Method for reducing surface copper powder in copper rod rolling process

Abstract

The invention relates to the technical field of copper material preparation, in particular to a method for reducing surface copper powder in the rolling process of a copper rod, which comprises the following steps: dividing the rolling process into a plurality of process sections, and supplying different lubricants to each process section; establishing a correlation model of the process parameters and the copper rod performance, and generating a process parameter feasible region according to the target performance of the copper rod; roller surface parameters of all the process sections are monitored respectively, and the surface topography variation is calculated; when the surface topography variation is larger than a set value, rolling process parameters are adjusted; monitoring the surface topography variation corresponding to the adjusted rolling process parameters, and comparing the surface topography variation with a set value; the surface topography variation is smaller than a set value; and when the rolling process parameters are adjusted to the feasible region boundary and the surface topography variation is greater than a set value, alarming. According to the method, a whole-process closed-loop control system of process segmentation optimization-dynamic monitoring adjustment-rapid emergency response is constructed, fundamental control over the copper powder problem is achieved, and the rolling process is better promoted to be transformed to be intelligent and refined.

Description

Technical Field

[0001] The present invention relates to the technical field of copper material preparation, and particularly to a method for reducing surface copper powder during copper rod rolling. Background Art

[0002] Copper rod rolling is a pressure processing process in which a copper billet is passed through rolls with a certain pass profile and its cross-section is reduced and its length is increased after continuous extrusion by the rolls. During the process, a large frictional force will be generated in the contact area between the copper rod and the rolls. Therefore, it is necessary to lubricate and cool the copper rod and the rolls in the rolling mill.

[0003] Currently, the commonly used lubricant is emulsified oil. However, in application, since the rolling of copper rods involves multiple passes, a single lubricant is difficult to adapt to the frictional requirements of different rolling stages, and often due to excessive friction or insufficient lubrication, the generation amount of copper powder increases. Especially in the rough rolling section and the finish rolling section, due to the large reduction, high rolling speed, high frictional force and temperature, it is easy to cause damage or local melting on the surface of the copper rod, thus forming copper powder, which not only affects the surface quality of the copper rod, but also increases the production cost.

[0004] Moreover, the amount of lubricant used is not the more the better. In the rough rolling section, due to the large reduction (about 40%) and high contact pressure, the lubricant is easily extruded, resulting in direct contact between metals and a sharp increase in the amount of copper powder; while in the finish rolling section during high-speed rolling, the residue of high-viscosity lubricant increases the surface roughness (Ra > 1.6 μm), making the lubricant unable to flexibly adapt to the requirements of different process sections. Some lubricants are prone to failure at high temperatures, resulting in reduced lubrication effect.

[0005] Some technologies optimize the surface quality of copper rods by adjusting rolling parameters, thereby reducing the generation of copper powder. For example, by reducing the rolling speed and reduction to reduce friction and lower the surface temperature. However, excessive adjustment of rolling parameters may lead to a reduction in production efficiency and even affect the final quality and specifications of the copper rod. And the coating on the roll surface deteriorates continuously during rolling, making the equivalent stress on the copper rod increase continuously. The parameter adjustment of the existing technologies lags behind the change of the actual working conditions, and it is difficult to avoid the risk of excessive copper powder under extreme working conditions. In addition, the tiny changes in the surface topography are often difficult to predict, and simply relying on the adjustment of rolling parameters may not be able to completely solve the copper powder problem.

[0006] Therefore, how to flexibly adjust process parameters and the use of lubricants according to the characteristics of different process sections has become the key to further reducing the generation of copper powder, improving the surface quality of copper rods and production efficiency.

[0007] The information disclosed in this background art section is only intended to deepen the understanding of the overall background art of the present invention, and should not be regarded as an admission or any form of implication that this information constitutes the prior art known to those skilled in the art. Summary of the Invention

[0008] The first object of the present invention is to provide a method for reducing surface copper powder during copper rod rolling, which flexibly adjusts process parameters and the use of lubricants according to the characteristics of different process sections, and significantly reduces the amount of copper powder generated through process segmentation and differential supply of lubricants.

[0009] The above technical object of the present invention is achieved through the following technical solutions:

[0010] A method for reducing surface copper powder during copper rod rolling, comprising the following steps:

[0011] Divide the rolling process into multiple process sections and supply different lubricants to each process section;

[0012] Establish a correlation model between process parameters and the properties of the copper rod, and generate a feasible region of process parameters according to the target properties of the copper rod;

[0013] Monitor the surface parameters of the rolls in each process section respectively and calculate the change amount of the surface topography;

[0014] When the change amount of the surface topography is greater than the set value, adjust the rolling process parameters;

[0015] Monitor the change amount of the surface topography corresponding to the adjusted rolling process parameters and compare it with the set value; until the change amount of the surface topography is less than the set value;

[0016] When the rolling process parameters are adjusted to the boundary of the feasible region and the change amount of the surface topography is greater than the set value, give an alarm.

[0017] Furthermore, divide the rolling process into a rough rolling section, a finish rolling section, and a sizing section along the rolling direction, and the reduction ratios of the rough rolling section, the finish rolling section, and the sizing section gradually decrease.

[0018] Specifically, the reduction ratio of the rough rolling section is 35% - 45%, the reduction ratio of the finish rolling section is 15% - 25%, and the reduction ratio of the sizing section is 3% - 8%.

[0019] Preferably, the lubricant supply amount is set differently according to the process section: in the rough rolling section, use a sulfur-containing extreme pressure lubricant with a viscosity of 120 - 150 cSt and a supply amount ≥ 20 L / min; in the finish rolling section, use a low-viscosity mineral oil with a viscosity of 40 - 60 cSt and a supply amount of 10 - 15 L / min; in the sizing section, use a water-based emulsion with a concentration of 8% - 12% and a supply amount of 5 - 8 L / min;

[0020] As a preference, establishing a correlation model between process parameters and the properties of the copper rod includes:

[0021] (1) Extract the historical process parameter dataset from the production database, including rolling speed, rolling force, reduction ratio, lubricant supply, initial roll roughness, as well as the corresponding copper powder amount, tensile strength, and elongation rate on the surface of the copper rod;

[0022] When cleaning the data, exclude the following abnormal records: the section where the instantaneous fluctuation of the rolling force exceeds ±10% and lasts for more than 5 seconds; invalid data where the measured value of the surface copper powder amount exceeds the upper limit of the sensor range; production data during the overload alarm of the main motor of the rolling mill;

[0023] (2) For records with a process parameter missing rate less than 20%, after passing the time series continuity verification, use the cubic spline interpolation method to complete the data;

[0024] The time series continuity verification requires simultaneously meeting the following conditions: (a) the rolling speed fluctuation within 10 seconds before and after the missing data segment ≤ ±3%; (b) the rolling force change rate ≤ 50 kN / s; (c) the reduction ratio adjustment interval ≥ 30 seconds.

[0025] Based on the extreme values of the process parameters, use the rolling power formula P = F × v and the minimum lubrication amount constraint Q ≥ 0.1F to construct boundary virtual data points and expand the coverage of the dataset;

[0026] (3) Based on the expanded dataset, use the random forest algorithm to construct a relationship model between the process parameters and the performance of the copper rod;

[0027] (4) Incorporate prior knowledge constraints, eliminate interfering variables that violate physical laws through feature importance ranking, and then locally correct the abnormal points in the random forest prediction results that conflict with the knowledge constraints.

[0028] Furthermore, generate the process parameter feasible region according to the target performance of the copper rod, including:

[0029] Set the target performance constraint conditions for the copper rod;

[0030] Based on the correlation model, screen the set of process parameter combinations that meet all constraints to form a high-dimensional feasible region sample set;

[0031] Cluster the feasible region sample set, eliminate discrete noise points, and use the convex hull of the largest cluster as the initial feasible region;

[0032] Fit the feasible region boundary function, and at a 95% confidence level, output the parameter safe operation interval where all parameter combinations simultaneously meet the target performance requirements to obtain the process parameter feasible region.

[0033] Specifically, in the human-machine interface, the allowed operation range is marked in the green area. When the parameter point enters within 5% of the feasible region boundary, the interface triggers a red warning area and displays a list of adjustment suggestions, including preferentially increasing the lubricant amount or reducing the rolling speed.

[0034] Furthermore, the roll surface parameters include groove depth, surface roughness, and oil storage rate, and the surface topography change amount is the weighted deviation value of the groove depth, surface roughness, and oil storage rate from the initial design values.

[0035] Furthermore, the calculation formula for the surface topography change amount is:

[0036]

[0037] where ΔK is the surface topography change amount, d is the groove depth, Ra is the surface roughness, η is the oil storage rate, d0 is the initial design value of the groove depth, Ra0 is the initial design value of the surface roughness, η0 is the initial design value of the oil storage rate, and α, β, and γ are weight coefficients, and the weight coefficients α, β, and γ are dynamically adjusted according to the rolling stage.

[0038] Furthermore, the values of the weight coefficients α, β, and γ are:

[0039] Rough rolling stage: α = 0.6, β = 0.2, γ = 0.2;

[0040] Finish rolling stage: α = 0.3, β = 0.5, γ = 0.2;

[0041] Sizing stage: α = 0.2, β = 0.3, γ = 0.5.

[0042] Furthermore, the method for determining the set value is: statistically analyze the distribution of the surface topography change amount when copper powder exceeds the standard in historical production, and take the upper limit value of the 95% confidence interval.

[0043] Furthermore, the priority order for adjusting process parameters is: first increase the lubricant supply amount in the current process section, and the single adjustment range of the lubricant supply amount ≤ 15%; if the surface topography change amount corresponding to the adjusted rolling process parameters is still greater than the set value, then reduce the rolling speed, and the single adjustment range of the rolling speed ≤ 10%.

[0044] Furthermore, a base layer, a transition layer, and a functional layer are sequentially arranged on the roll surface from bottom to top. The thickness of the base layer is 50 - 80 μm, the thickness of the transition layer is 10 - 15 μm, and the thickness of the functional layer is 3 - 5 μm.

[0045] Furthermore, if the parameters are adjusted to the boundary of the feasible region and the surface topography change amount is still out of limit, trigger an alarm and switch to the standby roll set.

[0046] Preferably, the deviation of the surface topography parameters D0, Ra0, and η0 of the standby roll set from the initial design values of the current roll is ≤ 5%, and the switching action is completed within 10 seconds to ensure production continuity.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] (1) By constructing a full-process closed-loop control system of process segment optimization - dynamic monitoring and adjustment - rapid emergency response, the present invention systematically solves the problems in traditional processes, such as severe metal adhesion wear, deteriorated surface roughness, and difficulty in removing residual copper powder caused by large reduction and frictional heat in high-speed rolling. Specifically, the rolling process is divided into three segments: rough rolling, finish rolling, and sizing. Parameters are independently adjusted for each segment process. Further, the lubricant type and supply amount are matched according to tribological characteristics. In the rough rolling segment, a high-viscosity sulfur-containing extreme pressure lubricant is used. The extreme pressure anti-wear agent reduces direct metal contact, maximizing the reduction of copper powder generation. In the finish rolling segment, a low-viscosity mineral oil forms a stable oil film under high-speed rolling, improving the surface roughness (Ra). In the sizing segment, a water-based emulsion uses its cleaning characteristics to remove surface residual particles, significantly reducing the attached amount of finished copper powder.

[0049] (2) By dynamically monitoring the change amount of the roll surface topography, collecting data on groove depth, roughness, and oil storage rate, and introducing a dynamic weight coefficient, the present invention can give an early warning of the risk of excessive copper powder by 5 - 10 seconds. This avoids the situation where parameter adjustment lags behind the actual working condition changes. Once it is monitored that △K exceeds the threshold, the system preferentially adjusts the lubrication amount to quickly restore the lubrication state. Secondly, it reduces the rolling speed to control frictional heat. If the parameter adjustment is still ineffective when reaching the boundary of the feasible region, the standby roll group is automatically switched within 10 seconds, improving the continuous operation rate of the production line and reducing the annual fault shutdown time.

[0050] (3) By constructing a correlation model between process parameters and copper rod properties based on historical production data, the prediction model based on the random forest algorithm combines prior physical constraints to generate the feasible region of process parameters. Operators can intuitively obtain the dynamic boundaries of the green safety zone and the red warning zone, avoiding fluctuations in tensile strength caused by blind parameter adjustment. The human-machine interface displays the parameter position and adjustment suggestions in real time, which can improve the operation efficiency and shorten the debugging cycle. The present invention not only realizes the fundamental control of the copper powder problem, but also promotes the transformation of the rolling process towards intelligence and refinement, providing a reliable technical guarantee for the large-scale production of high-precision copper rods. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0052] Figure 1 It is a flowchart of the method for reducing surface copper powder in the copper rod rolling process of the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0053] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, a method for reducing surface copper powder during the rolling process of copper rods proposed according to the present invention, its specific implementation manners, features and effects are described in detail as follows. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0054] As Figure 1 shown, a method for reducing surface copper powder during the rolling process of copper rods includes the following steps:

[0055] S1 Divide the rolling process into multiple process sections along the rolling direction. Specifically, the rolling process can be divided into a rough rolling section, a finish rolling section and a sizing section according to the number of rolling stands, and different lubricants are supplied to each process section;

[0056] It should be noted that the reduction ratios of the rough rolling section, the finish rolling section and the sizing section gradually decrease. Among them, the rolling speed in the rough rolling section is relatively low, the reduction ratio is 35% - 45%, the reduction amount is relatively large, and the friction effect is relatively strong, which is likely to cause relatively large wear on the roll and the surface of the copper rod. The lubricant is required to have high viscosity and anti-wear ability, which can effectively reduce friction and lower the temperature rise. The reduction ratio in the finish rolling section is 15% - 25%. Although the contact pressure is relatively low, the speed is relatively fast and the friction force is still relatively large. It is necessary to reduce the heat generated by friction and maintain the surface quality. The lubricant should be able to reduce the friction heat during the high-speed rolling process, lower the surface roughness, and at the same time avoid excessive copper powder on the surface of the copper rod. The reduction ratio in the sizing section is 3% - 8%.

[0057] Specifically, in this embodiment, the lubricant supply amount is set differently according to the process section: in the rough rolling section, a sulfur-containing extreme pressure lubricant is used, with a viscosity of 120 - 150 cSt and a supply amount ≥ 20 L / min; in the finish rolling section, a low-viscosity mineral oil is used, with a viscosity of 40 - 60 cSt and a supply amount of 10 - 15 L / min; to reduce the surface temperature fluctuation of the roll and reduce the copper powder shedding caused by thermal brittle cracks on the copper surface. In the sizing section, a water-based emulsion is used, with a concentration of 8% - 12% and a supply amount of 5 - 8 L / min; by matching the lubricant viscosities in stages, the shear stress between the copper and the roll is reduced, and the copper powder peeling is inhibited.

[0058] To extend the service life of the roll, a substrate layer, a transition layer, and a functional layer are sequentially arranged on the roll surface from bottom to top. The thickness of the substrate layer is 50 - 80 μm, the thickness of the transition layer is 10 - 15 μm, and the thickness of the functional layer is 3 - 5 μm. With the multi-layer coating design on the roll surface of the present invention, the substrate layer provides high-strength support, and the low friction coefficient of the functional layer reduces copper adhesion and the risk of roll failure. Among them, the substrate layer can be an electroplated nickel layer or a WC-Co cemented carbide, the transition layer can be a chemically vapor-deposited Cr layer or the transition layer is NiCrAlY, and the functional layer can be a magnetron-sputtered Ti-Al-N coating or a CrN coating, with a surface hardness ≥ 2500 HV, which can be adjusted by those skilled in the art according to the actual situation.

[0059] S2 Establish a correlation model between process parameters and the properties of copper rods, and generate a feasible region of process parameters according to the target properties of copper rods;

[0060] Specifically, establishing a correlation model between process parameters and the properties of copper rods includes:

[0061] (1) Connect the MES system and the SCADA system, extract production data for six consecutive months from the production database, and extract the historical process parameter dataset, including rolling speed, rolling force, reduction ratio, lubricant supply amount, initial roll roughness, and the corresponding copper powder amount, tensile strength, and elongation of the copper rod surface;

[0062] When cleaning the data, the following abnormal records are excluded: the section where the instantaneous fluctuation of the rolling force exceeds ±10% and lasts for more than 5 seconds; the invalid data where the measured value of the surface copper powder amount exceeds the upper limit of the sensor range; the production data during the overload alarm of the main motor of the rolling mill; excluding the sudden change data of the rolling force can eliminate the noise interference caused by equipment vibration or sensor failure; excluding the data during motor overload avoids misleading the model training due to abnormal working conditions. After data cleaning, the signal-to-noise ratio of the training set is increased by ≥ 3 times, the generalization ability of the model is enhanced, and the misjudgment rate for sudden working conditions is reduced.

[0063] (2) For records with a process parameter missing rate less than 20%, after passing the time series continuity verification, the missing data is filled in by the cubic spline interpolation method; among them, the cubic spline interpolation retains the change trend of the process parameters and avoids the step distortion caused by linear interpolation; the virtual data forces the model to learn the boundary of the physically feasible region and prevents the prediction result from violating the equipment capacity limit.

[0064] The time series continuity verification requires simultaneously meeting the following conditions: (a) The rolling speed fluctuation within 10 seconds before and after the missing data segment ≤ ±3% (for example, the speed changes from 20 m / s to 20.6 m / s within the window); (b) The change rate of the rolling force ≤ 50 kN / s (calculated by first-order difference); (c) The adjustment interval of the reduction ratio ≥ 30 seconds (verified by the time stamp difference).

[0065] Based on the extreme values of process parameters, such as the rolling power boundary and the lower limit of lubricant quantity, boundary virtual data points are constructed using the rolling power formula P(kW) = F(kN) × v(m / s) / 1000 and the minimum lubricant quantity constraint Q ≥ 0.1F to expand the coverage of the dataset.

[0066] (3) Based on the expanded dataset, a relationship model between process parameters and the performance of copper rods is constructed using the random forest algorithm; the non-linear fitting ability of the random forest can capture the complex coupling relationship between process parameters and performance indicators.

[0067] (4) Incorporate prior knowledge constraints, eliminate interfering variables that violate physical laws through feature importance ranking, and then locally correct the abnormal points in the random forest prediction results that conflict with the knowledge constraints.

[0068] Specifically, the prior knowledge constraints are rules related to the friction situation when the copper rod contacts the rolling mill. For example, set the lower limit of lubricant supply quantity Q ≥ 0.1F, and establish the correlation constraint between rolling speed and reduction ratio. The method of locally correcting the conflicting data points is: find the 5 nearest compliant samples in the feature space and replace the original predicted value with a weighted average.

[0069] Among them, the feasible region of process parameters is generated according to the target performance of the copper rod, including:

[0070] Set the target performance constraint conditions of the copper rod; that is, set the target performance input module of the copper rod in the HMI interface of the industrial control system, including inputting the target performance of the copper rod (such as tensile strength ≥ 260 MPa, elongation ≥ 42%, surface copper powder content ≤ 15 mg / m 2 ); The constraint conditions are transmitted to the server side in JSON format and automatically converted into computable mathematical boundaries to provide a benchmark for parameter screening.

[0071] Based on the association model, a set of process parameter combinations that meet all constraints is screened to form a high-dimensional feasible region sample set; specifically, by calling the random forest model of the above components, the parameter space (rolling speed, rolling force, reduction ratio, etc.) is traversed to generate multiple groups of simulated parameter combinations; the tensile strength, elongation and copper powder content of each group are predicted by parallel computing, and the combinations that simultaneously meet all target constraints are screened (about 12 - 15% of the data is retained). The feasible region modeling covers the interactive effects of multiple parameters and avoids the chain process fluctuations caused by single parameter optimization.

[0072] After Min-Max normalization of the feasible region sample set, the DBSCAN algorithm (neighborhood radius ε = 0.3, minimum number of samples = 50) is applied for clustering. After removing discrete noise points, the convex hull of the largest cluster is used as the initial feasible region; the boundary of the convex hull of the largest cluster is extracted.

[0073] Specifically, discrete noise points are removed as follows: noise points (accounting for about 7-9%) and small clusters (clusters with the number of samples < 5% of the total data volume) are removed, parameter outliers caused by sensor noise or process anomalies are eliminated, the robustness of the feasible region is improved, and the convex hull boundary constraint transforms the non-linear constraint into a convex optimization problem to ensure the global optimality of the parameter adjustment path.

[0074] Then, the high-dimensional feasible region is fitted by support vector regression (SVR) to form a feasible region boundary function, and the boundary function is stored in the database in the form of a coefficient matrix. The specific coefficients are generated by fitting the training data. At a 95% confidence level, the parameter safety operation interval where all parameter combinations meet the target performance requirements is output to obtain the process parameter feasible region.

[0075] Specifically, the allowable operation range is marked in green in the human-machine interface. When the parameter point enters within 5% of the feasible region boundary, the interface triggers a red warning area and displays a list of adjustment suggestions, including preferentially increasing the lubrication amount or reducing the rolling speed. In the present invention, the three-dimensional visualization reduces the understanding threshold of multi-parameter coupling and helps the operator quickly locate anomalies; the distance threshold warning can provide a buffer interval to avoid frequent alarms.

[0076] S3: Monitor the surface parameters of the rolls in each process section respectively and calculate the change amount of the surface topography.

[0077] Furthermore, the roll surface parameters include groove depth, surface roughness, and oil storage rate, and the change amount of the surface topography is the weighted deviation value of the groove depth, surface roughness, and oil storage rate from the initial design values. Specifically for each process section, a laser profiler or a laser scanner is used to online detect the groove depth, the surface roughness is sampled by a contact roughness meter, and the oil storage rate is indirectly calculated by an oil film thickness sensor. The oil storage rate = measured oil film thickness / theoretical maximum oil film thickness × 100%. The data sampling interval is to collect once every 50-100 meters of rolling. By performing a weighted sum of these three surface parameters, the change of the roll surface topography can be quantified, thereby judging the risk of copper powder generation. If the change amount of the surface topography is greater than the set value, it indicates that the rolling process needs to be adjusted.

[0078] The calculation formula for the change amount of the surface topography is:

[0079]

[0080] Among them, ΔK is the surface topography change, d is the groove depth, Ra is the surface roughness, η is the oil storage rate, d0 is the initial design value of the groove depth, Ra0 is the initial design value of the surface roughness, η0 is the initial design value of the oil storage rate. For example, the initial values are designed as d0 = 50 ± 5 μm; Ra0 = 0.6 ± 0.1 μm; η0 ≥ 0.75; α, β, γ are weight coefficients, and the weight coefficients α, β, γ are dynamically adjusted according to the rolling stage. The dynamic allocation of weight coefficients makes the ΔK index more suitable for the core failure modes of each process section, and the abnormal detection response time is shortened by more than 30%.

[0081] In the initial setting, for example, in the rough rolling section, the ability to mechanically lock the lubricant in the grooves is emphasized. The values of the weight coefficients α, β, γ in the rough rolling section are α = 0.6, β = 0.2, γ = 0.2. In the finish rolling section, the surface finish is given priority, and Ra is mainly controlled to prevent scratches on the surface of the copper rod. The values of the weight coefficients α, β, γ in the finish rolling section are α = 0.3, β = 0.5, γ = 0.2. In the sizing section, the oil storage rate is strengthened to maintain the integrity of the oil film. The values of the weight coefficients α, β, γ in the sizing section are α = 0.2, β = 0.3, γ = 0.5. The weight coefficients are stored in the database parameter table and support manual fine-tuning (±0.1 range) on the HMI interface.

[0082] During the production process, in the finish rolling section, if the rolling speed is increased to more than 20 m / s, the β weight is corrected in real time through an on-line friction coefficient detector (μ value): β 新 = β 原 + 0.1×(μ 实测 - μ 基准 ) / μ 基准 . Through real-time process feedback, when the friction coefficient exceeds the reference value by 10%, β increases from 0.5 to 0.6, strengthening the monitoring of surface roughness and avoiding the out-of-control of copper powder caused by the sudden increase in frictional heat under high-speed rolling.

[0083] S4 Select the production data for 3 consecutive months, statistically analyze the ΔK distribution when the copper powder exceeds the standard in historical production, calculate the 95% quantile of ΔK based on kernel density estimation (KDE), and take the upper limit value of the 95% confidence interval as the set value. When the surface topography change is greater than the set value, through the PLC control module, adjust the rolling process parameters according to the priority of first adjusting the lubricant supply amount and then adjusting the rolling speed; the specific priority order is: first increase the lubricant supply amount of the current process section, and the single adjustment range of the lubricant supply amount ≤ 15%; if the surface topography change corresponding to the adjusted rolling process parameters is still greater than the set value, then reduce the rolling speed, and the single adjustment range of the rolling speed ≤ 10%. When reducing the speed, use an exponential decay curve instead of a step-by-step decrease to avoid process oscillations caused by parameter mutations.

[0084] Monitor the change amount of the surface topography corresponding to the adjusted rolling process parameters and compare it with the set value; until the change amount of the surface topography is less than the set value; the present invention can dynamically correct the process parameters based on the real-time feedback of the change amount of the surface topography, and avoid the increase of the copper powder amount index caused by the cumulative roll wear in the traditional open-loop control.

[0085] S6 When the rolling process parameters are adjusted to the boundary of the feasible region and the change amount of the surface topography is greater than the set value, an alarm is given. Specifically, the alarm signal can be set as an audible and visual alarm, and can be pushed to the industrial control system interface to prompt to switch to the standby roll group. The boundary constraint of the feasible region ensures that the adjusted parameters always meet the mechanical property requirements of the copper rod, and avoids the risk of rod breakage caused by excessive parameter adjustment.

[0086] In order to reduce the impact of roll change on the copper rod, the deviation of the surface topography parameters D0, Ra0, and η0 of the standby roll group from the initial design value of the current roll is ≤5%, and the switching action is completed within 10 seconds to ensure production continuity.

[0087] The present invention constructs a full-process closed-loop control system of process segmented optimization - dynamic monitoring and adjustment - rapid emergency response, and systematically solves the problems in the traditional process such as serious metal adhesive wear, deteriorated surface roughness, and difficult removal of residual copper powder caused by large reduction and high-speed rolling friction heat. Specifically, the rolling process is divided into three sections: rough rolling, finish rolling, and sizing. The process parameters are independently adjusted for each section, and further, the lubricant type and supply amount are matched according to the tribological characteristics. In the rough rolling section, a high-viscosity sulfur-containing extreme pressure lubricant is used to reduce the direct contact of metals through extreme pressure anti-wear agents, and the generation amount of copper powder is maximally reduced; in the finish rolling section, a low-viscosity mineral oil forms a stable oil film under high-speed rolling to improve the surface roughness (Ra), and in the sizing section, a water-based emulsion uses its cleaning characteristics to remove surface residual particles, significantly reducing the adhesion amount of finished copper powder.

[0088] Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for reducing surface copper powder during copper rod rolling, characterized in that: The following steps are involved: Divide the rolling process into multiple process sections and supply different lubricants to each process section; Establish a correlation model between process parameters and copper rod performance, and generate a feasible domain of process parameters based on the target performance of the copper rod; Monitor the roller surface parameters in each process section and calculate the surface morphology change; When the surface morphology change is greater than a set value, adjusting the rolling process parameters; Monitor the surface morphology change amount corresponding to the adjusted rolling process parameters and compare it with the set value; until the surface morphology change amount is less than the set value; When the rolling process parameters are adjusted to the boundary of the feasible region and the surface morphology change is greater than the set value, an alarm is issued.

2. A method for reducing surface copper powder during copper rod rolling according to claim 1, characterized in that: The rolling process is divided into a rough rolling section, a finishing rolling section and a shaping section along the rolling direction, and the reduction rates of the rough rolling section, the finishing rolling section and the shaping section are gradually reduced.

3. A method for reducing surface copper powder during copper rod rolling according to claim 2, characterized in that: The lubricant supply amount is set differently according to the process sections. The rough rolling section uses sulfur-containing extreme pressure lubricant with a supply amount of ≥20L / min; the finishing section uses low-viscosity mineral oil with a supply amount of 10-15L / min; the shaping section uses water-based emulsion with a supply amount of 5-8L / min.

4. The method for reducing surface copper powder during copper rod rolling according to claim 1, characterized in that: The generating of the process parameter feasible domain according to the target performance of the copper rod includes: Set copper rod target performance constraints; Based on the association model, a combination set of process parameters that meets all constraints is selected to form a high-dimensional feasible domain sample set; Clustering the feasible domain sample set, removing discrete noise points, and taking the convex hull of the largest cluster as the initial feasible domain; The feasible domain boundary function is fitted, and at a confidence level of 95%, the safe operating range of parameters for all parameter combinations that simultaneously meet the target performance requirements is output to obtain the feasible domain of process parameters.

5. The method for reducing surface copper powder during copper rod rolling according to claim 1, characterized in that: The roller surface parameters include groove depth, surface roughness and oil storage rate, and the surface morphology change is the weighted deviation value of the groove depth, the surface roughness and the oil storage rate from the initial design value.

6. A method for reducing surface copper powder during copper rod rolling according to claim 5, characterized in that: The calculation formula of the surface morphology change is: Among them, ΔK is the change in surface morphology, d is the groove depth, Ra is the surface roughness, η is the oil storage rate, d0 is the initial design value of the groove depth, Ra0 is the initial design value of the surface roughness, η0 is the initial design value of the oil storage rate, α, β, γ are weight coefficients, and the weight coefficients α, β, γ are dynamically adjusted according to the rolling stage.

7. A method for reducing surface copper powder during copper rod rolling according to claim 6, characterized in that: The method for determining the set value is: statistically analyzing the distribution of surface morphology changes when copper powder exceeds the standard in historical production, and taking the upper limit value of the 95% confidence interval.

8. The method for reducing surface copper powder during copper rod rolling according to claim 1, characterized in that: The priority order of adjusting the process parameters is: prioritize increasing the lubricant supply in the current process section, and the single adjustment range of the lubricant supply is ≤15%; if the surface morphology change corresponding to the adjusted rolling process parameters is still greater than the set value, reduce the rolling speed, and the single adjustment range of the rolling speed is ≤10%.

9. The method for reducing surface copper powder during copper rod rolling according to claim 1, characterized in that: The roller surface is provided with a base layer, a transition layer and a functional layer in sequence from bottom to top, the base layer has a thickness of 50 to 80 μm, the transition layer has a thickness of 10 to 15 μm, and the functional layer has a thickness of 3 to 5 μm.

10. The method for reducing surface copper powder during copper rod rolling according to claim 1, characterized in that: If the parameters are adjusted to the boundary of the feasible region and the surface morphology change still exceeds the limit, an alarm is triggered and the roll group is switched to the spare roll group.

Images