A positioning tool for crystal rod cutting and a crystal rod cutting method

Abstract

The present invention belongs to the technical field of sapphire processing, and discloses a positioning tool for cutting a crystal rod and a method for cutting a crystal rod, comprising: a base; a fixed frame, arranged at one end of the upper surface of the base, for supporting the crystal rod; a push structure, arranged at the other end of the upper surface of the base, for pushing the crystal rod forward; a clamping and positioning mechanism, slidably connected to the base, for clamping and fixing both sides of the position of the crystal rod to be cut; a control system, for controlling the positioning tool to perform adaptive clamping and positioning on the crystal rod; the control system comprises: a first data acquisition module, a first data processing module, a second data processing module, a second data acquisition module and an analysis and control module. The present invention can significantly improve the cutting accuracy and quality by predicting the cutting size of the crystal rod according to the characteristics of the crystal rod to be cut; by accurately predicting and determining the clamping position and clamping force, the adjustment time in the cutting process is reduced, and the overall production efficiency is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of sapphire processing, and in particular relates to a positioning tool for crystal rod cutting and a crystal rod cutting method. Background Art

[0002] Sapphire is a single crystal material whose main component is aluminum oxide. Due to its excellent physical and chemical properties, sapphire has been widely used in optics, electronics, semiconductors and other fields. Sapphire crystal rods are usually used to manufacture LED substrates, optical windows and high-performance electronic devices. Its high hardness, high thermal conductivity and excellent corrosion resistance make sapphire an ideal material for high-precision cutting and processing. During the cutting process of sapphire crystal rods, positioning fixtures are needed to stably clamp and position them to ensure cutting accuracy.

[0003] For example, the patent application with publication number CN105538527A discloses a sapphire crystal rod cutting tool, which places the crystal rod in a positioning groove and then presses it into position with a pressure plate to fix the crystal rod firmly. During cutting, the crystal rod will not jump and cause uneven rear end surface and edge collapse after cutting, thereby improving processing quality and efficiency.

[0004] For another example, a Chinese patent with authorization announcement number CN219686170U discloses a positioning tool for crystal rod cutting. By utilizing the mechanical clamping of the fixed chuck and the movable chuck in the clamping mechanism, it has fast and efficient clamping and fixing performance, can reduce the operating intensity of the staff, and at the same time has a stable clamping function. Through the combination of the transmission screw and the pushing slide, the mechanical clamping of the fixed chuck and the movable chuck is pushed, which has a firm and stable positioning and clamping function.

[0005] The above patents also have the following problems:

[0006] When the crystal rod is positioned and cut, due to the different characteristics of the crystal rod, the processing purposes of the crystal rod are also different, and thus the cutting size is also different. The above patent does not accurately position the crystal rod by the cutting size, so that the crystal rod is not stable enough during the cutting process and is prone to shaking, thereby affecting the cutting quality and cutting efficiency; and the existing positioning tooling cannot adapt the clamping force according to the characteristics of the crystal rod itself when clamping and positioning the crystal rod, and is prone to excessive clamping causing surface damage or increased internal stress of the crystal rod, and during the cutting process, it cannot monitor the positioning effect of the crystal rod in real time, and has a low degree of intelligence, which needs to be improved. Summary of the invention

[0007] The object of the present invention is to provide a positioning tool for crystal rod cutting, which is used to solve the technical problems raised in the existing background technology.

[0008] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a positioning tool for cutting a crystal rod, comprising: a base; a fixing frame, arranged at one end of the upper surface of the base, used to support the crystal rod; a pushing structure, arranged at the other end of the upper surface of the base, used to push the crystal rod forward; a clamping and positioning mechanism, arranged between the fixing frame and the pushing structure, and slidably connected to the base, used to clamp and fix both sides of the position where the crystal rod is to be cut; a control system, used to control the positioning tool to perform adaptive clamping and positioning on the crystal rod;

[0009] The control system comprises:

[0010] A first data collection module is used to collect a historical training data set of the positioning tool, wherein the historical training data set includes comprehensive information of the crystal rod and cutting dimensions;

[0011] A first data processing module is used to train a machine learning model for predicting cutting size based on a historical training data set, collect comprehensive information of the real-time crystal rod, and input it into the trained machine learning model to predict the cutting size;

[0012] The second data processing module collects historical cutting parameters and trains a machine learning model for predicting clamping force based on the historical training data set and the historical cutting parameters;

[0013] A second data acquisition module is used to collect comprehensive impact information of the positioning tooling in real time, wherein the comprehensive impact information includes clamping force, vibration impact coefficient and clamping distance;

[0014] The analysis control module is used to generate a clamping stability coefficient according to the comprehensive impact information, and to analyze and generate optimization instructions according to the clamping stability coefficient.

[0015] Preferably, the clamping and positioning mechanism includes: a base plate, slidably arranged on the base; a first clamping structure, arranged on one side of the upper surface of the base plate, and used to clamp and fix the crystal rod; a second clamping structure, which is the same as the first clamping structure and is symmetrically arranged on the other side of the upper surface of the base plate with the first clamping structure; a bidirectional lead screw, rotatably mounted on the base plate, and used to drive the first clamping structure and the second clamping structure to move; a lead screw nut, arranged on the bidirectional lead screw, and respectively connected to the first clamping structure and the second clamping structure; a drive motor, arranged on one side of the base plate, and the output end of which is fixedly connected to one end of the bidirectional lead screw.

[0016] Preferably, the first clamping structure includes: a clamping seat, which is slidably arranged on the base plate and has a V-shaped groove on its upper surface; a lower pressure block, which is arranged above the clamping seat and has a V-shaped groove on its lower surface; a gantry frame, which is fixedly arranged on the upper end of the clamping seat; a first electric push rod, which is fixedly arranged on the gantry frame, and whose extending end is fixedly connected to the top end of the lower pressure block, so as to adjust the distance between the lower pressure block and the clamping seat.

[0017] Preferably, the method of training the machine learning model to predict the cutting size is as follows:

[0018] Convert the collected comprehensive information of the crystal rod into a corresponding set of feature vectors;

[0019] Each group of feature vectors is used as the input of the machine learning model. The machine learning model takes the cutting size corresponding to each group of crystal rod comprehensive information as the output, takes the cutting size actually corresponding to each group of crystal rod comprehensive information as the prediction target, and takes minimizing the loss function value of the machine learning model as the training target; training is stopped when the loss function value of the machine learning model is less than or equal to the preset target loss value.

[0020] Preferably, the training method of the machine learning model for predicting the clamping force includes:

[0021] Convert the historical training data set and the historical cutting parameters into a corresponding set of feature vectors, use each set of feature vectors as the input of the machine learning model, the machine learning model uses a set of clamping forces corresponding to each set of historical training data set and historical cutting parameters as output, uses a set of clamping forces actually corresponding to each set of historical training data set and historical cutting parameters as prediction targets, and uses minimizing the loss function value of the machine learning model as the training target; stop training when the loss function value of the machine learning model is less than or equal to the preset target loss value.

[0022] Preferably, the vibration influence coefficient is generated as follows:

[0023] Parameters related to the vibration influence coefficient include: vibration frequency, vibration amplitude and vibration duration;

[0024] The vibration duration indicates the time from the start of vibration to the end of sampling;

[0025]

[0026] In the formula, Zd represents the vibration influence coefficient, Zp represents the vibration frequency, Zf represents the vibration amplitude, and Zs represents the vibration duration. are weight coefficients, Both are greater than 0.

[0027] Preferably, the clamping stability coefficient is generated as follows:

[0028]

[0029] In the formula, Wd represents the clamping stability coefficient, Jc represents the clamping force, Zd represents the vibration influence coefficient, and Jl represents the clamping distance; are weight coefficients, Both are greater than 0.

[0030] Preferably, the method of analyzing and generating optimization instructions according to the clamping stability coefficient is as follows:

[0031] Preset clamping stability factor threshold Wd 1 ;

[0032] If Wd≤Wd 1 , it means that the clamping stability coefficient is less than or equal to the clamping stability coefficient threshold, and the stability of the crystal rod being clamped is poor. At this time, an optimization instruction is generated;

[0033] If Wd>Wd 1 , it means that the clamping stability coefficient is greater than the clamping stability coefficient threshold, and the stability of the crystal rod being clamped is good. At this time, no optimization instruction is generated.

[0034] Preferably, the optimization instruction includes:

[0035] Set the thresholds corresponding to the clamping force and clamping distance, which are recorded as the maximum clamping force and the minimum clamping distance respectively;

[0036] The priority of setting the clamping distance is higher than the priority of the clamping force;

[0037] When the clamping distance and the clamping force have not reached their respective corresponding thresholds, the clamping distance is first reduced until the clamping stability coefficient is greater than the clamping stability coefficient threshold, and then the reduction of the clamping distance is stopped;

[0038] When the clamping distance is reduced to the minimum clamping distance, and the clamping stability coefficient is still less than or equal to the clamping stability coefficient threshold, the clamping force is then increased until the clamping stability coefficient is greater than the clamping stability coefficient threshold, and the clamping force stops increasing.

[0039] A method for cutting a crystal rod, applied to the above-mentioned positioning tool for cutting a crystal rod, comprises the following steps:

[0040] Step 1: Place the V end of the crystal rod to be cut on the push plate and the other end on the fixed frame;

[0041] Step 2: Collect historical training data sets of the positioning tooling, which include comprehensive information of the crystal rod and cutting dimensions;

[0042] Step 3: Based on the historical training data set, a machine learning model is trained to predict the cutting size, and the comprehensive information of the real-time crystal rod is collected and input into the trained machine learning model to predict the cutting size, and based on the predicted cutting size, the clamping positioning mechanism is moved to the corresponding position;

[0043] Step 4: Collect historical cutting parameters, and train a machine learning model that predicts the clamping force based on the historical training data set and the historical cutting parameters. Based on the predicted clamping force, use a clamping positioning mechanism to clamp and fix the crystal rod;

[0044] Step 5: Using a cutting knife to cut the clamped crystal rod at the corresponding cutting point;

[0045] Step 6: During the cutting process, comprehensive influence information of the positioning tooling is collected in real time, and the comprehensive influence information includes clamping force, vibration influence coefficient and clamping distance;

[0046] Step 7: Generate a clamping stability coefficient based on the comprehensive impact information, analyze and generate optimization instructions based on the clamping stability coefficient.

[0047] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0048] 1. The present invention can significantly improve cutting accuracy and quality and optimize material utilization by predicting the cutting size of the crystal rod according to the length, diameter, shape, crystal orientation and crystal quality grade of the crystal rod to be cut; by accurately predicting and determining the clamping position and clamping force, the adjustment time in the cutting process is reduced and the overall production efficiency is improved; and appropriate clamping force can avoid excessive clamping that causes surface damage to the crystal rod or increased internal stress, and appropriate clamping force and a stable cutting process can reduce equipment wear.

[0049] 2. The present invention can provide real-time feedback on the stability of the crystal rod during cutting by real-time monitoring of the clamping stability coefficient during the crystal rod cutting process, and optimize and adjust according to the results of real-time monitoring, which can significantly reduce the impact of vibration on the cutting accuracy, ensure a smooth and flat cutting surface, and reduce downtime adjustment time caused by vibration or instability. Real-time optimization and adjustment of the clamping force can prevent damage to the crystal rod caused by excessive clamping, and real-time optimization and adjustment of the clamping distance can reduce vibration during the cutting process, improve the safety of the operating environment, and ensure the cutting quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A perspective view of the present invention is shown;

[0052] Figure 2 A schematic diagram of the structure of the clamping and positioning mechanism of the present invention is shown;

[0053] Figure 3 A module diagram of Embodiment 2 of the present invention is shown;

[0054] Figure 4 A module diagram of Embodiment 3 of the present invention is shown.

[0055] Figure numerals: 100, base; 200, fixed frame; 201, second electric push rod; 300, push structure; 301, hydraulic cylinder; 302, push plate; 400, clamping and positioning mechanism; 401, bottom plate; 402, first clamping structure; 402a, clamping seat; 402b, lower pressure block; 402c, gantry; 402d, first electric push rod; 403, second clamping structure; 404, bidirectional screw; 405, screw nut; 406, driving motor. DETAILED DESCRIPTION

[0056] 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.

[0057] Example 1

[0058] Reference Figure 1As shown, a positioning tool for cutting a crystal rod comprises: a base 100, a fixing frame 200, a push structure 300 and a clamping and positioning mechanism 400, wherein the base 100 is arranged in a rectangular plate shape as a whole, the fixing frame 200 is arranged at one end of the upper surface of the base 100, and is used to support the crystal rod; the push structure 300 is arranged at the other end of the upper surface of the base 100, the push structure 300 comprises a hydraulic cylinder 301 and a push plate 302, the extended end of the hydraulic cylinder 301 is fixedly connected to the push plate 302, the push plate 302 is arranged in an L shape as a whole, supports one end of the crystal rod to be clamped, and the push structure 300 is used as a whole to push the crystal rod forward; the clamping and positioning mechanism 400 is arranged between the fixing frame 200 and the push structure 300, and is slidably connected to the base 100, and is used to clamp and fix both sides of the position where the crystal rod is to be cut;

[0059] In this embodiment, one end of the crystal rod to be clamped can be placed on the push plate 302 and the other end can be placed on the fixing frame 200. Then, the clamping and positioning mechanism 400 is used to clamp and fix the two sides of the position where the crystal rod is to be cut, thereby ensuring that the cut crystal rod is clamped stably.

[0060] Specifically, refer to Figure 1 and Figure 2 As shown, the clamping and positioning mechanism 400 includes: a bottom plate 401, a first clamping structure 402, a second clamping structure 403, a bidirectional lead screw 404, a lead screw nut 405 and a driving motor 406, wherein the bottom plate 401 is arranged in a rectangular plate shape as a whole, a slider is arranged at the bottom end of the bottom plate 401, a slide groove is opened on the upper surface of the base 100, the bottom plate 401 is slidably arranged on the base 100 through the slider and the slide groove, a second electric push rod 201 is arranged between the bottom plate 401 and the fixing frame 200, and the extending end of the second electric push rod 201 is fixedly connected to the bottom plate 401; the first clamping structure 402 is arranged on the bottom plate 40 1, used for clamping and fixing the crystal rod; the second clamping structure 403 is the same as the first clamping structure 402, and is symmetrically arranged on the other side of the upper surface of the bottom plate 401 with the first clamping structure 402; the bidirectional lead screw 404 is rotatably installed on the bottom plate 401, and is used to drive the first clamping structure 402 and the second clamping structure 403 to move; the lead screw nut 405 is arranged on the bidirectional lead screw 404, and is respectively connected to the first clamping structure 402 and the second clamping structure 403; the driving motor 406 is arranged on one side of the bottom plate 401, and the output end is fixedly connected to one end of the bidirectional lead screw 404;

[0061] In this embodiment, the second electric push rod 201 can be extended or contracted to drive the base plate 401 to slide, thereby moving the entire clamping and positioning mechanism 400 to the cutting position of the crystal rod, and the first clamping structure 402 and the second clamping structure 403 are respectively used to clamp and fix the two sides of the cutting point of the crystal rod. At the same time, the setting of the bidirectional lead screw 404 and the lead screw nut 405 can drive the first clamping structure 402 and the second clamping structure 403 to move toward or away from each other, and then the distance between the two clamping points can be adjusted so that the clamping point is close to the cutting point, thereby further increasing the stability of the clamping.

[0062] Further, see Figure 2 As shown, the first clamping structure 402 includes: a clamping seat 402a, a lower pressure block 402b, a gantry frame 402c and a first electric push rod 402d, wherein the clamping seat 402a is slidably set on the bottom plate 401, and a V-shaped groove is provided on the upper surface of the sliding seat; the lower pressure block 402b is set above the clamping seat 402a, and a V-shaped groove is provided on the lower surface; the gantry frame 402c is fixedly set on the upper end of the clamping seat 402a; the first electric push rod 402d is fixedly set on the gantry frame 402c, and the extending end is fixedly connected to the top end of the lower pressure block 402b, which is used to adjust the distance between the lower pressure block 402b and the clamping seat 402a.

[0063] In this embodiment, the crystal rod to be clamped can be placed inside the V-shaped groove on the clamping seat 402a, and then the first electric push rod 402d and the extension push the lower pressure block 402b downward, so that the two V-shaped grooves cooperate with each other to clamp and fix the crystal rod. In order to increase the stability of the crystal rod being clamped and reduce the vibration of the crystal rod, a shock-absorbing pad is provided on the inner wall of the V-shaped groove to protect the crystal rod and further increase the stability of the crystal rod. At the same time, a pressure sensor is also provided on the inner wall of the V-shaped groove to monitor the pressure of the crystal rod being clamped in real time.

[0064] Working principle: when it is necessary to clamp the crystal rod to be cut, one end of the crystal rod to be clamped is placed on the push plate 302, and the other end is placed on the fixed frame 200, and then the clamping and positioning mechanism 400 is moved according to the cutting size, and the second electric push rod 201 is extended or contracted to drive the bottom plate 401 to slide, so that the entire clamping and positioning mechanism 400 is moved to the cutting position of the crystal rod, and the first electric push rod 402d and the extension are used to push the lower pressing block 402b downward, so that the two V-shaped grooves cooperate with each other to clamp and fix the crystal rod, and then the first clamping structure 402 and the second clamping structure 403 are used to clamp and fix the two sides of the cutting point of the crystal rod respectively, and at the same time, the setting of the bidirectional lead screw 404 and the lead screw nut 405 can drive the first clamping structure 402 and the second clamping structure 403 to move toward or away from each other, and then the distance between the two clamping points can be adjusted, so that the clamping point is close to the cutting point, thereby further increasing the stability of the clamping, and the cutting stability can be guaranteed, and the cutting accuracy is high.

[0065] Example 2

[0066] Reference Figure 3 As shown, in order to accurately control the clamping position and clamping force of the crystal rod and ensure the stability of cutting, the present invention further includes a first data acquisition module, a first data processing module and a second data processing module on the basis of Example 1, and the modules are connected in a wired / wireless manner to realize data transmission;

[0067] A first data collection module is used to collect a historical training data set of the positioning tool, wherein the historical training data set includes comprehensive information of the crystal rod and cutting dimensions;

[0068] Specifically, the comprehensive information of the crystal rod includes the length, diameter, shape, crystal orientation, and crystal quality grade of the crystal rod to be cut;

[0069] Among them, the length of the crystal rod can be obtained in real time by a laser rangefinder; the diameter of the crystal rod can be obtained by real-time scanning by a laser scanner;

[0070] The shapes of the crystal rods include cylindrical crystal rods, hexagonal crystal rods, square crystal rods, conical crystal rods and polyhedral crystal rods;

[0071] The crystal orientation of the crystal rod includes C-plane, R-plane, A-plane, M-plane and N-plane. Different crystal orientations of the crystal rod will affect the physical and chemical properties of the crystal rod, and further affect its processing and application;

[0072] The crystal quality grades of the crystal rod include A, B, C, D and E, and the grades A, B, C, D and E decrease in sequence. Different applications have different requirements for crystal quality, so the quality grade standards are also different;

[0073] Among them, the shape of the crystal rod, the crystal orientation of the crystal rod and the crystal quality grade of the crystal rod can be input through a screen page that interacts with the positioning tooling.

[0074] A first data processing module is used to train a machine learning model for predicting cutting size based on a historical training data set, collect comprehensive information of the real-time crystal rod, and input it into the trained machine learning model to predict the cutting size;

[0075] The way to train a machine learning model to predict the cut size is as follows:

[0076] Convert the collected comprehensive information of the crystal rod into a corresponding set of feature vectors;

[0077] Each set of feature vectors is used as the input of the machine learning model. The machine learning model uses the cutting size corresponding to each set of crystal rod comprehensive information as the output, and the actual cutting size corresponding to each set of crystal rod comprehensive information as the prediction target. Minimizing the machine learning model loss function value is used as the training target. When the machine learning model loss function value is less than or equal to the preset target loss value, the training is stopped.

[0078] The machine learning model can be one of the models such as SVM regression, random forest regression or neural network regression.

[0079] The loss function value of the machine learning model is the mean square error;

[0080] Mean square error is one of the commonly used loss functions. Minimization is used as the goal to train the model so that the machine learning model can better fit the data, thereby improving the performance and accuracy of the model;

[0081] In the loss function, MSE is the loss function value of the machine learning model, x is the feature vector group number; m is the number of feature vector groups; y is the number of feature vector groups; x is the predicted cutting size for the xth group of feature vectors, is the actual cutting size corresponding to the xth group of eigenvectors;

[0082] Other model parameters of the machine learning model, target loss value, optimization algorithm, training set test set validation set ratio, and loss function optimization are all achieved through actual engineering implementation and continuous experimental tuning.

[0083] According to the predicted cutting size, the clamping and positioning mechanism 400 is moved to the corresponding position to clamp and position the crystal rod to be cut, which can ensure accurate positioning and achieve better cutting effect;

[0084] For example, if the cutting size is 3 cm, it means that the point extending 3 cm from one end of the crystal rod to the other end of the crystal rod is marked as the cutting point. At this time, the clamping and positioning mechanism 400 can be moved to the positions on both sides of the cutting point to clamp and fix the crystal rod to ensure stable clamping.

[0085] The second data processing module collects historical cutting parameters and trains a machine learning model for predicting clamping force based on the historical training data set and historical cutting parameters. The historical cutting parameters include the frequency and amplitude of vibration generated during cutting, and the temperature and humidity of the cutting environment.

[0086] The vibration frequency is obtained as follows: a vibration sensor is installed at a corresponding position of the positioning tool to collect a signal of vibration of the clamped crystal rod, wherein the vibration sensor can be one of an acceleration sensor, a velocity sensor and a displacement sensor, and the collected vibration signal is converted into a digital signal for storage and analysis, and the collected vibration data is preprocessed, including filtering, noise reduction and other operations to remove interference and extract effective vibration signals, and the time domain vibration signal is converted into a frequency domain signal using Fourier transform (or fast Fourier transform, FFT), and the main frequency component is found in the spectrum, that is, the main frequency component of the vibration signal is the vibration frequency;

[0087] The vibration amplitude can be expressed by calculating the root mean square value of the vibration signal as follows:

[0088]

[0089] In the formula, Zf represents the vibration amplitude, T represents the vibration signal acquisition time, and x(t) represents the sampling value of the vibration signal;

[0090] High frequency and large amplitude vibrations will have an adverse effect on the stability of the crystal ingot, requiring higher clamping force to keep the crystal ingot fixed.

[0091] The cutting environment temperature can be obtained through a temperature sensor. Temperature changes will cause thermal expansion and contraction of the crystal rod and the equipment, so it may be necessary to adjust the clamping force.

[0092] The humidity of the cutting environment can be obtained through a humidity sensor. A high humidity environment may affect the friction on the surface of the crystal rod, requiring a higher clamping force to maintain stability.

[0093] The training method of the machine learning model that predicts the clamping force includes:

[0094] Convert the historical training data set and the historical cutting parameters into a corresponding set of feature vectors, use each set of feature vectors as the input of the machine learning model, the machine learning model uses a set of clamping forces corresponding to each set of historical training data set and historical cutting parameters as output, uses a set of clamping forces actually corresponding to each set of historical training data set and historical cutting parameters as prediction targets, and uses minimizing the loss function value of the machine learning model as the training target; when the loss function value of the machine learning model is less than or equal to the preset target loss value, stop training, and the machine learning model can be one of the models such as support vector machine regression, random forest regression or neural network regression.

[0095] The loss function value of the machine learning model is the mean square error.

[0096] Mean square error is one of the commonly used loss functions. Minimization is used as the goal to train the model so that the machine learning model can better fit the data, thereby improving the performance and accuracy of the model;

[0097] MSE in loss function 1 is the loss function value of the machine learning model, s is the feature vector group number; q is the number of feature vector groups; w S is the clamping force predicted by the sth group of feature vectors, is the actual clamping force corresponding to the sth group of eigenvectors.

[0098] Other model parameters of the machine learning model, target loss value, optimization algorithm, training set test set validation set ratio, and loss function optimization are all achieved through actual engineering implementation and continuous experimental tuning.

[0099] The crystal rod to be cut is effectively clamped according to the predicted clamping force, thereby ensuring the stability of the crystal rod during cutting and ensuring a good cutting effect.

[0100] In this embodiment, by predicting the cutting size of the crystal rod according to the length, diameter, shape, crystal orientation, and crystal quality grade of the crystal rod to be cut, the cutting accuracy and quality can be significantly improved, the material utilization rate can be optimized, the production efficiency can be improved, and the personalized needs can be better met; and the position of the crystal rod clamping and positioning is determined by the predicted cutting size, ensuring that the crystal rod is always in the best position during the cutting process, reducing displacement and deviation, and the accurate clamping position can ensure that the crystal rod is stable during the cutting process, thereby improving the cutting accuracy and consistency;

[0101] By predicting the clamping force of the crystal rod based on the historical training data set and historical cutting parameters, excessive clamping that may cause damage to the crystal rod surface or increase in internal stress can be avoided. Appropriate clamping force and a stable cutting process can reduce equipment wear. By accurately predicting and determining the clamping position and clamping force, the adjustment time during the cutting process can be reduced, thereby improving overall production efficiency.

[0102] Example 2

[0103] Reference Figure 4 As shown, in order to further optimize the stability of the crystal rod clamping and positioning during the cutting process, the present invention further includes the following on the basis of Example 1:

[0104] The second data acquisition module is used to collect comprehensive impact information of the positioning tooling in real time, and the comprehensive impact information includes clamping force, vibration impact coefficient and clamping distance;

[0105] The clamping force is the force applied by the clamping mechanism when clamping the crystal rod, which can be obtained in real time through the pressure sensor;

[0106] Specifically, the vibration influence coefficient is generated as follows:

[0107] Parameters related to the vibration influence coefficient include: vibration frequency, vibration amplitude and vibration duration;

[0108] The vibration duration indicates the time from the start of vibration to the end of sampling;

[0109]

[0110] In the formula, Zd represents the vibration influence coefficient, Zp represents the vibration frequency, Zf represents the vibration amplitude, and Zs represents the vibration duration. are weight coefficients, Both are greater than 0.

[0111] It should be noted that the higher the vibration frequency, the larger the vibration amplitude and the longer the vibration duration, the greater the impact of the vibration on the stability of the crystal rod clamping, and vice versa.

[0112] Specifically, the clamping distance represents the distance between the clamping point of the clamping mechanism and the cutting point, which can be obtained in real time by a laser rangefinder.

[0113] The analysis control module is used to generate a clamping stability coefficient according to the comprehensive impact information, and to analyze and generate optimization instructions according to the clamping stability coefficient.

[0114] Specifically, the clamping stability coefficient is generated as follows:

[0115]

[0116] In the formula, Wd represents the clamping stability coefficient, Jc represents the clamping force, Zd represents the vibration influence coefficient, and Jl represents the clamping distance; are weight coefficients, Both are greater than 0.

[0117] It should be noted that the greater the clamping force Jc, the larger the clamping stability coefficient Wd, and the higher the clamping stability. The larger the vibration influence coefficient Zd and the larger the clamping distance Jl, the smaller the clamping stability coefficient Wd, and the lower the clamping stability.

[0118] It should be noted that the formulas involved in the above-mentioned are all calculated by removing dimensions and taking numerical values. They are a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The weight coefficients in the formulas and the various preset thresholds in the analysis process are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data; the size of the weight coefficient is to quantify each parameter to obtain a specific value for subsequent comparison. The size of the weight coefficient depends on the amount of sample data and the corresponding processing coefficients initially set by technical personnel in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantized values.

[0119] The method of analyzing and generating optimization instructions based on the clamping stability coefficient is as follows:

[0120] Preset clamping stability factor threshold Wd 1 ;

[0121] If Wd≤Wd 1 , it means that the clamping stability coefficient is less than or equal to the clamping stability coefficient threshold, and the stability of the crystal rod being clamped is poor. At this time, an optimization instruction is generated;

[0122] If Wd>Wd 1 , it means that the clamping stability coefficient is greater than the clamping stability coefficient threshold, and the stability of the crystal rod being clamped is good. At this time, no optimization instruction is generated.

[0123] Specifically, the optimization instructions include:

[0124] Set the thresholds corresponding to the clamping force and clamping distance, which are recorded as the maximum clamping force and the minimum clamping distance respectively.

[0125] Setting the clamping distance takes precedence over the clamping force.

[0126] When the clamping distance and the clamping force have not reached their respective corresponding thresholds, the clamping distance is first reduced until the clamping stability coefficient is greater than the clamping stability coefficient threshold, and then the clamping distance is stopped from being reduced.

[0127] When the clamping distance is reduced to the minimum clamping distance, and the clamping stability coefficient is still less than or equal to the clamping stability coefficient threshold, the clamping force is then increased until the clamping stability coefficient is greater than the clamping stability coefficient threshold, and the clamping force stops increasing.

[0128] By setting the above priorities, damage to the crystal rod caused by excessive clamping force can be avoided as much as possible, and stable clamping positioning can be achieved.

[0129] In this embodiment, by real-time monitoring of the clamping stability coefficient during the crystal rod cutting process, real-time feedback of the stability of the crystal rod during cutting can be obtained, and optimization and adjustment can be performed based on the results of real-time monitoring, which can significantly reduce the impact of vibration on cutting accuracy, ensure a smooth and flat cutting surface, and reduce downtime adjustment time caused by vibration or instability. Real-time optimization and adjustment of the clamping force can prevent damage to the crystal rod caused by excessive clamping, and also avoid displacement and damage of the crystal rod caused by insufficient clamping force. Real-time optimization and adjustment of the clamping distance can reduce vibration during the cutting process, improve the safety of the operating environment, reduce equipment failures and safety accidents caused by vibration, reduce material waste caused by unstable cutting, and improve material utilization.

[0130] Example 4

[0131] A method for cutting a crystal rod, applied to the above-mentioned positioning tool for cutting a crystal rod, comprises the following steps:

[0132] Step 1: Place one end of the crystal rod to be cut on the push plate 302 and the other end on the fixing frame 200;

[0133] Step 2: Collect historical training data sets of the positioning tooling, which include comprehensive information of the crystal rod and cutting dimensions;

[0134] Step 3: Based on the historical training data set, a machine learning model is trained to predict the cutting size, and the comprehensive information of the real-time crystal rod is collected and input into the trained machine learning model to predict the cutting size, and based on the predicted cutting size, the clamping positioning mechanism 400 is moved to the corresponding position;

[0135] Step 4: Collect historical cutting parameters, and train a machine learning model for predicting the clamping force based on the historical training data set and the historical cutting parameters, and clamp and fix the crystal rod using the clamping positioning mechanism 400 based on the predicted clamping force;

[0136] Step 5: Using a cutting knife to cut the clamped crystal rod at the corresponding cutting point;

[0137] Step 6: During the cutting process, comprehensive influence information of the positioning tooling is collected in real time, and the comprehensive influence information includes clamping force, vibration influence coefficient and clamping distance;

[0138] Step 7: Generate a clamping stability coefficient based on the comprehensive impact information, analyze and generate optimization instructions based on the clamping stability coefficient.

[0139] In this embodiment, when cutting the crystal rod, the position near the cutting point is precisely clamped and positioned, and an appropriate clamping force is used to increase the stability during cutting. During the cutting process, the clamping stability coefficient is monitored in real time, and real-time feedback on the stability of the crystal rod during cutting can be obtained. Optimization and adjustment are performed based on the results of real-time monitoring, which can significantly reduce the impact of vibration on cutting accuracy and ensure a good cutting effect.

[0140] 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.

[0141] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A positioning tool for cutting a crystal rod, characterized in that: include: Base (100); A fixing frame (200) is arranged on one end of the upper surface of the base (100) and is used to support the crystal rod; A push structure (300) is arranged at the other end of the upper surface of the base (100) and is used to push the crystal rod forward; the push structure (300) comprises a hydraulic cylinder (301) and a push plate (302); an extended end of the hydraulic cylinder (301) is fixedly connected to the push plate (302); the push plate (302) is arranged in an L shape as a whole, and the push plate (302) supports one end of the crystal rod to be clamped; A clamping and positioning mechanism (400) is arranged between the fixing frame (200) and the pushing structure (300), and is slidably connected to the base (100), and is used to clamp and fix two sides of the position of the crystal rod to be cut; A control system, used to control the positioning tooling to adaptively clamp and position the crystal rod; The control system comprises: A first data collection module is used to collect a historical training data set of the positioning tool, wherein the historical training data set includes comprehensive information of the crystal rod and cutting dimensions; A first data processing module is used to train a machine learning model for predicting cutting size based on a historical training data set, collect comprehensive information of the real-time crystal rod, and input it into the trained machine learning model to predict the cutting size; The second data processing module collects historical cutting parameters and trains a machine learning model for predicting clamping force based on the historical training data set and the historical cutting parameters; A second data acquisition module is used to collect comprehensive impact information of the positioning tooling in real time, wherein the comprehensive impact information includes clamping force, vibration impact coefficient and clamping distance; An analysis control module is used to generate a clamping stability coefficient according to the comprehensive impact information, and to analyze and generate optimization instructions according to the clamping stability coefficient; The vibration influence coefficient is generated as follows: Parameters related to the vibration influence coefficient include: vibration frequency, vibration amplitude and vibration duration; The vibration duration indicates the time from the start of vibration to the end of sampling; In the formula, Zd represents the vibration influence coefficient, Zp represents the vibration frequency, Zf represents the vibration amplitude, and Zs represents the vibration duration. are weight coefficients, All are greater than 0; The clamping stability coefficient is generated as follows: In the formula, Wd represents the clamping stability coefficient, Jc represents the clamping force, Zd represents the vibration influence coefficient, and Jl represents the clamping distance; are weight coefficients, Both are greater than 0.

2. A positioning tool for cutting a crystal rod according to claim 1, characterized in that: The clamping and positioning mechanism (400) comprises: A bottom plate (401) slidably disposed on the base (100); A first clamping structure (402) is arranged on one side of the upper surface of the bottom plate (401) and is used to clamp and fix the crystal rod; A second clamping structure (403), which is identical to the first clamping structure (402) and is symmetrically arranged on the other side of the upper surface of the bottom plate (401) with the first clamping structure (402); A bidirectional lead screw (404) is rotatably mounted on the base plate (401) and is used to drive the first clamping structure (402) and the second clamping structure (403) to move; A lead screw nut (405), arranged on the bidirectional lead screw (404), and connected to the first clamping structure (402) and the second clamping structure (403) respectively; The driving motor (406) is arranged on one side of the base plate (401), and the output end is fixedly connected to one end of the bidirectional lead screw (404).

3. A positioning tool for cutting a crystal rod as claimed in claim 2, characterized in that: The first clamping structure (402) comprises: A clamping seat (402a) is slidably disposed on the bottom plate (401) and has a V-shaped groove on its upper surface; A lower pressing block (402b) is arranged above the clamping seat (402a) and has a V-shaped groove on its lower surface; A gantry frame (402c) fixedly arranged on the upper end of the clamping seat (402a); The first electric push rod (402d) is fixedly arranged on the gantry frame (402c), and its extended end is fixedly connected to the top of the lower pressing block (402b), and is used to adjust the distance between the lower pressing block (402b) and the clamping seat (402a).

4. A positioning tool for cutting a crystal rod according to claim 1 or 3, characterized in that: The method of training a machine learning model to predict the cutting size is as follows: Convert the collected comprehensive information of the crystal rod into a corresponding set of feature vectors; Each set of feature vectors is used as input of the machine learning model. The machine learning model uses the cutting size corresponding to each set of comprehensive information of the crystal rod as output, uses the cutting size actually corresponding to each set of comprehensive information of the crystal rod as a prediction target, and uses minimizing the loss function value of the machine learning model as a training target; When the loss function value of the machine learning model is less than or equal to the preset target loss value, training stops.

5. A positioning tool for cutting a crystal rod as claimed in claim 4, characterized in that: The training method of the machine learning model for predicting the clamping force includes: Convert the historical training data set and the historical cutting parameters into a corresponding set of feature vectors, use each set of feature vectors as the input of the machine learning model, the machine learning model uses a set of clamping forces corresponding to each set of historical training data set and historical cutting parameters as output, uses a set of clamping forces actually corresponding to each set of historical training data set and historical cutting parameters as prediction targets, and uses minimizing the loss function value of the machine learning model as the training target; stop training when the loss function value of the machine learning model is less than or equal to the preset target loss value.

6. A positioning tool for cutting a crystal rod as claimed in claim 5, characterized in that: The method of analyzing and generating optimization instructions according to the clamping stability coefficient is as follows: Preset clamping stability coefficient threshold Wd1; If Wd≤Wd1, it means that the clamping stability coefficient is less than or equal to the clamping stability coefficient threshold, and the stability of the crystal rod being clamped is poor. At this time, an optimization instruction is generated; If Wd>Wd1, it means that the clamping stability coefficient is greater than the clamping stability coefficient threshold, and the stability of the crystal rod being clamped is good. At this time, no optimization instruction is generated.

7. A positioning tool for cutting a crystal rod according to claim 6, characterized in that: The optimization instructions include: Set the thresholds corresponding to the clamping force and clamping distance, which are recorded as the maximum clamping force and the minimum clamping distance respectively; The priority of setting the clamping distance is higher than the priority of the clamping force; When the clamping distance and the clamping force have not reached their respective corresponding thresholds, the clamping distance is first reduced until the clamping stability coefficient is greater than the clamping stability coefficient threshold, and then the reduction of the clamping distance is stopped; When the clamping distance is reduced to the minimum clamping distance, and the clamping stability coefficient is still less than or equal to the clamping stability coefficient threshold, the clamping force is then increased until the clamping stability coefficient is greater than the clamping stability coefficient threshold, and the clamping force stops increasing.

8. A method for cutting a crystal rod, characterized in that: A positioning tool for cutting a crystal rod as claimed in claim 7, comprising the following steps: Step 1: placing one end of the crystal rod to be cut on the push plate (302) and the other end on the fixing frame (200); Step 2: Collect historical training data sets of the positioning tooling, which include comprehensive information of the crystal rod and cutting dimensions; Step 3: Based on the historical training data set, a machine learning model is trained to predict the cutting size, and the comprehensive information of the real-time crystal rod is collected and input into the trained machine learning model to predict the cutting size, and based on the predicted cutting size, the clamping positioning mechanism (400) is moved to a corresponding position; Step 4: collecting historical cutting parameters, and training a machine learning model for predicting clamping force based on the historical training data set and the historical cutting parameters, and clamping and fixing the crystal rod using a clamping positioning mechanism (400) based on the predicted clamping force; Step 5: Using a cutting knife to cut the clamped crystal rod at the corresponding cutting point; Step 6: During the cutting process, comprehensive influence information of the positioning tooling is collected in real time, and the comprehensive influence information includes clamping force, vibration influence coefficient and clamping distance; Step 7: Generate a clamping stability coefficient based on the comprehensive impact information, analyze and generate optimization instructions based on the clamping stability coefficient.

Images