pixels in the entire wear region and dividing by the total area A, an average value representing the brightness characteristics of the wear region can be obtained.

[0026] Based on the wear area S and the average brightness L_avg, a quantification parameter Q = S * L_avg is defined as a quantitative index to describe the degree of wear on the mold surface. This index reflects the combined influence of the size of the wear area and its brightness characteristics. Q provides a single value to quantify the degree of wear, facilitating subsequent correlation analysis with other process parameters.

[0027] The quantification parameter Q is correlated with process parameters (such as speed V and pressure P) to form a dataset D_1={Q,V,P}, which is used for subsequent analysis of the impact of the wear pattern on the wire drawing process. By correlating the quantification parameter Q with key process parameters in actual operation (such as speed V and pressure P), a direct link between wear state and production process can be established.

[0028] Step 3: Calculate the three-dimensional morphological change value of the mold surface wear based on the quantification parameters, specifically including: Based on the quantization parameter Q, the average depth D_avg of the wear area on the mold surface is first calculated. The formula D_avg = (∑d_i) / N is used, where d_i represents the depth value of the i-th point within the wear area, and N is the number of points within the wear area; this formula is used to calculate the average depth within the wear area. The average depth D_avg is obtained by summing the depth values ​​d_i of all measured points and dividing by the total number of these points N. This helps to understand the overall impact of wear on the mold surface.

[0029] Using the average depth D_avg and the wear area S, the wear volume V_wear=D_avg*S is calculated to describe the overall three-dimensional morphological change of the wear area; this helps to quantify the material loss caused by wear to the mold.

[0030] Based on the wear volume V_wear and the quantization parameter Q, the three-dimensional morphological change rate R_change = V_wear / Q is further calculated. This rate of change measures how the wear morphology changes over time or with the degree of use. The three-dimensional morphological change rate provides a method to compare the proportional relationship between the wear volume and the quantization parameter under different wear levels. This helps to identify the trend and rate of wear development.

[0031] The three-dimensional morphology change rate R_change is associated with the process parameters (such as speed V and pressure P) recorded during the wire drawing process to form a dataset D_2={R_change,V,P}, so as to analyze the need for process parameter adjustment due to the wear three-dimensional morphology change.

[0032] Step 4: Based on the three-dimensional morphology change values, establish a dynamic correlation between the process parameters and the three-dimensional morphology change values, specifically including: Based on the three-dimensional morphology change rate R_change, the influence of the process parameters V and P on the wear morphology changes is first determined. An influence factor IF = R_change / (V*P) is defined to quantify the effect of the process parameters on wear morphology changes; the influence factor is calculated by dividing the three-dimensional morphology change rate R_change by the product of the current velocity V and pressure P. This helps to understand the rate of wear development under given process conditions, thereby identifying which process conditions need to be adjusted to reduce wear.

[0033] Using the influence factor IF and the three-dimensional morphology change rate R_change, the adjustment coefficient AC = IF * R_change is calculated. Here, the adjustment coefficient AC is used to dynamically adjust the process parameters according to the wear changes. This coefficient reflects the amount of process parameter adjustment required based on the current wear condition, and helps to dynamically adjust the speed and pressure in the production process.

[0034] Based on the adjustment coefficient AC, and combined with the speed V and pressure P values ​​recorded in the historical dataset, a dynamic correlation relationship DR=V+P-AC is established between the process parameters (V,P) and the three-dimensional morphology change values, where DR represents the result value of the dynamic correlation relationship, thereby reflecting how the process parameters should be adjusted accordingly under different wear conditions. This relationship aims to dynamically adjust the process parameters so that they can adapt to the constantly changing wear conditions and maintain production efficiency and quality.

[0035] The dynamic correlation DR is applied to the real-time monitoring system. By continuously updating the three-dimensional morphology change rate R_change and the corresponding process parameters, a dataset D_3={DR,R_change,V,P} is formed to continuously optimize the working status of the wire drawing machine.

[0036] Step 5: Calculate the real-time drift compensation amount of the process parameters based on the dynamic correlation, specifically including: Based on the dynamic correlation DR, the deviation value between the current process parameters and the ideal state is first calculated. The deviation value DV is defined as DV = DR - DR_ideal, where DR_ideal represents the dynamic correlation value under the ideal state without wear. By comparing the current dynamic correlation DR with the correlation value DR_ideal under the ideal wear-free state, the degree to which the current process parameters deviate from the ideal state can be quantified. This deviation value DV reflects the changes in process parameters caused by mold wear.

[0037] Using the deviation value DV and the three-dimensional morphology change rate R_change, the basic amount BC for real-time drift compensation is calculated as DV / R_change. Here, the basic amount BC is used to quantify the degree of process parameter drift caused by wear. This formula helps to determine the specific degree of process parameter drift caused by wear, thereby guiding how to perform compensation.

[0038] Based on the aforementioned baseline quantity BC, and combined with the actual speed V and pressure P, the specific compensation amount for each process parameter is further determined. The compensation amounts CP_V = BC * V and CP_P = BC * P correspond to the compensation amounts for speed and pressure, respectively.

[0039] The compensation amounts CP_V and CP_P are applied to adjust the actual values ​​of the process parameters to form a new adjustment instruction AI={V+CP_V,P+CP_P}, and the adjustment instruction is sent to the wire drawing machine control system to achieve real-time correction of the process parameters.

[0040] Step Six: Superimpose the real-time drift compensation amount onto the baseline value of the process parameters to generate the corrected process parameter instructions, specifically including: Based on the process parameter combination instruction PI, the corrected speed value V_new and pressure value P_new in the instruction are first analyzed. By analyzing the instruction, the specific process parameter values ​​that need to be adjusted can be clearly identified, ensuring that the generation and transmission of subsequent control signals can be accurate.

[0041] The parsed V_new and P_new are converted into control signals CS_V=k_V*V_new and CS_P=k_P*P_new, where k_V and k_P are the conversion coefficients corresponding to speed and pressure, respectively, to match the input signal format required by the wire drawing machine control system. Using the control signals CS_V and CS_P, these signals are sent to the wire drawing machine control terminal via a communication interface, forming an adjustment command AC={CS_V,CS_P}. This ensures that the adjustment command can be correctly received and executed by the wire drawing machine control system; through a reliable communication mechanism, seamless integration from data processing to actual equipment control can be achieved, ensuring that the adjustment command can be executed in a timely and accurate manner.

[0042] After receiving the adjustment command AC, the wire drawing machine control system updates the current operating parameters according to the command, realizing real-time adjustment of the die drawing process parameters, and records the actual adjustment result RA={V_current,P_current}, where V_current and P_current represent the actual speed and pressure values ​​after adjustment, respectively. This step ensures closed-loop feedback in the adjustment process, facilitating subsequent analysis and optimization.

[0043] Step 7: Send the revised process parameter instructions to the wire drawing machine control terminal to adjust the wire drawing process parameters of the die, specifically including: Based on the compensation values ​​CP_V and CP_P, the baseline values ​​of the current process parameters are first determined. Let the speed baseline value be V_0 and the pressure baseline value be P_0; the baseline values ​​(V_0, P_0) represent the initial process parameters set under no-wear or ideal conditions. By comparing these baseline values ​​with the real-time drift compensation values ​​(CP_V, CP_P), the specific values ​​that need adjustment can be calculated.

[0044] The corrected speed value V_new = V_0 + CP_V is calculated using the compensation amount CP_V and the reference value V_0. Similarly, the corrected pressure value P_new = P_0 + CP_P is calculated using the compensation amount CP_P and the reference value P_0; these are used to calculate the new values ​​of speed and pressure after compensation. This step, by superimposing the compensation amount onto the reference value, achieves precise adjustment of the process parameters.

[0045] Based on the corrected speed value V_new and pressure value P_new, a new process parameter combination instruction PI={V_new,P_new} is generated. This instruction guides the wire drawing machine to adjust its parameters. The process parameter combination instruction PI includes the compensated new speed value V_new and new pressure value P_new. This instruction is directly transmitted to the control system, instructing it to operate according to the new parameters.

[0046] The process parameter combination command PI is sent to the wire drawing machine control system, and the adjustment result AR={V_new-V_0,P_new-P_0} is recorded for subsequent analysis of the impact of the real-time drift compensation on actual production.

[0047] Step 8: By using the difference between the adjusted finished wire diameter and the preset standard value, the parameter weights in the dynamic correlation are corrected in reverse, and the baseline values ​​of the process parameters are updated. Specifically, this includes: Based on the difference between the actual diameter D_real of the finished wire drawing and the preset standard value D_std, calculate the difference value ΔD = D_real - D_std. This step is used to quantify the deviation between the actual production result and the expected target. The difference value ΔD and the dynamic association relationship DR are used to adjust the parameter weight W_new in the dynamic association relationship. The formula W_new=W_old+K*ΔD is used, where W_old represents the original parameter weight and K is the adjustment coefficient, to reflect the impact of the difference on the parameter weight and help the system dynamically adjust the parameter weight according to the actual situation. Based on the updated parameter weights W_new, and combined with the speed V_new and pressure P_new in the corrected process parameter instruction PI, the baseline values ​​B_V=V_new / W_new and B_P=P_new / W_new are recalculated to ensure that the new baseline values ​​can adapt to the production process requirements under the current wear conditions. By dividing the corrected speed and pressure values ​​by the updated parameter weights, new baseline values ​​that adapt to the current wear conditions can be obtained.

[0048] The newly calculated baseline values ​​B_V and B_P are applied to subsequent process parameter adjustments to form an updated dataset D_4={B_V,B_P,ΔD}, so as to continuously monitor and optimize the working status of the wire drawing machine.

[0049] This method not only increases the level of automation in production but also significantly enhances the system's responsiveness and stability, ensuring optimal production even under conditions of wear and tear. Real-time adjustments and recording allow for dynamic monitoring of process parameter changes, further improving production stability and product quality.

[0050] On the other hand, this invention proposes an optimization system for wire drawing machines based on big data analysis, such as... Figure 2 As shown, it includes: The real-time data acquisition module is used to acquire real-time image data of the micro-wear morphology on the die surface and the corresponding process parameters during the wire drawing process; The wear morphology quantification analysis module is used to extract morphological features from the real-time image data, generate quantification parameters of wear morphology, and calculate the three-dimensional morphological change value of the wear on the mold surface based on the quantification parameters. The dynamic correlation and compensation calculation module is used to establish a dynamic correlation between the process parameters and the three-dimensional morphology change values ​​based on the three-dimensional morphology change values, and to calculate the real-time drift compensation amount of the process parameters based on the dynamic correlation. The control command sending module is used to execute the superposition of the real-time drift compensation amount to the reference value of the process parameters, generate the corrected process parameter command, and send the corrected process parameter command to the wire drawing machine control terminal to adjust the wire drawing process parameters of the die. The parameter update module is used to reverse-correct the parameter weights in the dynamic correlation by adjusting the difference between the adjusted wire drawing diameter and the preset standard value, and update the baseline value of the process parameters.

[0051] In addition, the modules mentioned above are also used to implement other steps of the above-mentioned optimization method for wire drawing machines based on big data analysis, as follows: Suppose a large copper processing plant has introduced a new wire drawing machine to produce copper wires with a diameter of 3.5mm. The plant needs to produce approximately 50 tons of copper wire per day. Continuous production leads to severe wear on the molds, affecting product quality stability. The aforementioned wire drawing machine optimization method based on big data analysis can be used to improve production efficiency and product consistency.

[0052] On-site configuration and data acquisition: Equipment parameter settings: wire drawing speed setting value V0=6m / s, die pressure setting value P0=18MPa.

[0053] A high-resolution industrial camera (2048×2048 pixels) was installed 5cm directly above the mold, with the image acquisition frequency set to once per minute.

[0054] Initial state: The mold surface has no obvious wear, the standard value of the produced copper wire diameter is D_std=3.5mm, and the initial weight of the dynamic correlation relationship is W_old=1.

[0055] Day 1: Running & Building the Basic Dataset Step 1: Image preprocessing and edge detection The image of the current mold surface captured by the camera is then processed. Grayscale conversion and noise reduction, formula: I_new=(I_old-N) / G; Example calculation: Assume that the original brightness of a certain pixel is I_old=180, the noise estimate is N=10, and G=2, then I_new=(180-10) / 2=85.

[0056] Edge detection calculation formula: E=sqrt((dI / dx)^2+(dI / dy)^2); Example calculation: If the brightness change in the x-direction is dI / dx = 20, and the brightness change in the y-direction is dI / dy = 15, then E=sqrt(20^2+15^2)=sqrt(625)=25.

[0057] Generate an associated dataset D={E,V,P}; Record the edge intensity, velocity, and pressure values ​​at each time point, for example: D={(25,6,18),(26,6,18),...,(27,6,18)} Step 2: Quantitative Analysis of Wear Morphology Take one set of data as an example: The area S of the wear zone is calculated using the formula: S = ∑(x_i - x_{i-1}) * (y_i - y_{i-1}); Assuming the boundary coordinates are (0,0), (2,0), (2,1), (0,1), then S=(2-0)*(0-0)+(2-2)*(1-0)+(0-2)*(1-1)+(0-0)*(0-1)=0+0+0+0=2; The average luminance L_avg is calculated using the formula: L_avg=(∑I_new) / A; If the total brightness of the image is 12000, then the total area A = 2048 × 2048 = 4,194,304. L_avg=12000 / 4194304=0.00286.

[0058] Define the wear quantification index Q = S * L_avg; Calculation result: Q = 2 * 0.00286 = 0.00572; Create a new dataset D1={Q,V,P}, such as (0.00572,6,18).

[0059] Step 3: Calculation of 3D morphological change values The average depth D_avg is calculated as follows: assuming the measured depth values ​​of points within the wear area are 0.01mm, 0.012mm, 0.011mm, ... for a total of 10 points, the formula is: D_avg=(∑d_i) / N; D_avg=(0.01+0.012+...+0.011) / 10=0.011mm.

[0060] The wear volume V_wear is calculated using the formula: V_wear = D_avg * S = 0.011 mm * 2 mm² = 0.022 mm³.

[0061] The formula for calculating the rate of change of three-dimensional topography, R_change, is as follows: R_change=V_wear / Q=0.022 / 0.00572=3.845mm³ / unit; Construct the dataset D2={R_change,V,P}.

[0062] Step 4: Establishing Dynamic Relationships The formula for calculating the impact factor (IF) is as follows: IF=R_change / (V*P)=3.845 / (6*18)=3.845 / 108=0.0356; The adjustment factor AC is calculated using the formula: AC = IF * R_change = 0.0356 * 3.845 = 0.137.

[0063] The dynamic correlation DR is calculated using the formula: DR = V + P - AC = 6 + 18 - 0.137 = 23.863.

[0064] Construct the dataset D3={DR,R_change,V,P}.

[0065] Step 5: Calculation of Drift Compensation Amount The deviation value DV is calculated (ideal DR_ideal=24) using the following formula: DV=DR-DR_ideal=23.863-24=-0.137.

[0066] The basic compensation amount BC is calculated using the formula: BC = DV / R_change = -0.137 / 3.845 = -0.0356.

[0067] The formulas for calculating the process parameter compensation amounts CP_V and CP_P are as follows: CP_V = BC * V = -0.0356 * 6 = -0.214; CP_P=BC*P=-0.0356*18=-0.641; The update instruction AI={V+CP_V,P+CP_P}={6-0.214,18-0.641}={5.786,17.359}.

[0068] Step Six: Sending and Executing Control Signals Input the new speed value V_new=5.786 and pressure value P_new=17.359 into the control system; Control signal conversion formula: CS_V=k_V*V_new(k_V=0.8mA / m·s -1 )→CS_V=0.8*5.786=4.629mA.

[0069] CS_P=k_P*P_new(k_P=0.5V / MPa)→CS_P=0.5*17.359=8.68V.

[0070] Control signals are sent to the PLC controller via the RS485 communication interface to update the actual operating parameters.

[0071] Step 7: Complete closed-loop feedback The actual measured finished wire diameter D_real = 3.505 mm; The difference ΔD = D_real - D_std = 3.505 - 3.5 = 0.005 mm; Parameter weight update formula: W_new=W_old+K*ΔD (K=0.2); W_new=1+0.2*0.005=1.001.

[0072] New benchmark value calculation: B_V=V_new / W_new=5.786 / 1.001=5.780m / s; B_P=P_new / W_new=17.359 / 1.001=17.342MPa; Update the dataset D4={B_V,B_P,ΔD}.

[0073] wire drawing speed 6m / s 5.786m / s ↓3.56% Wire drawing pressure 18MPa 17.359MPa ↓3.56% Finished product diameter error 0mm +0.005mm well controlled mold life Replace every 8 hours Replace every 12 hours ↑50% After the system ran continuously for a week, statistics showed that the fluctuation range of copper wire diameter decreased from ±0.02mm to ±0.008mm, the mold replacement frequency decreased by about 40%, and the average energy consumption decreased by 3.2%.

[0074] This real-time wear monitoring and adaptive adjustment method based on big data analysis has effectively improved product quality stability and equipment operating efficiency in actual production, demonstrating its significant engineering application value.

[0075] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An optimization method for a wire drawing machine based on big data analysis, characterized in that, Includes the following steps: Real-time image data of the microscopic wear morphology on the die surface and corresponding process parameters were collected during the wire drawing process; The morphological features of the real-time image data are extracted to generate quantitative parameters of wear morphology, and the three-dimensional morphological change value of the wear on the mold surface is calculated based on the quantitative parameters. Based on the three-dimensional morphology change value, a dynamic correlation between the process parameters and the three-dimensional morphology change value is established, and the real-time drift compensation amount of the process parameters is calculated based on the dynamic correlation. The real-time drift compensation is superimposed on the baseline value of the process parameters to generate a corrected process parameter instruction. The corrected process parameter instruction is then sent to the wire drawing machine control terminal to adjust the wire drawing process parameters of the die. By adjusting the difference between the finished wire diameter and the preset standard value, the parameter weights in the dynamic correlation are corrected in reverse, and the baseline values ​​of the wire drawing process parameters are updated.

2. The optimization method for a wire drawing machine based on big data analysis according to claim 1, characterized in that, The real-time image data of the microscopic wear morphology on the die surface during the wire drawing process and the corresponding process parameters include: During the operation of the wire drawing machine, a camera is installed above the mold to capture the microscopic wear patterns on the surface of the mold; The images acquired by the camera are preprocessed, including grayscale conversion and noise filtering, to enhance the contrast of wear features; Based on the preprocessed image data, the boundary of the wear area on the mold surface is determined using an edge detection method; The coordinate information of the wear area boundary and the corresponding process parameters are recorded synchronously to form an associated dataset.

3. The optimization method for a wire drawing machine based on big data analysis according to claim 2, characterized in that, The step of extracting morphological features from the real-time image data to generate quantified parameters of wear morphology includes: Calculate the area of ​​the wear region based on the coordinate information of the wear region boundary; Using the area of ​​the wear region and the image brightness value obtained by the camera, the average brightness is calculated to reflect the overall brightness characteristics of the wear pattern. Based on the area and average brightness of the wear region, a quantitative parameter is defined to describe the degree of wear on the mold surface.

4. The optimization method for a wire drawing machine based on big data analysis according to claim 3, characterized in that, The three-dimensional morphological change value of mold surface wear is calculated based on the quantification parameters, including: Based on the quantification parameters, the average depth of the wear area on the mold surface is first calculated; The wear volume is calculated using the average depth and the area of ​​the wear region, thereby describing the overall three-dimensional morphological changes of the wear region; Based on the wear volume and the quantification parameters, the three-dimensional morphology change rate of the wear is calculated. The change rate is used to measure the change of wear morphology over time or the degree of use. The three-dimensional morphology change rate is correlated with the process parameters recorded during the wire drawing process to form a dataset.

5. The optimization method for a wire drawing machine based on big data analysis according to claim 4, characterized in that, Based on the three-dimensional morphology change values, a dynamic correlation between the process parameters and the three-dimensional morphology change values ​​is established, including: Based on the three-dimensional morphology change rate, the influence of speed and pressure in the process parameters on the wear three-dimensional morphology change is first determined, and an influence factor is defined to quantify the effect of the process parameters on the wear morphology change. The adjustment coefficient is calculated using the influencing factor and the rate of change of the three-dimensional morphology; Based on the adjustment coefficient, and combined with the speed and pressure values ​​recorded in the historical dataset, a dynamic correlation between the process parameters and the three-dimensional morphology change values ​​is established. The dynamic correlation is applied to the real-time monitoring system. By continuously updating the three-dimensional morphology change rate and the corresponding process parameters, a dataset is formed for continuous optimization of the wire drawing machine's working status.

6. The optimization method for a wire drawing machine based on big data analysis according to claim 5, characterized in that, Calculating the real-time drift compensation amount of the process parameters based on the dynamic correlation includes: Based on the dynamic correlation, the deviation between the current process parameters and the ideal state is first calculated; Using the deviation value and the three-dimensional morphology change rate, the basic quantity of real-time drift compensation is calculated to quantify the degree of process parameter drift caused by wear; Based on the aforementioned basic quantities, and combined with the speed and pressure in actual operation, the specific compensation amount for each process parameter is determined. The compensation amount is applied to adjust the actual value of the process parameter to form a new adjustment instruction, which is then sent to the wire drawing machine control system to achieve real-time correction of the process parameter.

7. The optimization method for a wire drawing machine based on big data analysis according to claim 6, characterized in that, The real-time drift compensation is superimposed on the baseline value of the process parameters to generate a corrected process parameter instruction, including: Based on the compensation amount, the baseline value of the current process parameters is first determined, and the corrected speed and pressure values ​​are calculated using the compensation amount and the baseline value. Based on the corrected speed and pressure values, a new combination of process parameters is generated to guide the wire drawing machine in parameter adjustments. The process parameter combination command is sent to the wire drawing machine control system, and the adjustment results are recorded.

8. The optimization method for a wire drawing machine based on big data analysis according to claim 7, characterized in that, The step of sending the corrected process parameter instructions to the wire drawing machine control terminal to adjust the wire drawing process parameters of the die includes: Based on the process parameter combination instruction, the corrected speed and pressure values ​​are first analyzed; The obtained speed and pressure values ​​are converted into control signals to match the input signal format required by the wire drawing machine control system. Using the aforementioned control signals, these signals are sent to the wire drawing machine control terminal via a communication interface to generate adjustment commands; After receiving the adjustment command, the wire drawing machine control system updates the current working parameters according to the adjustment command, realizes the real-time adjustment of the die wire drawing process parameters, and records the actual adjustment results.

9. The optimization method for a wire drawing machine based on big data analysis according to claim 8, characterized in that, By adjusting the difference between the finished wire diameter and the preset standard value, the parameter weights in the dynamic correlation are corrected in reverse, and the baseline values ​​of the process parameters are updated, including: Based on the difference between the actual diameter of the finished wire drawing product and the preset standard value, the difference value is calculated to quantify the deviation between the actual production results and the expected target. The parameter weights in the dynamic association are adjusted using the difference values ​​and the dynamic association. Based on the updated parameter weights, and combined with the speed and pressure values ​​in the revised process parameter instructions, the baseline values ​​are recalculated. The newly calculated baseline values ​​are applied to subsequent process parameter adjustments to form an updated dataset, enabling continuous monitoring and optimization of the wire drawing machine's operating status.

10. An optimization system for a wire drawing machine based on big data analysis for implementing the method as described in any one of claims 1-9, characterized in that, include: The real-time data acquisition module is used to acquire real-time image data of the micro-wear morphology on the die surface and the corresponding process parameters during the wire drawing process; The wear morphology quantification analysis module is used to extract morphological features from the real-time image data, generate quantification parameters of wear morphology, and calculate the three-dimensional morphological change value of the wear on the mold surface based on the quantification parameters. The dynamic correlation and compensation calculation module is used to establish a dynamic correlation between the process parameters and the three-dimensional morphology change values ​​based on the three-dimensional morphology change values, and to calculate the real-time drift compensation amount of the process parameters based on the dynamic correlation. The control command sending module is used to execute the superposition of the real-time drift compensation amount to the reference value of the process parameters, generate the corrected process parameter command, and send the corrected process parameter command to the wire drawing machine control terminal to adjust the wire drawing process parameters of the die. The parameter update module is used to reverse-correct the parameter weights in the dynamic correlation by adjusting the difference between the adjusted wire drawing diameter and the preset standard value, and update the baseline value of the process parameters.

Images