Full-automatic shaver welding control system and method

Abstract

The invention discloses a full-automatic shaver welding method which comprises the following steps: workpiece positioning: adopting a high-precision visual positioning system and combining with a multi-axis mechanical arm to carry out precise positioning and welding parameter intelligent setting on each part of a shaver; laser welding is executed, specifically, a laser beam with high energy density is used for welding a shaver part under set welding parameters, and welding quality is monitored in real time, specifically, a multi-sensor fusion technology is adopted and comprises an infrared thermal imaging sensor and an ultrasonic flaw detection sensor, and temperature distribution and weld defect parameters in the welding process are monitored in real time; self-adaptive adjustment: monitoring data in real time according to the welding quality, and if the welding quality does not reach a preset standard; high-precision positioning and intelligent parameter setting are achieved, the advanced technology is adopted for real-time monitoring and adjustment, the welding quality and efficiency are improved, the cost is reduced, and automatic and intelligent development of shaver welding is promoted.

Description

Technical Field

[0001] The invention relates to the technical field of razor welding control systems, and in particular to a full-automatic razor welding control system and method. Background Art

[0002] In modern manufacturing, razors are a common personal care product with growing market demand. However, the traditional razor welding process has many problems, which seriously restricts product quality and production efficiency.

[0003] Traditional welding methods usually rely on manual operation, and welders rely on experience and skills to adjust welding parameters and perform welding operations. This method is not only labor-intensive, but also difficult to ensure welding quality and is easily affected by human factors, such as worker fatigue and skill level differences. Due to the limitations of manual operation, welding accuracy is often low, which is difficult to meet the requirements of modern razors for high-precision welding, resulting in the product's sealing, stability and durability being affected.

[0004] In terms of setting welding parameters, traditional processes mainly rely on empirical formulas and simple experiments, which lack scientificity and accuracy. Different razor parts have different materials, shapes and sizes, requiring different welding parameters, but traditional methods are difficult to achieve accurate parameter matching. For example, for some razor parts made of special materials, such as welding high-strength plastics and metals, traditional processes are difficult to find suitable welding parameters, which can easily lead to problems such as loose welding or overheating and damage to parts.

[0005] In terms of monitoring and controlling the welding process, traditional processes mainly rely on visual observation and simple detection tools, which cannot monitor the welding quality in real time and accurately. This makes it difficult to find and solve defects and problems in the welding process in a timely manner, and often requires a lot of rework and repair in the subsequent quality inspection process, which increases production costs and time costs.

[0006] With the continuous advancement of science and technology, laser welding technology has gradually been applied to the manufacturing industry. Laser welding has the advantages of concentrated energy, fast welding speed, and small heat-affected zone, making it very suitable for high-precision welding tasks. At the same time, the development of computer technology, sensor technology, and artificial intelligence technology has made it possible to achieve fully automatic welding.

[0007] In this context, there is an urgent need for an innovative fully automatic razor welding control system and method that can achieve high-precision workpiece positioning, intelligent welding parameter setting, real-time welding quality monitoring and adaptive adjustment control to improve welding quality and production efficiency, reduce production costs, and meet the market demand for high-quality razors. Summary of the invention

[0008] The present invention provides a self-service swab testing system and a sampling method to solve the problems mentioned in the above-mentioned prior art.

[0009] In order to achieve the above object, the present invention adopts the following technical solution: a fully automatic razor welding method, comprising the following steps:

[0010] S1. Workpiece positioning: Use a high-precision visual positioning system, combined with a multi-axis robotic arm, to accurately position each component of the razor, set the positioning accuracy error within the range of ±0.01 mm, place the parts to be welded in the welding work area, and set the number of positioning mark points in the work area to N. p ;

[0011] S2. Intelligent setting of welding parameters: By analyzing the material, shape and size characteristics of the razor parts, the built-in welding parameter database and intelligent algorithm are used to automatically set the optimal welding parameters. The welding parameters include welding current I, welding voltage V, and welding time T. The calculation formula of welding current I is I=k1×M+k2×S, where M is the welding current coefficient of the component material, S is the surface area of ​​the welding part, k1 and k2 are adjustment parameters obtained by fitting based on a large amount of experimental data; welding voltage and welding current satisfy the relationship V=k3×I 0.5 +k4, where k3 and k4 are empirical coefficients;

[0012] S3. Laser welding execution: Use a high energy density laser beam to weld the razor parts under the set welding parameters. Suppose the power of the laser beam is P, and the power P is dynamically adjusted according to the welding quality feedback data monitored in real time during the welding process. The adjustment formula is P=P0+ΔP×f(Q), where P0 is the initial laser power, ΔP is the power adjustment step, and f(Q) is the adjustment function determined according to the welding quality index;

[0013] S4. Real-time monitoring of welding quality: Multi-sensor fusion technology is used, including infrared thermal imaging sensors and ultrasonic flaw detection sensors, to monitor the temperature distribution and weld defect parameters in real time during the welding process. The resolution of temperature monitoring is set to R t , the accuracy of defect detection is A d ;

[0014] S5, Adaptive adjustment: According to the real-time monitoring data of welding quality, if it is found that the welding quality does not meet the preset standard, the welding parameters are automatically adjusted or the welding operation is performed again. The upper limit of the adjustment times is set to N. a .

[0015] Preferably, a system of a fully automatic razor welding control method comprises:

[0016] Positioning module: It consists of a high-precision visual sensor and a multi-axis robotic arm, used to accurately position the razor components;

[0017] Parameter setting module: contains welding parameter database and intelligent algorithm processor, used to automatically set welding parameters;

[0018] Welding execution module: uses high-power laser welding equipment to perform welding operations according to set parameters;

[0019] Quality monitoring module: integrated with multiple sensors to monitor welding quality in real time;

[0020] Control module: connects the above modules and performs adaptive adjustment and control of the system according to quality monitoring data.

[0021] Preferably, in step S1, a deep learning algorithm is used to train the visual positioning system, and the number of samples in the training data set is N. s , improve the accuracy and adaptability of positioning of shaver components;

[0022] In step S2, the influence of temperature and humidity environmental factors on welding parameters is considered, and the environmental factor correction coefficient is set as C e , dynamically correct the welding parameters.

[0023] Preferably, in step S3, the shape and energy distribution of the laser beam are optimized by using beam shaping technology, and the parameter adjustment method of the beam shaping is set as M b , used to improve the quality and efficiency of welding;

[0024] The welding trajectory prediction algorithm is introduced to predict the welding trajectory in advance according to the shape of the razor parts and the welding requirements. The prediction accuracy is set to A p , used to improve the accuracy and consistency of welding.

[0025] Preferably, in step S4, a data fusion algorithm is applied to process the multi-sensor data, and the complexity of the fusion algorithm is assumed to be C f , used to improve the accuracy and reliability of quality monitoring;

[0026] Establish a quality data storage and analysis system to store and analyze the quality data of each welding. Suppose the data storage capacity is C d , providing data support for subsequent process optimization and quality traceability.

[0027] Preferably, the multi-axis robot in the positioning module is driven by a high-precision servo motor. The precision parameter of the motor is P m , used to improve the motion accuracy and stability of the robotic arm;

[0028] The intelligent algorithm processor in the parameter setting module adopts a parallel computing architecture, and the number of designed computing cores is N. c , used to improve the speed and efficiency of parameter setting.

[0029] Preferably, the laser welding equipment in the welding execution module is equipped with an automatic focusing system, and the focusing accuracy is set to F a , ensuring that the laser beam is always focused on the welding part;

[0030] Use gas protection device to provide appropriate shielding gas during welding. Assume that the gas flow control accuracy is F g , improve welding quality and weld aesthetics;

[0031] Preferably, in the deep learning algorithm training, the transfer learning technology is used to pre-train using the existing visual positioning data in the related field, and the similarity threshold of the source field data of the transfer learning is set to S t , speed up training and improve positioning performance;

[0032] When considering environmental factors, an environmental monitoring sensor network is established to monitor welding environmental parameters in real time. The number of nodes in the sensor network is set as N. e .

[0033] Preferably, in the beam shaping technology, an adjustable optical element is used, and the adjustment range and accuracy of the optical element are set to R o , achieving flexible control of the laser beam.

[0034] The welding trajectory prediction algorithm combines artificial intelligence technology to continuously optimize the prediction model by learning and analyzing a large amount of historical welding data. The frequency of model optimization is set as F o .

[0035] Preferably, in the data fusion algorithm, a weighted fusion strategy is adopted to assign different weights according to the performance and reliability of different sensors, and the weight assignment function is assumed to be W(S), where S is the sensor type.

[0036] The quality data storage and analysis system adopts a distributed storage architecture. Suppose the number and distribution of storage nodes are N d , improving data storage security and access efficiency.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] The present invention uses a high-precision visual positioning system combined with the workpiece positioning steps of a multi-axis robotic arm to achieve precise positioning with a positioning accuracy error within the range of ±0.01 mm, and can accurately place various complex-shaped razor parts in the welding work area, greatly improving the accuracy and consistency of welding. This not only reduces welding defects caused by inaccurate positioning, but also lays a solid foundation for subsequent high-quality welding, effectively improving the overall quality and performance of the product.

[0039] In the intelligent setting step of welding parameters, the built-in welding parameter database and intelligent algorithm are used to automatically set the optimal welding parameters according to the material, shape, size and other characteristics of the parts, and dynamically correct them considering environmental factors. This enables the welding parameters to be accurately matched to each specific welding task, ensuring the stability and reliability of the welding process, whether for common razor parts or parts of special materials and structures. At the same time, the precise calculation of welding current and voltage and the function of adjusting the laser power in real time according to quality feedback greatly improve the quality and efficiency of welding and reduce energy waste and material loss.

[0040] In the laser welding execution step, a high-energy-density laser beam is used for welding, and the shape and energy distribution of the laser beam are optimized through beam shaping technology, and the welding trajectory prediction algorithm is introduced to further improve the quality and efficiency of welding. Beam shaping technology can adjust the laser beam according to different welding requirements, so that it can better adapt to various complex welding parts and improve the uniformity and firmness of welding. The welding trajectory prediction algorithm can plan the welding path in advance, ensure the accuracy and continuity of welding, reduce errors and repeated operations in the welding process, and improve production efficiency.

[0041] The real-time monitoring step of welding quality using multi-sensor fusion technology, combined with infrared thermal imaging sensors and ultrasonic flaw detection sensors, can monitor the temperature distribution and weld defects and other quality parameters in real time and accurately during the welding process. High-resolution temperature monitoring and high-accuracy defect detection functions can promptly detect problems in the welding process and provide accurate data support for the adaptive adjustment step. This not only helps to adjust welding parameters or re-perform welding operations in a timely manner, ensuring the stability of welding quality, but also provides detailed data records for product quality traceability.

[0042] The adaptive adjustment step automatically adjusts welding parameters or re-performs welding operations based on real-time monitoring data, effectively reducing scrap and rework rates. By setting an upper limit on the number of adjustments, the efficiency and stability of the production process are ensured, avoiding the waste of time and resources due to excessive adjustments.

[0043] In terms of the construction and technical application of the entire system, such as the training of the visual positioning system by the deep learning algorithm, the application of the transfer learning technology, the establishment of the environmental monitoring sensor network, the optimization of the data fusion algorithm, and the quality data storage and analysis system of the distributed storage architecture, the performance and intelligence level of the system have been further improved. The comprehensive application of these technologies makes the fully automatic razor welding control system and method of the present invention highly flexible, adaptable and reliable, able to meet the requirements of different production environments and product needs, and provides strong support for the automation and intelligent development of the razor manufacturing industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a schematic block diagram of the fully automatic razor welding control system proposed by the present invention;

[0045] Figure 2 This is a schematic block diagram of the fully automatic razor welding control method proposed in the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0048] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0049] Reference Figure 1-2 :A fully automatic razor welding method, comprising the following steps:

[0050] S1. Workpiece positioning: Use a high-precision visual positioning system, combined with a multi-axis robotic arm, to accurately position each component of the razor, set the positioning accuracy error within the range of ±0.01 mm, place the parts to be welded in the welding work area, and set the number of positioning mark points in the work area to N. p ;

[0051] S2. Intelligent setting of welding parameters: By analyzing the material, shape and size characteristics of the razor parts, the built-in welding parameter database and intelligent algorithm are used to automatically set the optimal welding parameters. The welding parameters include welding current I, welding voltage V, and welding time T. The calculation formula of welding current I is I=k1×M+k2×S, where M is the welding current coefficient of the component material, S is the surface area of ​​the welding part, k1 and k2 are adjustment parameters obtained by fitting based on a large amount of experimental data; welding voltage and welding current satisfy the relationship V=k3×I 0.5 +k4, where k3 and k4 are empirical coefficients;

[0052] S3. Laser welding execution: Use a high energy density laser beam to weld the razor parts under the set welding parameters. Suppose the power of the laser beam is P, and the power P is dynamically adjusted according to the welding quality feedback data monitored in real time during the welding process. The adjustment formula is P=P0+ΔP×f(Q), where P0 is the initial laser power, ΔP is the power adjustment step, and f(Q) is the adjustment function determined according to the welding quality index;

[0053] S4. Real-time monitoring of welding quality: Multi-sensor fusion technology is used, including infrared thermal imaging sensors and ultrasonic flaw detection sensors, to monitor the temperature distribution and weld defect parameters in real time during the welding process. The resolution of temperature monitoring is set to R t , the accuracy of defect detection is A d ;

[0054] S5, Adaptive adjustment: According to the real-time monitoring data of welding quality, if it is found that the welding quality does not meet the preset standard, the welding parameters are automatically adjusted or the welding operation is performed again. The upper limit of the adjustment times is set to N. a .

[0055] Preferably, a system of a fully automatic razor welding control method comprises:

[0056] Positioning module: It consists of a high-precision visual sensor and a multi-axis robotic arm, used to accurately position the razor components;

[0057] Parameter setting module: contains welding parameter database and intelligent algorithm processor, used to automatically set welding parameters;

[0058] Welding execution module: uses high-power laser welding equipment to perform welding operations according to set parameters;

[0059] Quality monitoring module: integrated with multiple sensors to monitor welding quality in real time;

[0060] Control module: connects the above modules and performs adaptive adjustment and control of the system according to quality monitoring data.

[0061] In the present invention, in step S1, a deep learning algorithm is used to train the visual positioning system. Assume that the number of samples in the training data set is N. s , improve the accuracy and adaptability of positioning of shaver components;

[0062] In step S2, the influence of temperature and humidity environmental factors on welding parameters is considered, and the environmental factor correction coefficient is set as C e , dynamically correct the welding parameters.

[0063] In the present invention, in step S3, the shape and energy distribution of the laser beam are optimized by using the beam shaping technology, and the parameter adjustment method of the beam shaping is set as M b , used to improve the quality and efficiency of welding;

[0064] The welding trajectory prediction algorithm is introduced to predict the welding trajectory in advance according to the shape of the razor parts and the welding requirements. The prediction accuracy is set to A p , used to improve the accuracy and consistency of welding.

[0065] In the present invention, in step S4, a data fusion algorithm is applied to process multi-sensor data. The complexity of the fusion algorithm is assumed to be C f , used to improve the accuracy and reliability of quality monitoring;

[0066] Establish a quality data storage and analysis system to store and analyze the quality data of each welding. Suppose the data storage capacity is C d , providing data support for subsequent process optimization and quality traceability.

[0067] In the present invention, the multi-axis mechanical arm in the positioning module is driven by a high-precision servo motor. The precision parameter of the motor is P m , used to improve the motion accuracy and stability of the robotic arm;

[0068] The intelligent algorithm processor in the parameter setting module adopts a parallel computing architecture, and the number of designed computing cores is N. c , used to improve the speed and efficiency of parameter setting.

[0069] In the present invention, the laser welding equipment in the welding execution module is equipped with an automatic focusing system, and the focusing accuracy is set to F a , ensuring that the laser beam is always focused on the welding part;

[0070] Use gas protection device to provide appropriate shielding gas during welding. Assume that the gas flow control accuracy is F g , improve welding quality and weld aesthetics;

[0071] In the present invention, in the deep learning algorithm training, the transfer learning technology is adopted to use the existing visual positioning data in the related field for pre-training, and the similarity threshold of the source field data of the transfer learning is set as S t , speed up training and improve positioning performance;

[0072] When considering environmental factors, an environmental monitoring sensor network is established to monitor welding environmental parameters in real time. The number of nodes in the sensor network is set as N. e .

[0073] In the present invention, in the beam shaping technology, adjustable optical elements are innovatively used to achieve highly flexible control of the laser beam. These optical elements include but are not limited to deformable mirrors, microlens arrays, and diffractive optical elements.

[0074] The deformable mirror has high-precision surface adjustment capability. Through electrical or mechanical control, the shape of the mirror can be precisely changed to finely adjust the wavefront of the laser beam. Assuming the minimum resolution of its shape adjustment is at the micron level, it can achieve arbitrary changes in the focus, defocus and beam profile of the laser beam. For example, when welding razor parts of different shapes and sizes, the laser beam can be adjusted to a circular, rectangular or other specific shape according to actual needs to adapt to complex welding locations and ensure the uniformity and quality of welding.

[0075] The microlens array is composed of multiple tiny lenses. By adjusting the arrangement and focal length of the microlenses, the laser beam can be split and recombined. Assuming the diameter of the microlens is at the millimeter level, the spacing can be customized according to different application requirements. During the welding process, the microlens array can be used to split the laser beam into multiple sub-beams and operate multiple welding points at the same time, greatly improving the welding efficiency. At the same time, by adjusting the parameters of the microlens, the energy distribution and focusing characteristics of the sub-beams can be changed to meet the requirements of different welding tasks.

[0076] Diffractive optical elements use the principle of light diffraction to perform complex phase and amplitude modulation on laser beams. By designing specific diffraction patterns, special shaping effects on laser beams can be achieved, such as generating Bessel beams and vortex beams. If the manufacturing precision of diffractive optical elements reaches the nanometer level, the propagation characteristics and energy distribution of laser beams can be accurately controlled. For example, when high-precision micro welding or welding of special materials is required, the special beams generated by diffractive optical elements can achieve better welding effects and reduce heat-affected zones and welding defects.

[0077] For these adjustable optical components, the adjustment range and accuracy are F o . The adjustment range covers from small local adjustments to large-scale beam transformations, which can meet the needs of various welding scenarios. In terms of accuracy, it can achieve precise control of laser beam parameters to ensure the stability and consistency of the laser beam during the welding process. At the same time, advanced automatic control technology is used to achieve fast and accurate adjustment of optical components. For example, through integration with the welding control system, the parameters of the optical components are automatically adjusted according to the real-time welding quality feedback data and welding trajectory prediction results to achieve dynamic optimization of the laser beam.

[0078] The welding trajectory prediction algorithm innovatively combines artificial intelligence technology to achieve accurate prediction and continuous optimization of the welding trajectory. First, a large amount of historical welding data is used to train the artificial intelligence model. These historical welding data include characteristic information such as the shape, size, and material of the razor parts, as well as the corresponding actual welding trajectory and welding quality data. The scale of the training data set is assumed to be thousands or even tens of thousands of records, covering various types of razor parts and welding scenarios.

[0079] The welding trajectory prediction model is constructed by using architectures such as recurrent neural network (RNN) or long short-term memory network (LSTM) in deep learning algorithms. These models can learn the time series characteristics and spatial relationships in the welding process, so as to accurately predict the future welding trajectory. After sufficient training and optimization, the prediction accuracy of the model can reach more than 90%, greatly improving the accuracy and consistency of welding.

[0080] During the training and optimization process of the model, a large amount of historical welding data is continuously learned and analyzed. By regularly updating the training data set, new welding data is incorporated into the model training, so that the model can adapt to the ever-changing welding tasks and environment. Let the frequency of model optimization be F o , such as optimizing the model once a day or a week to ensure that the model is always in the best condition.

[0081] At the same time, reinforcement learning technology is introduced to allow the model to continuously adjust and optimize itself during the actual welding process. Based on real-time feedback data on welding quality, such as the width, depth, uniformity and other indicators of the weld, the model can automatically adjust its prediction strategy and parameters to improve the accuracy and adaptability of the prediction. For example, if a deviation is found between the actual welding trajectory and the predicted trajectory, the model will automatically analyze the cause of the deviation and adjust its internal weights and parameters to reduce future prediction errors.

[0082] In the present invention, the quality data storage and analysis system adopts a distributed storage architecture, assuming that the number and distribution of storage nodes are N d , improving data storage security and access efficiency.

[0083] In the data fusion algorithm, an innovative weighted fusion strategy is adopted to achieve more accurate data processing. First, a scientific and reasonable weight distribution system is constructed according to the performance and reliability differences of different sensors. For sensors with stable performance and high detection accuracy, such as high-precision infrared thermal imaging sensors, higher initial weights are assigned according to their performance in long-term practical applications and professional performance evaluation indicators. The initial weight value of this type of sensor is set to range from 0.6 to 0.8, and the specific value is dynamically adjusted according to its contribution to the accuracy of key quality parameters (such as temperature distribution monitoring) in different welding scenarios.

[0084] For ultrasonic flaw detection sensors that have unique advantages under specific welding conditions, considering their excellent ability in weld defect detection, they are given corresponding weights according to their detection accuracy and complementarity with other sensors. The weight range is set between 0.3 and 0.6, and when encountering complex weld structures or scenarios with high requirements for micro-defect detection, the weight will be appropriately increased.

[0085] The weight distribution function W(S), where S is the sensor type, is calculated by comprehensively considering multiple key factors. These include the accuracy statistics of the sensor's historical data, where the statistical period is set to be one week or one month, and the evaluation is based on the consistency between the sensor data and the actual welding quality results during this period; the sensor's response time, where the sensor with a shorter response time will have a certain advantage in the weight calculation; and the sensor's anti-interference ability, where a higher weight adjustment coefficient is given to sensors that can maintain stable performance in complex electromagnetic environments or harsh working environments such as high temperature and high humidity.

[0086] In the actual real-time monitoring process of welding quality, the weight of each sensor is dynamically adjusted according to the characteristics of different welding stages and welding parts. For example, in the initial stage of welding, temperature changes are more critical. At this time, the weight of the infrared thermal imaging sensor is appropriately increased to more accurately monitor the temperature rise in the welding area and ensure timely adjustment of welding parameters. In the later stage of welding, weld defect detection becomes the focus. At this time, the weight of the ultrasonic flaw detection sensor is correspondingly increased to achieve accurate evaluation of weld quality.

[0087] The quality data storage and analysis system innovatively adopts a distributed storage architecture to meet the needs of efficient storage and fast access to large-scale welding quality data. The number and distribution of storage nodes are designed based on the scale of the razor welding production line and the speed of data generation.

[0088] Assume the number of storage nodes is N d ,In a medium-sized razor welding factory, dozens of distributed ,storage nodes are initially set up according to the daily welding task volume and ,data generation frequency. These storage nodes are distributed in different ,physical locations, such as near different welding workstations or in different areas of the ,data processing center, and connected through a high-speed network to form a ,distributed storage network.

[0089] Each storage node has a certain independent storage and data processing capabilities. The storage nodes use advanced storage technologies, such as solid-state drives (SSDs) and large-capacity hard disk arrays. The storage capacity of a single storage node is set between hundreds of GB and several TB to meet the long-term storage needs of a large amount of welding quality data. At the same time, the storage node is equipped with an efficient data caching mechanism. When new welding quality data is generated, it can quickly store the data in the cache and synchronize it to the entire distributed storage system at the right time to ensure the timeliness and integrity of the data.

[0090] In the process of data storage, data sharding and redundant backup technology are used. The quality data of each welding is stored in shards according to certain rules, such as classified storage according to welding time, product batch or welding part, and each data shard is stored on multiple different storage nodes to improve data reliability and availability. The level of data redundant backup is set to at least two backups to ensure that when a storage node fails, the data can still be quickly restored from other backup nodes to ensure data security.

[0091] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A fully automatic razor welding method, characterized in that: The following steps are involved: S1. Workpiece positioning: Use a high-precision visual positioning system, combined with a multi-axis robotic arm, to accurately position each component of the razor, set the positioning accuracy error within the range of ±0.01 mm, place the parts to be welded in the welding work area, and set the number of positioning mark points in the work area to N. p ; S2. Intelligent setting of welding parameters: By analyzing the material, shape and size characteristics of the razor parts, the built-in welding parameter database and intelligent algorithm are used to automatically set the optimal welding parameters. The welding parameters include welding current I, welding voltage V, and welding time T. The calculation formula of welding current I is I=k1×M+k2×S, where M is the welding current coefficient of the component material, S is the surface area of ​​the welding part, k1 and k2 are adjustment parameters obtained by fitting based on a large amount of experimental data; welding voltage and welding current satisfy the relationship V=k3×I 0.5 +k4, where k3 and k4 are empirical coefficients; S3. Laser welding execution: Use a high energy density laser beam to weld the razor parts under the set welding parameters. Suppose the power of the laser beam is P, and the power P is dynamically adjusted according to the welding quality feedback data monitored in real time during the welding process. The adjustment formula is P=P0+△P×f(Q), where P0 is the initial laser power, △P is the power adjustment step, and f(Q) is the adjustment function determined according to the welding quality index; S4. Real-time monitoring of welding quality: Multi-sensor fusion technology is used, including infrared thermal imaging sensors and ultrasonic flaw detection sensors, to monitor the temperature distribution and weld defect parameters in real time during the welding process. The resolution of temperature monitoring is set to R t , the accuracy of defect detection is A d ; S5, Adaptive adjustment: According to the real-time monitoring data of welding quality, if it is found that the welding quality does not meet the preset standard, the welding parameters are automatically adjusted or the welding operation is performed again. The upper limit of the adjustment times is set to N. a .

2. A system for implementing the fully automatic razor welding control method according to claim 1, characterized in that: include: Positioning module: It consists of a high-precision visual sensor and a multi-axis robotic arm, used to accurately position the razor components; Parameter setting module: contains welding parameter database and intelligent algorithm processor, used to automatically set welding parameters; Welding execution module: uses high-power laser welding equipment to perform welding operations according to set parameters; Quality monitoring module: integrated with multiple sensors to monitor welding quality in real time; Control module: connects the above modules and performs adaptive adjustment and control of the system according to quality monitoring data.

3. The fully automatic razor welding method according to claim 1, characterized in that: In step S1, the deep learning algorithm is used to train the visual positioning system. Assume that the number of samples in the training data set is N. s , improve the accuracy and adaptability of positioning of shaver components; In step S2, the influence of temperature and humidity environmental factors on welding parameters is considered, and the environmental factor correction coefficient is set as C e , dynamically correct the welding parameters.

4. The fully automatic razor welding method according to claim 1, characterized in that: In step S3, the beam shaping technology is used to optimize the shape and energy distribution of the laser beam. The parameter adjustment method of the beam shaping is set as M b , used to improve the quality and efficiency of welding; The welding trajectory prediction algorithm is introduced to predict the welding trajectory in advance according to the shape of the razor parts and the welding requirements. The prediction accuracy is set to A p , used to improve the accuracy and continuity of welding.

5. The fully automatic razor welding method according to claim 1, characterized in that: In step S4, the data fusion algorithm is applied to process the multi-sensor data. The complexity of the fusion algorithm is assumed to be C f , used to improve the accuracy and reliability of quality monitoring; Establish a quality data storage and analysis system to store and analyze the quality data of each welding. Suppose the data storage capacity is C d , providing data support for subsequent process optimization and quality traceability.

6. The fully automatic razor welding control system according to claim 2 is characterized in that: The multi-axis robot in the positioning module is driven by a high-precision servo motor. The motor's precision parameter is P m , used to improve the motion accuracy and stability of the robotic arm; The intelligent algorithm processor in the parameter setting module adopts a parallel computing architecture, and the number of designed computing cores is N. c , used to improve the speed and efficiency of parameter setting.

7. The fully automatic razor welding control system according to claim 2 is characterized in that: The laser welding equipment in the welding execution module is equipped with an automatic focusing system, and the focusing accuracy is set to F a , ensuring that the laser beam is always focused on the welding part; Use gas protection device to provide appropriate shielding gas during welding. Assume that the gas flow control accuracy is F g , improve welding quality and the aesthetics of welds.

8. The fully automatic razor welding method according to claim 3, characterized in that: In the deep learning algorithm training, the transfer learning technology is used to pre-train the existing visual positioning data in related fields. The similarity threshold of the source field data of the transfer learning is set as S t , speed up training and improve positioning performance; When considering environmental factors, an environmental monitoring sensor network is established to monitor welding environmental parameters in real time. The number of nodes in the sensor network is set as N. e .

9. The fully automatic razor welding method according to claim 4, characterized in that: In the beam shaping technology, an adjustable optical element is used, and the adjustment range and accuracy of the optical element are set as R. , to achieve flexible control of the laser beam; The welding trajectory prediction algorithm combines artificial intelligence technology to continuously optimize the prediction model by learning and analyzing a large amount of historical welding data. The frequency of model optimization is set as F o .

10. The fully automatic razor welding method according to claim 5, characterized in that: In the data fusion algorithm, a weighted fusion strategy is adopted to assign different weights according to the performance and reliability of different sensors. The weight assignment function is set to W(S), where S is the sensor type; The quality data storage and analysis system adopts a distributed storage architecture. Suppose the number and distribution of storage nodes are N d , improving data storage security and access efficiency.

Images