Crystal machining methods and systems
The method and system for crystal machining address the issue of non-uniform optical output and cracking by determining surface parameters for light and non-light-output surfaces, ensuring consistent performance and reducing costs through automated machining.
Patent Information
- Application Number
- US19/720748
- Authority / Receiving Office
- US · United States
- Patent Type
- Applications(United States)
- Current Assignee / Owner
- Filing Date
- 2026-06-25
- Publication Date
- 2026-11-12
AI Technical Summary
Scintillation crystals face challenges in achieving uniformity of optical output and are prone to cracking during machining due to variations in doping concentration and thermal stress, which affects their performance and increases production costs.
A method and system for crystal machining that determines light output and non-light-output surfaces based on target optical yield or output values, adjusting surface parameters to ensure consistent optical performance and reduce cracking, using automated machining with grinding and polishing parameters determined by machine learning models.
The method and system achieve consistent optical yield and output across the crystal, reducing production costs and improving machining efficiency by automating the process to meet performance requirements.
Smart Images

Figure US20260355896A1-D00000_ABST
Abstract
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a Continuation of International Application No. PCT / CN2023 / 141477, filed on 25 Dec. 2023, the entire contents of which are incorporated herein by reference.TECHNICAL FIELD
[0002] The present disclosure generally relates to the field of crystal machining, and in particular to a method and a system for crystal machining.BACKGROUND
[0003] A scintillation crystal refers to a crystal that, upon impact of high-energy particles such as X-rays, converts the kinetic energy of the high-energy particles into light energy to emit flashes. Scintillation crystals are widely used in nuclear medical imaging, such as X-ray computed tomography (XCT) and positron emission tomography (PET), nuclear detection technologies, such as industrial computed tomography (industrial CT), oil well exploration, nuclear physics, high-energy physics, environmental monitoring, security inspection, weapon fire control, guidance, and the like. In the fields of high-energy physics and nuclear medical imaging, scintillation crystals are required to have a high optical yield, strong gamma-ray absorption capability, short luminescence decay time, high radiation hardness, high density, high atomic number, etc. However, due to the large atomic radius of luminescent atoms in most scintillation crystals and the significant difference in atomic radii among different elements in the crystal, the segregation coefficient (K value) of certain elements in the crystal is low (e.g., far less than 1). Consequently, a deviation in the doping concentration of the elements between the head and tail of the crystal is large, which further leads to a deviation in the optical output at various points within the crystal, greatly affecting the uniformity of the scintillation performance of the crystal.
[0004] Since the crystal growth process, growth environment, volatilization conditions, and temperature field uniformity are difficult to be completely consistent, the uniformity of the crystal is difficult to guarantee, and it is also difficult to meet the requirement for consistency in the optical output of the crystal. Moreover, due to the existence of a temperature gradient during the crystal growth process, the thermal stress of the crystal is also relatively high, and crystal cracking often occurs during traditional machining processes.
[0005] Therefore, there is a need to provide a method for crystal machining and a system for crystal machining to reduce the crystal cracking rate and improve the consistency of optical output, thereby meeting the requirements for crystal performance and machining, and reducing production costs.SUMMARY
[0006] One or more embodiments of the present disclosure provide a method for crystal machining. The method includes: obtaining at least one of a target optical yield or a target optical output value of a crystal; determining a light output surface and a non-light-output surface of the crystal; determining a first surface parameter of the light output surface and a second surface parameter of the non-light-output surface based on at least one of the target optical yield or the target optical output value; and machining the crystal based on the first surface parameter and the second surface parameter.
[0007] One or more embodiments of the present disclosure provide a system for crystal machining. The system includes: an input module configured to obtain at least one of a target optical yield or a target optical output value of a crystal; a determination module configured to determine a light output surface and a non-light-output surface of the crystal; a processing module configured to determine a first surface parameter of the light output surface and a second surface parameter of the non-light-output surface based on at least one of the target optical yield or the target optical output value; and a machining module configured to machine the crystal based on the first surface parameter and the second surface parameter.
[0008] One or more embodiments of the present disclosure further provide a system for crystal machining. The system includes: a control host and a machining device, wherein the control host is configured to: obtain at least one of a target optical yield or a target optical output value of a crystal; determine a light output surface and a non-light-output surface of the crystal; determine a first surface parameter of the light output surface and a second surface parameter of the non-light-output surface based on at least one of the target optical yield or the target optical output value; and control the machining device to machine the crystal based on the first surface parameter and the second surface parameter.
[0009] Some embodiments of the present disclosure at least include the following beneficial effects: (1) By machining the light output surface based on the first surface parameter and / or machining the non-light-output surface based on the second surface parameter, the optical yield and / or optical output of each crystal can be made consistent by changing surface parameters of the crystal (e.g., surface roughness, etc.). (2) The surface parameters of the non-light-output surface and the light output surface of the crystal are automatically determined. Based on the determined surface parameters, polishing parameters, and grinding parameters are determined to meet both crystal performance requirements and machining requirements, thereby achieving fully automated machining of the crystal and reducing production costs.BRIEF DESCRIPTION OF THE DRAWINGS
[0010] This specification will be further explained by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, the same reference numerals denote the same structures, wherein:
[0011] FIG. 1 is a block diagram illustrating an exemplary system for crystal machining according to some embodiments of the present disclosure;
[0012] FIG. 2A is a schematic diagram illustrating an exemplary system for crystal machining according to some embodiments of the present disclosure;
[0013] FIG. 2B is a schematic diagram illustrating an exemplary machining device according to some embodiments of the present disclosure;
[0014] FIG. 3 is a flowchart illustrating an exemplary process for crystal machining according to some embodiments of the present disclosure;
[0015] FIG. 4 is a schematic diagram illustrating an exemplary process for crystal machining according to some embodiments of the present disclosure;
[0016] FIG. 5 is a flowchart illustrating an exemplary process for determining machining quality according to some embodiments of the present disclosure;
[0017] FIG. 6 is a schematic diagram illustrating an exemplary process for preprocessing according to some embodiments of the present disclosure.DETAILED DESCRIPTION
[0018] In order to more clearly describe the technical solutions of the embodiments of the present disclosure, a brief introduction will be given to the accompanying drawings to be used in the description of the embodiments. Apparently, the accompanying drawings in the following description are merely some examples or embodiments of the present disclosure, and for a person of ordinary skill in the art, without inventive efforts, the present disclosure may also be applied to other similar scenarios based on these drawings. Unless apparent from the context or otherwise specified, identical reference numerals in the drawings represent identical structures or operations.
[0019] It should be understood that the terms ‘system’, ‘device’, ‘unit’, and / or ‘module’ as used herein are a way for distinguishing different components, elements, parts, portions, or assemblies of different levels. However, if other terms can achieve the same purpose, the terms may be replaced by other expressions.
[0020] As shown in the present disclosure and the claims, unless the context clearly indicates otherwise, words such as ‘a’, ‘an’, ‘one’, and / or ‘the’ do not specifically refer to the singular, and may also include the plural. Generally speaking, the terms ‘comprising’ and ‘including’ only indicate that explicitly identified steps and elements are included, and these steps and elements do not constitute an exclusive enumeration, and a method or a device may also include other steps or elements.
[0021] Flowcharts are used in the present disclosure to illustrate operations performed by a system according to embodiments of the present disclosure. It should be understood that the preceding or subsequent operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. Meanwhile, other operations may also be added to these processes, or one or more operations may be removed from these processes.
[0022] FIG. 1 is a block diagram illustrating an exemplary system for crystal machining according to some embodiments of the present disclosure.
[0023] As shown in FIG. 1, in some embodiments, the system for crystal machining 100 may include an input module 110, a determination module 120, a processing module 130, and a machining module 140.
[0024] In some embodiments, the input module 110 is configured to obtain a target optical yield and / or a target optical output value of a crystal.
[0025] In some embodiments, the determination module 120 is configured to determine a light output surface and a non-light-output surface of the crystal. In some embodiments, the determination module 120 may be further configured to obtain an initial optical yield and / or an initial optical output value of each surface of the crystal, and determine the light output surface and the non-light-output surface based on the initial optical yield of each surface and the target optical yield; and / or determine the light output surface and the non-light-output surface based on the initial optical output value of each surface and the target optical output value.
[0026] In some embodiments, the processing module 130 is configured to determine a first surface parameter of the light output surface and a second surface parameter of the non-light-output surface based on the target optical yield and / or the target optical output value. In some embodiments, the processing module 130 may be further configured to obtain the initial optical yield and / or the initial optical output value of the light output surface, and an initial surface parameter, and determine the first surface parameter based on the initial optical yield of the light output surface, the initial surface parameter, and the target optical yield; and / or determine the first surface parameter based on the initial optical output value of the light output surface, the initial surface parameter, and the target optical output value. In some embodiments, the processing module 130 may be configured to determine the second surface parameter based on the first surface parameter and the target optical yield; and / or determine the second surface parameter based on the first surface parameter and the target optical output value.
[0027] In some embodiments, the machining module 140 is configured to machine the crystal based on the first surface parameter and the second surface parameter. In some embodiments, the machining module 140 may be further configured to determine a polishing parameter based on the first surface parameter, and machine the light output surface based on the polishing parameter. In some embodiments, the machining module 140 may also be configured to determine a grinding parameter based on the second surface parameter, and machine the non-light-output surface based on the grinding parameter. In some embodiments, the machining module 140 may be further configured to determine the polishing parameter and / or the grinding parameter through a machine learning model, and the grinding parameter includes a mesh size of abrasive grit.
[0028] In some embodiments, the machining module 140 may also be configured to obtain a vibration response of a polishing plate and / or a grinding plate during machining, and determine a machining quality of the light output surface and / or the non-light-output surface based on the vibration response. In some embodiments, the machining module 140 may be further configured to extract a characteristic frequency band from the vibration response and identify the characteristic frequency band to determine whether the machined crystal has a defect. In some embodiments, the machining module 140 may be further configured to process the characteristic frequency band through an anomaly detection model to determine whether the machined crystal has a defect, and the anomaly detection model is a machine learning model. In some embodiments, in response to the machining quality not satisfying a preset condition, the machining module 140 may also be configured to stop machining or output prompt information. In some embodiments, the machining module 140 may also be configured to obtain the vibration response and a reference vibration response of the polishing plate and / or the grinding plate during machining, and in response to a difference of at least one frequency response parameter between the vibration response and the reference vibration response not satisfying a preset response condition, stop machining or output prompt information.
[0029] In some embodiments, the machining module 140 may also be configured to machine the non-light-output surface based on a first grinding pressure, and obtain a first grinding feedback; in response to the first grinding feedback satisfying a first pressurization condition, machine the non-light-output surface based on a second grinding pressure, and obtain a second grinding feedback; in response to the second grinding feedback satisfying the first pressurization condition, machine the non-light-output surface based on a third grinding pressure; and the first grinding pressure is less than the second grinding pressure, and the second grinding pressure is less than the third grinding pressure. In some embodiments, each of the first grinding feedback and the second grinding feedback comprises a torque of a motor providing drive for the grinding plate. In some embodiments, each of the first grinding feedback and the second grinding feedback comprises a grinding efficiency, the grinding efficiency is determined based on a grinding amount and / or a number of grinding revolutions, and the grinding amount is obtained by a laser rangefinder disposed above the grinding plate.
[0030] In some embodiments, the machining module 140 may also be configured to machine the light output surface based on a first polishing pressure, and obtain a first polishing feedback; in response to the first polishing feedback satisfying a second pressurization condition, machine the light output surface based on a second polishing pressure, and obtain a second polishing feedback; in response to the second polishing feedback satisfying the second pressurization condition, machine the light output surface based on a third polishing pressure; and the first polishing pressure is less than the second polishing pressure, and the second polishing pressure is less than the third polishing pressure.
[0031] In some embodiments, the machining module 140 may also be configured to preprocess an initial blank to obtain the crystal. In some embodiments, the preprocessing may include a shaping process, and the machining module 140 may be further configured to arrange a plurality of initial blanks and grind cross-sections of the plurality of arranged initial blanks. In some embodiments, the preprocessing may include a combination process, and the machining module 140 may also be configured to bond the plurality of initial blanks to form a combined blank, and further bond a protective layer to the combined blank to obtain the crystal.
[0032] In some embodiments, the system for crystal machining 100 may also include a detection module. The detection module may be configured to detect the crystal or the initial blank. In some embodiments, the detection module may include an image recognition module and a measurement module. The image recognition module may be configured to detect a type and an appearance of the crystal or the initial blank. The measurement module may be configured to detect a surface roughness, the initial optical yield, and / or the initial optical output value of the crystal or the initial blank.
[0033] In some embodiments, the system for crystal machining 100 may be integrated into a processor. The processor may process data and / or information obtained from other devices or system components. The processor may execute program instructions based on the data, the information, and / or processing results to execute one or more functions described in the present disclosure. Merely by way of example, the processor may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or the like, or any combination thereof.
[0034] More descriptions regarding the various modules may be found elsewhere in the present disclosure (e.g., FIGS. 3 to 5 and related descriptions thereof).
[0035] It should be noted that the above description of the system for crystal machining and the modules thereof is merely for convenience of description, and does not limit the present disclosure to the scope of the illustrated embodiments. It can be understood that for a person skilled in the art, after understanding the principle of the system, the various modules may be arbitrarily combined, or a subsystem may be formed and connected with other modules, without departing from the principle. In some embodiments, the input module 110, the determination module 120, the processing module 130, and the machining module 140 disclosed in FIG. 1 may be different modules in a system, or a module may implement functions of two or more of the above-mentioned modules. For example, the various modules may share a storage module, or each of the various modules may have its own storage module. Variations of this kind are all within the scope of protection of the present disclosure.
[0036] FIG. 2A is a structural schematic diagram illustrating another exemplary system for crystal machining according to some embodiments of the present disclosure.
[0037] As shown in FIG. 2A, in some embodiments, a system for crystal machining 200 may include a control host 210, a machining device 220, and a roughness detection device 230.
[0038] The control host 210 may control the machining device 220 or the roughness detection device 230 to perform the method for crystal machining described in embodiments of the present disclosure. In some embodiments, the control host 210 may be a processor.
[0039] In some embodiments, the control host 210 is configured to obtain at least one of a target optical yield or a target optical output value of the crystal; determine a light output surface and a non-light-output surface of the crystal; determine a first surface parameter of the light output surface and a second surface parameter of the non-light-output surface based on at least one of the target optical yield or the target optical output value; and control the machining device to machine the crystal based on the first surface parameter and the second surface parameter. More descriptions regarding the method for crystal machining may be found elsewhere in the present disclosure (e.g., FIGS. 3-5 and related descriptions thereof).
[0040] The machining device 220 is a device configured to machine the crystal. In some embodiments, the machining device 220 may perform grinding processing, polishing processing, and other crystal-related machining processing, or the like.
[0041] In some embodiments, the grinding processing may refer to a processing operation of finishing a surface of the crystal, wherein the processing operation is performed by coating or embedding abrasive particles on a grinding tool and by relative movement between the grinding tool and the crystal under a certain pressure. Through grinding processing, the surface flatness of the crystal may be adjusted. The grinding surface refers to a surface obtained through grinding processing. In some embodiments, one or two end faces of the crystal may be subjected to grinding processing, to obtain one or two grinding surfaces. An end face refers to a plane at either end of the crystal along a length direction. In some embodiments, all surfaces of the crystal may be subjected to grinding processing, to obtain a plurality of grinding surfaces. For example, when the crystal is a cuboid, six surfaces of the crystal may be subjected to grinding processing. In some embodiments, a manner of grinding processing may include one or more of single-sided grinding, double-sided grinding, or the like. A grinding fluid may comprise abrasive medium having particle sizes of 100 mesh to 2500 mesh, preferably 100 mesh to 2000 mesh. Types of abrasive medium may be alumina, silica, ceria, silicon carbide, boron carbide, diamond, or the like.
[0042] In some embodiments, the polishing processing may refer to a processing operation of machining a surface of the crystal by mechanical, chemical, electrochemical manners, or the like, thereby reducing a surface roughness of the crystal to obtain a bright and flat surface. In some embodiments, at least one surface of the crystal may be subjected to polishing processing. For example, the machining device 220 may subject one or two end faces of the crystal to polishing processing. The polishing surface refers to a surface obtained through polishing processing. In some embodiments, polishing processing may be performed on a grinding surface to obtain the polishing surface. In some embodiments, polishing processing may also be performed on a non-grinding surface to obtain the polishing surface. In some embodiments, polishing processing may be performed on an end face to obtain a polishing surface. The polishing processing may use the polishing fluid with a finer abrasive medium to polish the surface of the crystal to a bright finish. The polishing process may be similar to the grinding process. A polishing fluid may comprise polishing medium particle sizes of 5 μm to 0.001 μm, and types of the polishing medium may be alumina, silica, ceria, silicon carbide, boron carbide, diamond, or the like. In some embodiments, before polishing, it may be necessary to attach a polishing pad to each of an upper grinding plate and a lower grinding plate, and the polishing pad may be a polyurethane material, or the like.
[0043] In some embodiments, the machining device 220 may also perform the preprocessing of the crystal, and the preprocessing may include the shaping processing and the combining processing. The shaping processing refers to performing grinding processing on a cross section of an initial blank after arranging and placing the initial blank. The combining processing refers to bonding initial blanks to form a combined blank, and bonding a protective layer to the combined blank. More descriptions regarding the shaping processing and the combining processing may be found elsewhere in the present disclosure (e.g., FIG. 3 and related descriptions thereof).
[0044] FIG. 2B is a structural schematic diagram illustrating an exemplary machining device according to some embodiments of the present disclosure.
[0045] As shown in FIG. 2B, in some embodiments, the machining device 220 includes a grinding plate and a polishing plate. In some embodiments, the grinding plate and the polishing plate may be disposed in two machining regions, to perform grinding processing and polishing processing in the two machining regions, respectively. In some embodiments, the grinding plate and the polishing plate may be disposed in the same machining region. When grinding processing is to be performed, the grinding plate is installed in the machining region; and when polishing processing is to be performed, the polishing plate is installed in the machining region.
[0046] As shown in FIG. 2B, in some embodiments, the grinding plate includes an upper grinding plate 221 and a lower grinding plate 223, and the polishing plate includes an upper polishing plate 222 and a lower polishing plate 224. The grinding plate is configured to perform grinding processing, and the polishing plate is configured to perform polishing processing. When grinding processing is performed, the crystal 225 may be placed between the upper grinding plate 221 and the lower grinding plate 223. When polishing processing is performed, the crystal 225 may be placed between the upper polishing plate 222 and the lower polishing plate 224. In some embodiments, the machining device 220 may also include a rotating shaft 226, wherein at least one of the upper grinding plate 221 or the upper polishing plate 222 is connected to the rotating shaft 226 of the machining device 220. In some embodiments, an upper side surface of at least one of the upper grinding plate 221 or the upper polishing plate 222 may be provided with at least one vibration sensor 227, and the vibration sensor 227 is configured to obtain a vibration response of the at least one of the upper grinding plate 221 or the upper polishing plate 222 during machining. More descriptions regarding the vibration response may be found elsewhere in the present disclosure (e.g., FIG. 5 and related descriptions thereof).
[0047] As shown in FIG. 2B, in some embodiments, the machining device 220 may also include a laser rangefinder 228 and a slip ring 229, and the laser rangefinder 228 is configured to measure a displacement of the upper grinding plate 221 or the upper polishing plate 222, thereby obtaining a grinding amount or a polishing amount. The slip ring 229 may be configured to lead out signal wires, thereby preventing the signal wires from tangling due to rotation.
[0048] When machining processing is performed, the upper grinding plate and the lower grinding plate are used in cooperation to perform grinding processing on the crystal; and the upper polishing plate and the lower polishing plate are used in cooperation to perform polishing processing on the crystal. For example, when the crystal is fixed between the upper grinding plate (or the upper polishing plate) and the lower grinding plate (or the lower polishing plate), and when the upper grinding plate (or the upper polishing plate) and / or the lower grinding plate (or the lower polishing plate) rotate, a position where the upper grinding plate (or the upper polishing plate) and / or the lower grinding plate (or the lower polishing plate) contact the crystal is rubbed. In some embodiments, a lower side surface of the upper grinding plate (e.g., a side surface of the upper grinding plate contacting the crystal) and an upper side surface of the lower grinding plate (e.g., a side surface of the lower grinding plate contacting the crystal) may be coated or embedded with abrasive material. e.g., abrasive sand or grinding fluid. In some embodiments, polishing media may be placed on the polishing plate, and surfaces of a lower side surface of an upper polishing plate and an upper side surface of a lower polishing plate may be coated with polishing material (e.g., polishing fluid, or the like).
[0049] In some embodiments, the system for crystal machining 200 includes the roughness detection device 230. The roughness detection device 230 is configured to detect a surface roughness of the crystal. For example, the roughness detection device 230 may detect the surface roughness of each surface of the crystal. In some embodiments, the roughness detection device 230 may detect the surface roughness of the crystal before and after grinding, during grinding, before and after polishing, and / or during polishing. In some embodiments, the roughness detection device 230 may transmit a detected value of the surface roughness to the control host 210, wherein the control host 210 may determine the first surface parameter and the second surface parameter based on the detected value of the surface roughness. The roughness detection device 230 may be any device capable of detecting a surface roughness of an object (e.g., a surface roughness measuring instrument, or the like), and is not limited herein.
[0050] In some embodiments, the system for crystal machining 200 may also include an image recognition device, an optical yield detection device, and an optical output value detection device. The image recognition device may be configured to detect a type and an appearance of the crystal or an initial blank. The optical yield detection device may be configured to detect an initial optical yield of each surface of the crystal. The optical output value detection device may be configured to detect an initial optical output value of each surface of the crystal. The image recognition device may be any device capable of recognizing and acquiring images and performing recognition, and the optical output value detection device may also be any device capable of detecting an optical output value. More descriptions regarding the initial optical yield and the initial optical output value may be found elsewhere in the present disclosure (e.g., FIG. 3 and related descriptions thereof).
[0051] FIG. 3 is a flowchart illustrating an exemplary process for crystal machining according to some embodiments of the present disclosure. As shown in FIG. 3, process 300 comprises the following steps. In some embodiments, the flow 300 may be performed by the system for crystal machining 100 or the system for crystal machining 200.
[0052] In 310, at least one of a target optical yield or a target optical output value of the crystal may be obtained. In some embodiments, operation 310 may be performed by the input module 110 or the control host 210.
[0053] The crystal refers to a crystal planned for machining. In some embodiments, the crystal may be a crystal to be machined with respect to the size and shape, and surface parameters (e.g., surface roughness, or the like). The machining refers to one or more processing operations that change the size and shape, and surface parameters of a crystal. In some embodiments, the crystal may include a cerium bromide crystal, a cerium-doped lanthanum bromide crystal, a cerium-doped lanthanum chloride crystal, a silicate scintillator crystal, a garnet scintillator crystal, or the like, or a combination thereof. The shape of the crystal may vary. For example, the crystal may be a crystal rod, a crystal block, or a crystal of other shapes.
[0054] The optical yield refers to the ability of a crystal to convert gamma photons into visible light. The optical output value refers to a ratio of a total number of photons emitted by a crystal to incident radiation energy absorbed by the crystal. The target optical yield refers to an optical yield that the crystal is expected to achieve. The target optical output value refers to an optical output value that the crystal is expected to achieve. The target optical yield and / or the target optical output value may be determined by user input, and may be set according to actual production requirements.
[0055] In some embodiments, the crystal may be a crystal obtained after the preprocessing. The preprocessing may be performed by the machining module 140. In some embodiments, the machining module 140 or the control host 210 may control the machining device 220 to preprocess at least one initial blank to obtain the crystal.
[0056] An initial blank refers to a crystal raw material that has not been machined. The crystal raw material may be a crystal rod, a crystal block, or crystal raw material of other shapes grown in a crystal growth device. The crystal raw material may have microcracks or small chipping during the growth. To avoid such appearance defects, when the crystal raw material is divided, it may result in crystal blocks of varying sizes being divided out. To make the appearance and dimensions of the respective crystal blocks consistent or to avoid appearance defects before grinding, it is necessary to preprocess the initial blank.
[0057] In some embodiments, the preprocessing may include shaping treatment. The shaping treatment refers to a processing manner that treats surface defects of the initial blank and maintains consistent crystal dimensions.
[0058] In some embodiments, the shaping treatment includes directly performing grinding treatment on a defective surface of a single initial blank to remove obvious defects on the surface.
[0059] In some embodiments, the shaping treatment includes arranging the plurality of initial blanks, and grinding the cross-sections of the arranged plurality of initial blanks. In some embodiments, the plurality of initial blanks with similar dimensions and / or similar optical yields may be arranged, and cross-sections of the arranged plurality of initial blanks may be ground. The similar dimensions refer to a case where a dimension deviation between the initial blanks is less than a corresponding dimension deviation threshold (e.g., 0.1 mm, or the like). For example, when a length deviation, a width deviation, and a height deviation of the plurality of initial blanks are respectively less than the corresponding dimension deviation threshold, it may be determined that the plurality of initial blanks have similar dimensions. The dimensions of the initial blank may be obtained by measurement. The similar optical yields refers to a case where an optical yield deviation between the initial blanks is less than a corresponding optical yield deviation threshold (e.g., 3600 Ph / Mev, or the like). The optical yield of the initial blank may be obtained by measurement using a scintillation detector system.
[0060] In some embodiments, before performing the preprocessing, an appearance of the initial blank also may be detected. For example, a detection module or an image recognition device may detect the type and the appearance of the initial blank. In some embodiments, the detection module or the image recognition device may detect whether a surface of the initial blank has cracks or notches, or the like. Further, a preprocessing module may preprocess the initial blank based on detection results. For example, the preprocessing module may grind a surface that has cracks or notches.
[0061] In some embodiments, defective surfaces of the plurality of initial blanks may be placed on the same side, and an adjacent surface corresponding to the defective surface of each of at least a portion of the plurality of initial blanks may be sequentially overlapped, thereby achieving an arrangement of the plurality of initial blanks.
[0062] In some embodiments, the plurality of initial blanks also may be arranged along axes of the initial blanks. For example, axes of the plurality of initial blanks are parallel after being arranged. Descriptions regarding the arrangement are merely exemplary descriptions and do not constitute a limitation on embodiments.
[0063] In some embodiments, each initial blank may be separated by a protective layer (e.g., a polyurethane skin, or the like). The protective layer between the initial blanks may prevent problems such as scratches or chipping, or the like, of the initial blanks during machining and grinding.
[0064] The cross-section refers to a surface perpendicular to an arrangement direction of the plurality of initial blanks, and is also a surface to be ground. In some embodiments, the cross-section may be a ‘large surface’ formed by arranging defective surfaces of the plurality of initial blanks.
[0065] Before grinding, the machining module 140 or the control host 210 may draw a reference grinding line (a virtual line) that spans each initial blank. The reference grinding line is parallel to the cross-sections of the arranged plurality of initial blanks. In some embodiments, the machining module 140 or the control host 210 may draw the reference grinding line based on defects existing in the plurality of initial blanks. For example, the reference grinding line may be drawn according to a defect degree of the initial blank (e.g., scratches, chipping, or cracking). The reference grinding line needs to be below a defect location of the initial blank. During the grinding process, the machining device 220 may perform grinding treatment until a cross-section position is ground to the reference grinding line. Merely by way of example, if a surface of the initial blank is detected to have one or more cracks or notches, the reference grinding line may be drawn below the one or more notches. After grinding to the reference grinding line, the one or more cracks or the one or more notches have been ground away, and defects of the initial blank are removed.
[0066] Merely by way of example, a manner of performing shaping treatment using the machining device 220 may include operations S1 and S2 as follows.
[0067] In S1, a plurality of initial blanks may be arranged according to operations S11-S14 as follows.
[0068] Operation S11 may include performing dimension measurement at both ends of a multi-line or an inner circle cutting surface of each initial blank, marking a highest point position of each initial blank (e.g., a highest point position in a certain surface), and using a surface with a greater defect degree (e.g., scratches, chipping, or cracking, or the like) as a cross-section of each initial blank.
[0069] Operation S12 may include placing limit blocks in three directions on the grinding machine workbench, sequentially placing the initial blanks on the grinding machine workbench, and placing the highest point positions in the same direction. One polyurethane skin may be placed between two adjacent initial blanks to prevent chipping caused by friction between the adjacent initial blanks.
[0070] Operation S13 may include sequentially limiting the initial blanks individually using the limit blocks, mutually pressing at both left and right ends of the initial blanks, thereby reducing the generation of gaps.
[0071] Operation S14 may include turning on a magnetic switch on a device control panel, and magnetically attracting the limit blocks on the grinding machine workbench.
[0072] In S2, the cross-sections of the arranged plurality of initial blanks may be ground according to operations S21-S23 as follows.
[0073] Operation S21 may include drawing the reference grinding line on the combined blank, moving a grinding wheel to a position 1 to 2 mm above a highest point of the combined blank; then moving the grinding wheel to a grinding position starting point (e.g., the highest point of the combined blank), requiring a horizontal distance between the grinding wheel and the crystal of 5 to 10 mm, to prevent damage to a grinding wheel surface.
[0074] Operation S22 may include determining processing parameters. A total grinding amount is set according to actual conditions; a grinding amount is 0.005 to 0.025 mm; a number of fine grinding passes is 1 to 3; a grinding width is a sum of a maximum length of the combined blank and 10 mm; and a mesh size of abrasive grit is 100 to 2500 mesh.
[0075] Operation S23 may include starting a grinding machine to perform a shaping treatment, and using an angle fixture to perform angle control during the shaping treatment.
[0076] After completing the shaping treatment for a single surface of the initial blank, the foregoing steps may be repeated for other surfaces to perform the shaping treatment.
[0077] In some embodiments, the preprocessing may include combined treatment. The combined treatment refers to bonding the plurality of initial blanks (e.g., initial blanks that have been or have not been subjected to shaping treatment) to form a combined blank. The bonding surface refers to an adjacent surface of a surface to be machined. For example, the plurality of initial blanks may be bonded by glue or a bonding medium. Each initial blank in the formed combined blank may be ground or polished together. In some embodiments, the machining module 140 or the control host 210 may control the machining device 220 to further bond a protective layer to the combined blank to obtain the crystal. For example, a protective layer may be bonded to each surface of the combined blank other than a machining surface. The hardness of the protective layer is higher than that of the initial blank to achieve a purpose of protecting the initial blank during machining. As shown in FIG. 6, a group of initial blanks 610 may form a combined blank 620 by bonding. A protective layer 630 may be bonded to peripheries of the combined blank 620 (e.g., other side surfaces except for a surface to be machined), thereby obtaining a crystal 640. For example, protective layers of the same size may be bonded to opposite surfaces, or protective layers may be bonded around the peripheries of the combined blank. The protective layer may be a polyurethane skin or other protective materials, and is not limited herein. The crystal obtained by the combined treatment is a combination of the plurality of initial blanks. Grinding these initial blanks together may reduce repetitive operations to improve machining efficiency, and also may make the dimensions of the machined crystal more uniform.
[0078] In some embodiments, the crystal also may be an unprocessed crystal raw material, and a crystal machining process described in embodiments of the present disclosure may be directly performed.
[0079] The initial blank after the preprocessing may remove defects that may exist in the crystal raw material and maintain the uniformity of the dimensions of a plurality of crystals. By adding a protective layer, the crystal also may be protected during subsequent machining, thereby improving machining quality and machining efficiency of the crystal.
[0080] In 320, at least one light output surface and at least one non-light-output surface of the crystal may be determined. In some embodiments, operation 320 may be executed by the determination module 120 or the control host 210.
[0081] A light output surface refers to a surface of the crystal on which light output is planned to be achieved. In some embodiments, the number of the at least one light output surface may be one or two. If there are two light output surfaces, the two light output surfaces may be opposite surfaces. Merely by way of example, if the crystal is a cuboid, then the light output surface of the crystal may be a top surface or a bottom surface of the cuboid. In some embodiments, the number of light output surfaces may be manually determined or input.
[0082] The at least one non-light-output surface refers to other surfaces of the crystal excluding the at least one light output surface. Merely by way of example, if the light output surface of the crystal is the top surface or the bottom surface of the cuboid, then the non-light-output surface of the crystal may be the four side surfaces of the cuboid.
[0083] In some embodiments, the determination module 120 or the control host 210 may determine the at least one light output surface and the at least one non-light-output surface of the crystal based on manual input.
[0084] In some embodiments, the determination module 120 or the control host 210 may determine the at least one light output surface and the at least one non-light-output surface based on the initial optical yield of each surface of the crystal and the target optical yield. In some embodiments, the determination module 120 or the control host 210 may obtain the initial optical yield of each surface of the crystal, and determine the at least one light output surface and the at least one non-light-output surface based on the initial optical yield of each surface and the target optical yield.
[0085] The initial optical yield refers to an optical yield of one surface of the crystal. Since the surface condition of each surface of the crystal is different, the initial optical yield corresponding to each surface may be different. In some embodiments, the initial optical yield of each surface of the crystal may be detected by the measurement module. In some embodiments, the optical yield (in units of phe / MeV) may be measured by a scintillation detector system.
[0086] In some embodiments, the determination module 120 or the control host 210 may determine the at least one light output surface and the at least one non-light-output surface based on the initial optical output value of each surface of the crystal and the target optical output value. In some embodiments, the determination module 120 or the control host 210 may obtain the initial optical output value of each surface of the crystal, and determine the at least one light output surface and the at least one non-light-output surface based on the initial optical output value of each surface and the target optical output value.
[0087] The initial optical output value refers to an optical output value of one surface of the crystal. For example, the initial optical output value may be a ratio of a total number of photons emitted by one surface of the crystal to incident radiation energy absorbed by the surface. Since the surface condition of each surface of the crystal is different, the initial optical output value corresponding to each surface may be different. In some embodiments, the initial optical output value of each surface of the crystal may be detected by the measurement module. In some embodiments, the light output performance of the crystal may be detected according to measurement manners in relevant industry standards. For example, in combination with the T / CAB 0180-2022 standard (standard name: ‘Measurement Method for Gadolinium Aluminum Gallium Garnet (GAGG) Crystal and Wafer Array Performance’) issued by the China Industry-University-Research Institute Collaboration Association, a full absorption peak manner and a Compton edge manner may be adopted for measurement. The measurement principle is as follows: when monoenergetic gamma radiation enters a scintillation detector, a distribution of an amplitude of an output pulse thereof is mainly composed of spectrum segments such as a Compton distribution and a full absorption peak (except for scintillators with low atomic number), or the like. The full absorption peak manner and the Compton edge manner respectively take the full absorption peak or a Compton distribution edge amplitude as a measure for determining the light output of the scintillator. In some embodiments, the optical output value may also be obtained through a correspondence relationship between the optical yield and the optical output value, wherein the correspondence relationship may be determined based on an experiment.
[0088] When the number of the at least one light output surfaces varies, the determination manner of the light output surfaces is also different.
[0089] In some embodiments, in response to the at least one light output surface being one, the determination module 120 or the control host 210 may determine the surface having an initial optical yield closest to the target optical yield as the light output surface, and the remaining surfaces as the non-light-output surfaces. In some embodiments, in response to the at least one light output surface being one, the determination module 120 or the control host 210 may determine the surface having an initial optical output value closest to the target optical output value as the light output surface, and the remaining surfaces as the non-light-output surfaces. In some embodiments, in response to the at least one light output surface being one, the determination module 120 or the control host 210 may determine the surface having an initial optical yield closest to the target optical yield and an initial optical output value closest to the target optical output value as the light output surface, and the remaining surfaces as the non-light-output surfaces.
[0090] In some embodiments, in response to the at least one light output surface being two opposite surfaces, the determination module 120 or the control host 210 may determine an average value of initial optical yields of each two opposite surfaces, and determine the two opposite surfaces having the average value of initial optical yields closest to the target optical yield as the light output surfaces, and the remaining surfaces as the non-light-output surfaces. In some embodiments, in response to the light output surface being two opposite surfaces, the determination module 120 or the control host 210 may determine an average value of initial optical output values of each two opposite surfaces, and determine the two opposite surfaces having the average value of initial optical output values closest to the target optical output value as the light output surfaces, and the remaining surfaces as the non-light-output surfaces. In some embodiments, in response to the light output surface being two opposite surfaces, the determination module 120 or the control host 210 may determine an average value of initial optical yields and an average value of initial optical output values of each two opposite surfaces, and determine the two opposite surfaces having the average value of initial optical yields closest to the target optical yield and the average value of initial optical output values closest to the target optical output value as the light output surfaces, and the remaining surfaces as the non-light-output surfaces. In some embodiments, the at least one light output surface and the at least one non-light-output surface may also be determined according to other manners. The present disclosure is not limited to the determination manner of the at least one light output surface and the at least one non-light-output surface.
[0091] According to some embodiments of the present disclosure, by using the target optical yield and the initial optical yield of each surface, and / or the target optical output value and the initial optical output value of each surface, the light output surface and the non-light-output surface with minimum machining demand may be determined, thereby increasing machining efficiency. When there are two opposite light output surfaces, determining the light output surfaces based on an average value of the optical yield or the optical output value may effectively avoid a problem of uneven light output.
[0092] In 330, a first surface parameter of the at least one light output surface and a second surface parameter of the at least one non-light-output surface may be determined based on at least one of the target optical yield or the target optical output value. In some embodiments, operation 330 may be performed by the processing module 130 or the control host 210.
[0093] The surface parameter refers to a parameter related to a surface condition of the crystal. In some embodiments, the surface parameter may include at least one of surface roughness (Ra), warpage (Wap), total thickness variation (TTV), perpendicularity, parallelism, or the like.
[0094] The first surface parameter refers to a parameter value that the surface parameter of the at least one light output surface is expected to reach after completing machining. The first surface parameters of different light output surfaces may be the same or different.
[0095] In some embodiments, the first surface parameter of a light output surface may include a target surface roughness of the light output surface. The target surface roughness of the light output surface refers to a parameter value that the Ra of the light output surface is expected to reach after completing machining.
[0096] The second surface parameter refers to a parameter value that the surface parameter of the at least one non-light-output surface is expected to reach after completing machining. The second surface parameters of different non-light-output surfaces may be the same or different.
[0097] In some embodiments, the second surface parameter of a non-light-output surface may include a target surface roughness of the non-light-output surface. The target surface roughness of the non-light-output surface refers to a parameter value that the Ra of the non-light-output surface is expected to reach after completing machining.
[0098] In some embodiments, the processing module 130 or the control host 210 may determine the first surface parameter and the second surface parameter in a plurality of manners based on at least one of the target optical yield or the target optical output value. For example, the processing module 130 or the control host 210 may determine the surface parameters (including the first surface parameter and / or the second surface parameter) through historical machining data. The processing module 130 or the control host 210 may determine, from the historical machining data, a portion of the historical machining data. The portion of the historical machining data may have a historical optical yield and / or a historical target optical output value that is similar to or the same as the current target optical yield and / or the current target optical output value. The processing module 130 or the control host 210 may determine the historical first surface parameter and the historical second surface parameter in the portion of the historical machining data as the first surface parameter and the second surface parameter corresponding to the current target optical yield and / or the current target optical output value, respectively.
[0099] In some embodiments, the control host 210 may first determine the first surface parameter and then determine the second surface parameter based on the first surface parameter.
[0100] In some embodiments, the processing module 130 or the control host 210 may obtain at least one of an initial optical yield or an initial optical output value of the light output surface, and an initial surface parameter of the light output surface; determine the first surface parameter based on the initial optical yield of the light output surface, the initial surface parameter, and the target optical yield; and / or determine the first surface parameter based on the initial optical output value of the light output surface, the initial surface parameter of the light output surface, and the target optical output value.
[0101] The initial surface parameter refers to an initial value of the surface parameter of the light output surface before machining. The initial surface parameter may be obtained by detecting the light output surface of the crystal with a related detection device (e.g., the roughness detection device, or the like).
[0102] In some embodiments, the processing module 130 or the control host 210 may determine the first surface parameter (e.g., a target surface roughness of the light output surface) in a plurality of manners based on the initial surface parameter, the initial optical yield, and the target optical yield of the light output surface. For example, a correspondence may be preset based on historical data or prior knowledge between different initial surface parameters, different initial optical yields, different target optical yields, and different target surface roughnesses. Based on the correspondence and the initial surface parameter, the initial optical yield, and the target optical yield of the current light output surface, the corresponding target surface roughness may be determined. In some embodiments, the processing module 130 or the control host 210 may determine the first surface parameter (e.g., a target surface roughness of the light output surface) in a plurality of manners based on the initial surface parameter, the initial optical output value, and the target optical output value of the light output surface. For example, a correspondence may be preset based on historical data or prior knowledge between different initial surface parameters, different initial optical output values, different target optical output values, and different target surface roughnesses. Based on the correspondence and the initial surface parameter, the initial optical output value, and the target optical output value of the current light output surface, the corresponding target surface roughness may be determined.
[0103] In some embodiments, the processing module 130 or the control host 210 may determine the first surface parameter through a first determination model. The first determination model is a machine learning model (e.g., a Convolutional Neural Network (CNN) model, or the like).
[0104] In some embodiments, an input of the first determination model may include at least one of the initial optical output value (and / or the initial optical yield), the initial surface parameter, and the target optical output value (and / or the target optical yield) of at least one light output surface, and an output may include the first surface parameter corresponding to the at least one light output surface. In some embodiments, the first determination model may be trained through a plurality of first training samples with first training labels. Merely by way of example, each of the plurality of first training samples with a first label may be input into an initial first determination model, and a loss function may be constructed through the first label and a result of the initial first determination model generated based on the each of the plurality of first training samples. The parameters of the initial first determination model may be iteratively updated based on the loss function. When the loss function of the initial first determination model satisfies a preset condition, the model training is completed, and a trained first determination model is obtained. The preset condition may be convergence of the loss function, the number of iterations reaching a threshold, or the like.
[0105] In some embodiments, a first training sample may include at least one of a sample initial optical output value (and / or a sample initial optical yield) of a sample light output surface of a sample crystal, a sample initial surface parameter of the sample light output surface of the sample crystal, and a sample target optical output value (and / or a sample target optical yield) of the sample crystal, and a first training label of the first training sample may be a first surface parameter corresponding to the sample light output surface. The first training label may be obtained based on manual annotation or may be automatically annotated based on historical data.
[0106] In some embodiments, the processing module 130 or the control host 210 may determine the second surface parameter based on the first surface parameter and the target optical output value (and / or the target optical yield). For example, the processing module 130 or the control host 210 may determine the second surface parameter by combining a number of the light output surfaces, based on the first surface parameter and the target optical output value (and / or the target optical yield).
[0107] In some embodiments, in response to the light output surface being two opposite surfaces, the processing module 130 or the control host 210 may determine the second surface parameter of the non-light-output surface based on the first surface parameter of the light output surface and the target optical output value (and / or the target optical yield) of the crystal. For example, the processing module 130 or the control host 210 may input the first surface parameters of the two opposite light output surfaces into a second determination model, thereby outputting the second surface parameter of the non-light-output surface. More descriptions regarding the second determination model may be found in the following related descriptions.
[0108] In some embodiments, in response to the light output surface being one surface, the processing module 130 or the control host 210 may first determine a second surface parameter of a surrounding non-light-output surface of the light output surface based on the first surface parameter of the light output surface and the target optical output value (and / or the target optical yield) of the crystal, and then determine a second surface parameter of an opposite non-light-output surface of the light output surface based on the first surface parameter of the light output surface and the target optical output value (and / or the target optical yield) of the crystal.
[0109] The surrounding non-light-output surface of the light output surface refers to a non-light-output surface adjacent to the light output surface, and the opposite non-light-output surface of the light output surface refers to a non-light-output surface opposite to the light output surface. Merely by way of example, when the crystal is a cuboid, if the light output surface is a top surface of the crystal, then the second surface parameters of the four side surfaces of the crystal may first be determined, and then the second surface parameter of a bottom surface of the crystal may be determined.
[0110] Some embodiments of the present disclosure may reduce the influence of an opposite non-light-output surface on the light output by distinguishing between the non-light-output surfaces, thereby improving consistency of the light output of the crystal. When the light output surface is a single surface, the influence of the surrounding non-light-output surface and the opposite non-light-output surface of the light output surface on the light output is different. Therefore, by first determining the second surface parameters of the remaining non-light-output surfaces (e.g., the remaining non-light-output surfaces not opposite to the light output surface), and then determining the second surface parameter of the non-light-output surface opposite to the light output surface, the accuracy of determining the second surface parameter may be improved.
[0111] In some embodiments, the processing module 130 or the control host 210 may determine the second surface parameter in a plurality of manners based on the first surface parameter and the target optical output value (and / or the target optical yield).
[0112] In some embodiments, the processing module 130 or the control host 210 may determine the second surface parameter based on the first surface parameter and the target optical output value (and / or the target optical yield) by querying a parameter lookup table. The parameter lookup table includes a correspondence between different first surface parameters, different target optical output values (and / or different target optical yields), and different second surface parameters, and the parameter lookup table may be determined based on historical data or prior knowledge.
[0113] In some embodiments, the processing module 130 or the control host 210 may determine the second surface parameter through the second determination model. The second determination model is a machine learning model (e.g., a CNN model, or the like).
[0114] In some embodiments, an input of the second determination model may include a first surface parameter of at least one light output surface and a target optical output value (and / or a target optical yield) of the crystal, and an output may include a second surface parameter of at least one non-light-output surface. In some embodiments, the second determination model may be trained through a plurality of second training samples with second training labels. The training process of the second determination model is similar to the training process of the first determination model, and more descriptions may be found in related descriptions of the training process of the first determination model.
[0115] In some embodiments, a second training sample may include a sample first surface parameter of at least one sample light output surface and a sample target optical output value (and / or a sample target optical yield) of a sample crystal, and a second training label may be a second surface parameter corresponding to at least one sample non-light-output surface. The second training label may be obtained based on manual annotation, or may be automatically annotated based on historical data. The sample light output surface involved in the second training sample and the sample non-light-output surface involved in the second training label belong to the same sample crystal.
[0116] Some embodiments of the present disclosure, by first determining the first surface parameter of the light output surface, and then determining the second surface parameter of the non-light-output surface, may improve the accuracy of the first surface parameter and / or the second surface parameter obtained by calculation, thereby improving the precision of the machining process. Since the surface roughness of each surface of the crystal needs to be within an appropriate range, and excessively large or small surface roughness may influence the crystal, the better the accuracy of the first surface parameter and the second surface parameter, the better the performance of the machined crystal. Furthermore, by determining the first surface parameter and the second surface parameter through the first determination model and the second determination model, respectively, the self-learning capability of machine learning models may be utilized to find rules from a large amount of historical data, and to obtain a relationship between the initial optical output value (and / or the initial optical yield), the initial surface parameter, the target optical output value (and / or the target optical yield), and the first surface parameter, as well as a relationship between the first surface parameter, the target optical output value (and / or the target optical yield), and the second surface parameter, thereby improving the accuracy and the efficiency of determining the first surface parameter and the second surface parameter.
[0117] In 340, the crystal may be machined based on the first surface parameter and the second surface parameter. In some embodiments, operation 340 may be performed by the machining device 220 controlled by the machining module 140 or the control host 210.
[0118] In some embodiments, the machining module 140 or the control host 210 may control the machining device 220 to perform polishing processing on the light output surface, thereby allowing the surface parameter of the light output surface to reach the first surface parameter, and to perform grinding processing on the non-light-output surface, thereby allowing the surface parameter of the non-light-output surface to reach the second surface parameter.
[0119] In some embodiments, the machining module 140 or the control host 210 may determine a polishing parameter based on the first surface parameter, and machine the light output surface based on the polishing parameter. In some embodiments, the machining module 140 or the control host 210 may determine a grinding parameter based on the second surface parameter, and machine the non-light-output surface based on the grinding parameter.
[0120] More descriptions regarding machining the crystal based on the polishing parameter and / or the grinding parameter may be found elsewhere in the present disclosure (e.g., FIG. 4 and FIG. 5 and related descriptions thereof).
[0121] According to some embodiments of the present disclosure, machining the light output surface based on the first surface parameter and / or machining the non-light-output surface based on the second surface parameter may achieve consistency in the optical output of each crystal by changing the surface parameters of the crystal (e.g., surface roughness, or the like), and further, based on satisfying a crystal performance demand and a machining demand, may achieve fully automatic machining of the crystal, thereby reducing production cost.
[0122] FIG. 4 is a schematic diagram illustrating an exemplary process for crystal machining according to some embodiments of the present disclosure.
[0123] As shown in FIG. 4, in some embodiments, the machining module 140 or the control host 210 may determine a polishing parameter 420 based on the first surface parameter 410, and machine the light output surface 430 based on the polishing parameter 420; and / or determine a grinding parameter 450 based on the second surface parameter 440, and machine the non-light-output surface 460 based on the grinding parameter 450.
[0124] The polishing parameter refers to a machining parameter related to polishing processing the light output surface. In some embodiments, the polishing parameter may include a relative speed of rotation of an upper polishing plate and a lower polishing plate, a flow rate of polishing liquid, the polishing pressure, a number of rotations of the upper polishing plate and the lower polishing plate, a type of polishing grit, or the like. The polishing grit may include alumina, silica, ceria, silicon carbide, boron carbide, or diamond, or the like. The polishing liquid may be formed by mixing at least one polishing grit (e.g., alumina, silica, ceria, silicon carbide, boron carbide, or diamond, or the like) with a liquid (e.g., water, or the like). The particle size of the polishing liquid may be 0.001 to 5 μm.
[0125] In some embodiments, the machining module 140 or the control host 210 may determine the polishing parameter based on the first surface parameter, through a plurality of manners. For example, the machining module 140 or the control host 210 may determine the polishing parameter through historical machining data according to operations including determining a portion of the historical machining data having a historical first surface parameter similar to or the same as a current first surface parameter from the historical machining data, and taking the corresponding polishing parameter in the portion of historical machining data as the polishing parameter corresponding to the current first surface parameter. More descriptions regarding the first surface parameter may be found elsewhere in the present disclosure (e.g., operation 330 and related descriptions thereof).
[0126] In some embodiments, the machining module 140 or the control host 210 may determine the polishing parameter through a machining parameter model. The machining parameter model is a machine learning model (e.g., a CNN model, or the like).
[0127] In some embodiments, the machining parameter model may include a polishing layer. The input of the polishing layer includes the first surface parameter of at least one light output surface, and the output of the polishing layer may include the polishing parameter of the at least one light output surface. In some embodiments, the polishing layer of the machining parameter model may be trained by a plurality of third training samples each of which has a third training label. The training process of the polishing layer is similar to the training process of the first determination model. More descriptions may be found elsewhere in the present disclosure (e.g., FIG. 3 and related descriptions thereof). In some embodiments, each of the third training samples may at least include a sample first surface parameter of a sample light output surface, and the training label of each of the third training samples may be the corresponding polishing parameter of the sample light output surface. The third training label may be obtained based on manual annotation, or may be automatically annotated based on historical data.
[0128] According to some embodiments of the present disclosure, determining the polishing parameter through a machine learning model may utilize a self-learning capability of the machine learning model to find a pattern from a large amount of historical data, and obtain the relationship between the first surface parameter and the polishing parameter, thereby improving the accuracy and efficiency of determining the polishing parameter.
[0129] In some embodiments, before performing the polishing processing, the machining module 140 or the control host 210 may control the machining device 220 to protect the surface not to be machined (e.g., the non-light-output surface) (e.g., by adding a protective layer, or the like).
[0130] In some embodiments, the machining module 140 or the control host 210 may control the machining device 220 to place a polishing medium and add polishing liquid in a machining region, e.g., on a polishing plate (e.g., an upper polishing plate and / or a lower polishing plate), place the crystal on the lower polishing plate, and then apply different polishing pressures to the machining region to perform the polishing processing. In some embodiments, the polishing pressure may be divided into three levels: low, medium, and high. The machining module 140 or the control host 210 may sequentially increase the polishing pressure from low to high to perform the polishing processing. Under different polishing pressures, a specific number of polishing revolutions may be configured. More descriptions regarding the polishing pressure and the number of polishing revolutions may be found elsewhere in the present disclosure (e.g., related descriptions of the grinding pressure and the number of grinding revolutions in FIG. 2B).
[0131] In some embodiments, the machining module 140 or the control host 210 may control the machining device 220 to machine the light output surface based on the first polishing pressure, and obtain a first polishing feedback; in response to the first polishing feedback satisfying the second pressurization condition, machine the light output surface based on the second polishing pressure, and obtain a second polishing feedback; in response to the second polishing feedback satisfying the second pressurization condition, machine the light output surface based on the third polishing pressure. The first polishing pressure is less than the second polishing pressure, and the second polishing pressure is less than the third polishing pressure.
[0132] In some embodiments, the first polishing pressure may be a low-level pressure, the second polishing pressure may be a medium-level pressure, and the third polishing pressure may be a high-level pressure. When initiating the polishing processing, the low-level pressure is applied first, and then the polishing pressure is gradually increased.
[0133] The first polishing feedback is the feedback obtained after applying the first polishing pressure, and the second polishing feedback is the feedback obtained after applying the second polishing pressure. In some embodiments, the first polishing feedback or the second polishing feedback may include polishing efficiency. The polishing efficiency may be determined based on the polishing amount and the polishing time. For example, the polishing efficiency may be a ratio of the polishing amount to the polishing time. The polishing time may be determined by statistically calculating the time length from a start time point of polishing to an end time point corresponding to the polishing amount. When a rotation speed is constant, the polishing time may be determined by the number of polishing revolutions. The polishing amount may be obtained based on a laser rangefinder disposed above the polishing plate. In some embodiments, the first polishing feedback may be the optical output value (and / or the optical yield) of the light output surface after the polishing processing, which may be obtained by the measurement module or a detection device.
[0134] The second pressurization condition is a condition for determining whether to continue applying a greater polishing pressure. The second pressurization condition is related to a type of polishing feedback. For example, the second pressurization condition may be that the polishing efficiency does not reach a preset value or that a rate of change of the polishing efficiency reaches a preset condition. When the rate of change of the polishing efficiency increases, it indicates that the surface flatness of the crystal has undergone a large change, and pressure may be applied to perform the next step of polishing processing. In another example, the second pressurization condition may be that the optical output value of the light output surface does not reach the target optical output value, or that the difference between the optical output value and the target optical output value is greater than a difference threshold. In yet another example, the second pressurization condition may be that the optical yield of the light output surface does not reach the target optical yield, or that the difference between the optical yield and the target optical yield is greater than a difference threshold.
[0135] In some embodiments, in response to the first polishing feedback satisfying the second pressurization condition, the machining module 140 or the control host 210 may control the machining device 220 to machine the light output surface based on the second polishing pressure. In response to the first polishing feedback not satisfying the second pressurization condition, the machining module 140 or the control host 210 may control the machining device 220 to continue machining the crystal under the first polishing pressure, or to stop machining.
[0136] In some embodiments, in response to the second polishing feedback satisfying the second pressurization condition, the machining module 140 or the control host 210 may control the machining device 220 to machine the light output surface based on the third polishing pressure. In response to the second polishing feedback not satisfying the second pressurization condition, the machining module 140 or the control host 210 may control the machining device 220 to continue machining the crystal under the second polishing pressure, or to stop machining.
[0137] In some embodiments, the specific values of the first polishing pressure, the second polishing pressure, and the third polishing pressure may be determined by operating personnel or relevant industry personnel.
[0138] According to some embodiments of the present disclosure, performing the polishing processing process by setting three levels of different polishing pressures may avoid damage to the crystal caused by a small force bearing area or uneven force bearing during the polishing processing process.
[0139] The grinding parameter refers to a machining parameter related to grinding processing the non-light-output surface. In some embodiments, the grinding parameter may include a grinding manner, a type of grinding liquid, or a mesh size of abrasive grit. The grinding manner may include dry grinding and wet grinding. Dry grinding uses an abrasive grit to perform grinding directly. Wet grinding needs to mix the abrasive grit with a liquid to form a grinding liquid for grinding. The type of grinding liquid may be the same as a type of polishing liquid, and the difference lies in the particle size (also referred to as the mesh size) of the abrasive grit (which is referred to as a polishing grit in the polishing liquid). In some embodiments, the grinding parameter may further include a ratio of the mesh size of abrasive grit to water. In some embodiments, the grinding parameter may further include a relative speed of rotation of the upper grinding plate and the lower grinding plate, a grinding liquid flow rate, the grinding pressure, the number of grinding revolutions, or the like.
[0140] In some embodiments, the machining module 140 or the control host 210 may determine the grinding parameter based on the second surface parameter in a plurality of manners. For example, the machining module 140 or the control host 210 may determine the grinding parameter through historical machining data. The process for determining the grinding parameter is similar to the process for determining the polishing parameter. More descriptions regarding the process for determining the grinding parameter may be found in the preceding related descriptions. More descriptions regarding the second surface parameter may be found elsewhere in the present disclosure (e.g., operation 330 and related descriptions thereof), which will not be repeated herein.
[0141] In some embodiments, the machining module 140 or the control host 210 may determine the grinding parameter through the machining parameter model.
[0142] In some embodiments, the machining parameter model may further include a grinding layer. The grinding layer and the polishing layer are two separate layers in the machining parameter model. In some embodiments, an input of the grinding layer includes the second surface parameter of at least one non-light-output surface, and an output of the grinding layer may include the grinding parameter of the at least one non-light-output surface. In some embodiments, the grinding layer of the machining parameter model may be obtained through training with a plurality of fourth training samples each of which has a fourth training label. The training process of the grinding layer is similar to the training process of the first determination model. More descriptions may be found elsewhere in the present disclosure (e.g., FIG. 3 and related descriptions thereof). In some embodiments, a fourth training sample may at least include a sample second surface parameter of a sample non-light-output surface, and the training label of the fourth training sample may be a corresponding grinding parameter of the sample non-light-output surface. The fourth training label may be obtained based on manual annotation, or may also be automatically annotated based on historical data. In some embodiments, the polishing layer and the grinding layer of the machining parameter model may be trained separately.
[0143] According to some embodiments of the present disclosure, determining the grinding parameter through the machine learning model may utilize a self-learning capability of the machine learning model to find patterns from a large amount of historical data and obtain a relationship between the second surface parameter and the grinding parameter, thereby improving the accuracy and efficiency of determining the grinding parameter.
[0144] In some embodiments, before performing the grinding processing, the machining module 140 or the control host 210 may protect a surface that is not to be machined (e.g., the light output surface) (e.g., by adding a protective layer or the like).
[0145] In some embodiments, the machining module 140 or the control host 210 may place the abrasive grit or the grinding liquid on a machining region, for example, on the grinding plate (e.g., the upper grinding plate and / or the lower grinding plate), place the crystal on the lower grinding plate, and then apply different grinding pressures to the machining region to perform the grinding processing. In some embodiments, the grinding pressure may include three levels: low, medium, and high. The machining module 140 or the control host 210 may sequentially increase the grinding pressure from low to high to perform the grinding processing, and a specific number of grinding revolutions may be set under different grinding pressures. More descriptions regarding the grinding pressure, the number of grinding revolutions, and the grinding process may be found elsewhere in the present disclosure (e.g., FIG. 2B and related descriptions thereof).
[0146] In some embodiments, during the grinding processing, the machining module 140 or the control host 210 may automatically add the abrasive grit to the grinding liquid or the grinding region according to the determined grinding parameter. According to some embodiments of the present disclosure, automatically adding the grinding liquid or abrasive grit may realize fully automatic machining of the crystal, thereby improving the machining efficiency.
[0147] In some embodiments, the machining module 140 or the control host 210 may machine the non-light-output surface based on the first grinding pressure and obtain the first grinding feedback; in response to the first grinding feedback satisfying the first pressurization condition, machine the non-light-output surface based on the second grinding pressure and obtain the second grinding feedback; and in response to the second grinding feedback satisfying the first pressurization condition, machine the non-light-output surface based on the third grinding pressure. The first grinding pressure is less than the second grinding pressure, and the second grinding pressure is less than the third grinding pressure.
[0148] In some embodiments, the first grinding pressure may be a low pressure level, the second grinding pressure may be a medium pressure level, and the third grinding pressure may be a high pressure level. When initially performing the grinding processing, pressure is applied starting from the low pressure level, and then the grinding pressure is gradually increased.
[0149] The first grinding feedback is the feedback obtained after applying the first grinding pressure, and the second grinding feedback is the feedback obtained after applying the second grinding pressure. The first pressurization condition is a condition configured to determine whether to continue applying a greater grinding pressure. The first pressurization condition is related to the type of grinding feedback.
[0150] In some embodiments, each of the first grinding feedback and the second grinding feedback may include a torque of a motor providing drive for the grinding plate. Correspondingly, the first pressurization condition may be the torque of the motor being greater than a torque threshold range.
[0151] In some embodiments, each of the first grinding feedback and the second grinding feedback may include the grinding efficiency. The grinding efficiency is determined based on at least one of the grinding amount or the number of grinding revolutions, and the grinding amount is obtained by the laser rangefinder disposed above the grinding plate. For example, the grinding efficiency may be determined based on the grinding amount and the grinding time. For example, the grinding efficiency may be a ratio of the grinding amount to the grinding time. The grinding time may be determined by statistically determining a time length between a start time point of grinding and an end time point corresponding to the grinding amount. When a rotation speed is fixed, the grinding time may be determined by the number of grinding revolutions. In some embodiments, the laser rangefinder may monitor a displacement of the upper grinding plate, which represents a grinding thickness, thereby determining the grinding amount under the grinding pressure. In some embodiments, the number of grinding revolutions is the number of rotations of the grinding plate (including the upper grinding plate and the lower grinding plate), and may be obtained based on detection.
[0152] In some embodiments, in response to each of the first grinding feedback and the second grinding feedback including the grinding efficiency, the first pressurization condition may be that the grinding efficiency is less than a preset value.
[0153] According to some embodiments of the present disclosure, determining the polishing parameter and / or the grinding parameter based on the first surface parameter and the second surface parameter, respectively, and performing the machining process according to the determined polishing parameter and / or grinding parameter, may realize automatic identification of the crystal and automatic acquisition of the required polishing liquid, the grinding liquid, or abrasive grit, thereby realizing fully automatic machining of the crystal.
[0154] In some embodiments, in response to the first grinding feedback satisfying the first pressurization condition, the machining module 140 or the control host 210 may control the machining device 220 to machine the non-light-output surface based on the second grinding pressure and obtain the second grinding feedback. In response to the first grinding feedback not satisfying the first pressurization condition, the machining module 140 or the control host 210 may control the machining device 220 to continue machining the crystal under the first grinding pressure, or stop machining.
[0155] In some embodiments, in response to the second grinding feedback satisfying the first pressurization condition, the machining module 140 or the control host 210 may control the machining device 220 to machine the non-light-output surface based on the third grinding pressure. In response to the second grinding feedback not satisfying the first pressurization condition, the machining module 140 or the control host 210 may control the machining device 220 to continue machining the crystal under the second grinding pressure, or stop machining.
[0156] In some embodiments, the first grinding pressure, the second grinding pressure, and the third grinding pressure may be determined by staff or personnel in related industries.
[0157] According to some embodiments of the present disclosure, performing the grinding process by setting three different pressure levels of grinding pressure may avoid damage to the crystal caused by a small force-receiving area or uneven force during the grinding process, due to unevenness of a target surface to be machined.
[0158] FIG. 5 is a flowchart illustrating an exemplary process for determining machining quality according to some embodiments of the present disclosure. In some embodiments, process 500 may be performed by a crystal machining system 100 (e.g., the machining module 140) or a crystal machining system 200 (e.g., the control host 210). As shown in FIG. 5, process 500 comprises the following operations.
[0159] During machining, the machining module 140 or the control host 210 may further determine whether a defect exists and perform real-time corrections, thereby ensuring the machining quality of the crystal being machined.
[0160] In 510, a vibration response of a polishing plate and / or a grinding plate during machining may be obtained.
[0161] The vibration response refers to a vibration condition of the polishing plate (e.g., an upper polishing plate and / or a lower polishing plate) and / or the grinding plate (e.g., an upper grinding plate and / or a lower grinding plate). For example, the vibration response may include time points when the polishing plate and / or the grinding plate vibrate, vibration amplitudes, vibration frequencies, or the like. In some embodiments, the vibration response may be obtained by a vibration sensor and / or other devices disposed on the polishing plate and / or the grinding plate. In some embodiments, when machining one of the surfaces (e.g., grinding and / or polishing), the machining module 140 or the control host 210 may determine a corresponding vibration curve as the vibration response based on the vibration amplitudes detected at the plurality of time points.
[0162] In some embodiments, the machining module 140 or the control host 210 may obtain the vibration response corresponding to each of the different light output surfaces and / or the different non-light-output surfaces.
[0163] In 520, the machining quality of the light output surface and / or the non-light-output surface may be determined based on the vibration response.
[0164] The machining quality may reflect whether a defect exists in the machining process. In some embodiments, the machining module 140 or the control host 210 may determine the machining quality based on the vibration response in various manners. For example, the machining module 140 or the control host 210 may determine the machining quality based on the number of time points where the vibration amplitudes in the vibration response are not within a normal amplitude range. The more such time points, the poorer the machining quality.
[0165] In some embodiments, the machining module 140 or the control host 210 may extract a characteristic frequency band from the vibration response and identify the characteristic frequency band to determine whether a defect exists in the machined crystal.
[0166] The characteristic frequency band is a frequency band with an abnormal amplitude in the vibration response. Merely by way of example, the vibration of the polishing plate and / or the grinding plate is generally periodic, so the obtained vibration response should also exhibit periodic characteristics. Therefore, the characteristic frequency band may be a non-periodic frequency band in the vibration response. As another example, the characteristic frequency band may be a mutation frequency band in the vibration response (e.g., a frequency band formed by time points corresponding to a mutated vibration amplitude and surrounding time points). The relevant descriptions of the characteristic frequency band above are merely for illustrative purposes and are not intended to limit the scope of the present disclosure.
[0167] In some embodiments, the machining module 140 or the control host 210 may extract the characteristic frequency band by extracting non-periodic frequency bands and / or mutation frequency bands.
[0168] In some embodiments, the machining module 140 or the control host 210 may determine whether a defect exists in the machined crystal and the type of the existing defect by querying a defect lookup table based on the characteristic frequency band (e.g., using a waveform as a query basis). The defect lookup table may include a corresponding relationship between different characteristic frequency bands and different defect types. The defect lookup table may be determined based on historical data and / or prior knowledge. When the machining module 140 or the control host 210 is unable to query a corresponding defect type in the defect lookup table, it is determined that no defect exists in the machined crystal. When the machining module 140 or the control host 210 queries a corresponding defect type in the defect lookup table, it may be determined that a defect exists in the machined crystal, and the type of the existing defect may be determined.
[0169] In some embodiments, in response to no defect existing in the machined crystal, the machining module 140 or the control host 210 may determine that the machining quality is good. For example, the quality score of the machining quality may be determined to be a full score. In response to a defect existing in the machined crystal, the machining module 140 or the control host 210 may determine that the machining quality is poor. For example, the machining module 140 or the control host 210 may determine that the quality score of the machining quality has not reached a full score or is 0, or the like. In some embodiments, the machining module 140 or the control host 210 may further determine a corresponding quality score based on the defect type. Different defect types correspond to different quality scores. The corresponding relationship between the defect types and the quality scores may be determined by the system or preset manually.
[0170] In some embodiments, the machining module 140 or the control host 210 may further determine the machining quality of the machined surface according to the machined surface to which the vibration response belongs (e.g., the light output surface and / or the non-light-output surface being machined).
[0171] In some embodiments of the present disclosure, the characteristic frequency band may represent segments prone to problems during the machining process. By identifying the characteristic frequency band to determine the machining quality, the determination range of whether problems occur may be narrowed. Therefore, determining the machining quality by extracting the characteristic frequency band may effectively improve the efficiency and accuracy of the calculation process.
[0172] In some embodiments, the machining module 140 or the control host 210 may process the characteristic frequency band through an anomaly detection model to determine whether a defect exists in the machined crystal.
[0173] In some embodiments, the anomaly detection model is a machine learning model, e.g., a CNN model, or the like. The input of the anomaly detection model may include the characteristic frequency band, and the output of the anomaly detection model may include a defect type corresponding to the machined crystal and a corresponding quality score. When the output of the anomaly detection model is 0, it may be determined that no defect exists in the machined crystal. In some embodiments, the anomaly detection model may be trained by a plurality of fifth training samples each of which has a fifth training label. The training process of the anomaly detection model is similar to a training process of the first determination model. More descriptions regarding the training process of the anomaly detection model may be found elsewhere in the present disclosure (e.g., FIG. 3 and related descriptions thereof). In some embodiments, each of the fifth training samples may include a sample characteristic frequency band, and the fifth training label of the each of the fifth training samples may be a defect type corresponding to the sample characteristic frequency band and a corresponding quality score. The fifth training labels may be obtained based on manual annotation or automatically annotated based on historical data.
[0174] In some embodiments of the present disclosure, by identifying problems in the characteristic frequency band through an anomaly detection model, the self-learning ability of a machine learning model may be utilized to find patterns from a large amount of historical data and obtain a relationship between the characteristic frequency band and a defect category, thereby effectively improving the efficiency of the calculation process and the accuracy of recognition results.
[0175] In 530, in response to the machining quality not satisfying a preset condition, machining processing may be stopped or prompt information may be outputted.
[0176] The preset condition refers to a condition for determining the machining quality. In some embodiments, the preset condition may be one or more of the crystal having a defect or damage, the crystal having a certain type of defect, or the quality score of the crystal being lower than a set score threshold, or the like. The preset condition may be set based on actual needs.
[0177] The prompt information refers to information configured to indicate a machining status. In some embodiments, the machining module 140 or the control host 210 may display the image or text-based prompt information via a display device, or may also output the voice-based prompt information. In some embodiments, the prompt information may include a defect type that may appear in the machined crystal. In some embodiments, the prompt information may also include related information of the machining process (e.g., the first surface parameter, the second surface parameter, the number of grinding revolutions, the polishing pressure or the grinding pressure, a polishing liquid or a grinding liquid, or the like). The machining module 140 or the control host 210 may output the prompt information via various feasible manners (e.g., in forms of text, image, or voice, or the like), without limitation thereto.
[0178] In some embodiments, the machining module 140 or the control host 210 may obtain a vibration response and a reference vibration response of at least one of a polishing plate or a grinding plate during the machining process; and in response to a difference of at least one frequency response parameter between the vibration response and the reference vibration response not satisfying a preset response condition, stop the machining processing or output the prompt information.
[0179] The reference vibration response refers to a standard vibration curve corresponding to the vibration response. The machining module 140 or the control host 210 may determine the machining quality based on a comparison between the vibration response and the reference vibration response. If a difference of at least one frequency response parameter between the vibration response and the reference vibration response is within a parameter threshold range, it indicates that the vibration response is similar to the reference vibration response, and the machining processing may have no problem. In response to determining that the difference of the at least one frequency response parameter between the vibration response and the reference vibration response is outside the parameter threhold range, a problem may exist in the machining processing.
[0180] The reference vibration response may be determined based on historical data. In some embodiments, the reference vibration response may be obtained through the historical vibration curves obtained during a plurality of standard machining processes having similar grinding parameters. The standard machining process may be a machining process where a machined crystal has no defect. For example, fitting may be performed on the historical vibration curves obtained during a plurality of standard machining processes having similar grinding parameters, thereby obtaining the reference vibration response.
[0181] The frequency response parameter refers to characteristic data of the vibration response. The frequency response parameter may include an amplitude of the vibration response, frequency positions of peaks / valleys, a peak interval, or the like.
[0182] In some embodiments, the preset response condition may include a difference of at least one frequency response parameter being within the parameter threshold range. The parameter threshold range refers to a threshold range corresponding to a frequency response parameter difference, and may be determined through manual operations or historical data. If the difference of at least one frequency response parameter between the vibration response and the reference vibration response is within the parameter threshold range, it indicates that the vibration response is similar to the reference vibration response, and the machining processing may have no problem. If the difference of at least one frequency response parameter between the vibration response and the reference vibration response is not within the parameter threshold range, it indicates that the vibration response deviates from the reference vibration response, and the machining processing may have a problem.
[0183] In some embodiments of the present disclosure, the machining quality of the machined crystal may be quickly and effectively determined by a comparison between the vibration response and the reference vibration response.
[0184] It should be noted that the above descriptions regarding flow 300 and flow 500 are merely for illustration and description, and do not limit the scope of application of the present disclosure. For those skilled in the art, various modifications and changes may be made to flow 300 and flow 500 under the guidance of the present disclosure. However, such modifications and changes are still within the scope of the present disclosure.
[0185] The basic concepts have been described above. Apparently, for those skilled in the art, the above detailed disclosure is merely by way of example and does not constitute a limitation on the present disclosure. Although not expressly stated herein, those skilled in the art may make various modifications, improvements, and revisions to the present disclosure. Such modifications, improvements, and revisions are contemplated in the present disclosure, and thus are still within the spirit and scope of the exemplary embodiments of the present disclosure.
[0186] Concurrently, the present disclosure uses certain terms to describe the embodiments of the present disclosure. For example, terms such as ‘an embodiment,’‘an embodiment,’ and / or ‘some embodiments’ mean a certain feature, structure, or characteristic related to at least one embodiment of the present disclosure. Therefore, it should be emphasized and noted that ‘an embodiment,’‘an embodiment,’ or ‘an alternative embodiment’ mentioned twice or more times in different locations in the present disclosure do not necessarily refer to the same embodiment. Furthermore, certain features, structures, or characteristics in one or more embodiments of the present disclosure may be suitably combined.
[0187] Moreover, unless expressly stated in the claims, the order of processing elements and sequences, the use of numerals and letters, or the use of other names described in the present disclosure is not used to limit the order of the flows and manners of the present disclosure. While certain presently preferred embodiments have been discussed through various examples in the foregoing disclosure, it should be understood that such details are merely for illustration, and the appended claims are not limited to the disclosed embodiments. Rather, the claims are intended to cover all modifications and equivalent combinations that fall within the spirit and scope of the embodiments of the present disclosure. For example, while the system components described above may be implemented by hardware devices, they may also be implemented solely by software solutions, such as by installing the described system on an existing server or mobile device.
[0188] Similarly, it should be noted that to simplify the expressions disclosed in the present disclosure, and thereby facilitate an understanding of one or more embodiments of the present disclosure, descriptions of the embodiments of the present disclosure sometimes group various features into a single embodiment, a figure, or descriptions thereof. However, this manner of disclosure does not mean that the features required by the subject matter of the present disclosure are more numerous than the features recited in the claims. Indeed, the features of the embodiments are fewer than all the features of the single embodiment disclosed above.
[0189] Some embodiments use numerical ranges and parameters describing the number of components or properties. It should be understood that such numbers used in the descriptions of the embodiments are modified by modifiers such as ‘about,’‘approximate,’ or ‘substantially’ in some examples. Unless otherwise indicated, ‘about,’‘approximate,’ or ‘substantially’ indicate that the stated numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the present disclosure and the claims are all approximate values, which approximate values may vary according to the required characteristics of individual embodiments. In some embodiments, the numerical parameters should consider specified significant digits and employ a general rounding manner. While numerical ranges and parameters used to confirm the breadth of their scope are approximate in some embodiments of the present disclosure, in specific embodiments, such values are set to be as precise as possible within a feasible range.
[0190] Each patent, patent application, patent application publication, and other materials cited in the present disclosure, such as articles, books, specifications, publications, documents, or the like, are hereby incorporated by reference in its entirety into the present disclosure. Excluding prosecution history documents that are inconsistent with or conflict with the content of the present disclosure, and also excluding documents that limit the broadest scope of the claims of the present disclosure (currently or later appended to the present disclosure). It should be noted that if there is any inconsistency or conflict between the descriptions, definitions, and / or use of terms in subsidiary materials of the present disclosure and the content described in the present disclosure, the descriptions, definitions, and / or use of terms of the present disclosure shall prevail.
[0191] Finally, it should be understood that the embodiments described in the present disclosure are merely for illustrating the principles of the embodiments of the present disclosure. Other variations may also fall within the scope of the present disclosure. Thus, by way of example and not limitation, alternative configurations of the embodiments of the present disclosure may be regarded as consistent with the teachings of the present disclosure. Accordingly, the embodiments of the present disclosure are not limited to the embodiments expressly introduced and described in the present disclosure.
Claims
1. A method for crystal machining, comprising:obtaining at least one of a target optical yield or a target optical output value of a crystal;determining at least one light output surface and at least one non-light-output surface of the crystal;determining a first surface parameter of the least one light output surface and a second surface parameter of the least one non-light-output surface based on at least one of the target optical yield or the target optical output value;machining the crystal based on the first surface parameter and the second surface parameter.
2. The method according to claim 1, wherein the determining the least one light output surface and the least one non-light-output surface of the crystal comprises:obtaining at least one of an initial optical yield or an initial optical output value of each surface of the crystal;determining the least one light output surface and the least one non-light-output surface based on the initial optical yield of each surface and the target optical yield; ordetermining the least one light output surface and the least one non-light-output surface based on the initial optical output value of each surface and the target optical output value.
3. The method according to claim 1, wherein the determining the first surface parameter of the least one light output surface and the second surface parameter of the least one non-light-output surface based on at least one of the target optical yield or the target optical output value comprises:obtaining at least one of an initial optical yield or an initial optical output value of the least one light output surface, and an initial surface parameter of the least one light output surface;determining the first surface parameter based on the initial optical yield and the initial surface parameter of the least one light output surface and the target optical yield; ordetermining the first surface parameter based on the initial optical output value and the initial surface parameter of the least one light output surface, and the target optical output value.
4. The method according to claim 3, wherein the determining the first surface parameter of the at least one light output surface and the second surface parameter of the at least one non-light-output surface based on at least one of the target optical yield or the target optical output value comprises:determining the second surface parameter based on the first surface parameter and the target optical yield; ordetermining the second surface parameter based on the first surface parameter and the target optical output value.
5. The method according to claim 1, wherein the machining the crystal based on the first surface parameter and the second surface parameter comprises:determining a polishing parameter based on the first surface parameter;machining the at least one light output surface based on the polishing parameter;or,determining a grinding parameter based on the second surface parameter;machining the at least one non-light-output surface based on the grinding parameter.
6. The method according to claim 5, wherein the method comprises:determining at least one of the polishing parameter or the grinding parameter through a machine learning model, wherein the grinding parameter comprises a mesh size of abrasive grit.
7. The method according to claim 1, wherein the machining the crystal based on the first surface parameter and the second surface parameter comprises:obtaining a vibration response of at least one of a polishing plate or a grinding plate during machining; anddetermining a machining quality of at least one of the at least one light output surface or the at least one non-light-output surface based on the vibration response.
8. The method according to claim 7, wherein the determining the machining quality of at least one of the at least one light output surface or the at least one non-light-output surface based on the vibration response comprises:extracting a characteristic frequency band from the vibration response; andidentifying the characteristic frequency band to determine whether a machined crystal has a defect.
9. The method according to claim 8, wherein the identifying the characteristic frequency band to determine whether the machined crystal has the defect comprises:processing the characteristic frequency band through an anomaly detection model to determine whether the machined crystal has the defect, wherein the anomaly detection model is a machine learning model.
10. The method according to claim 7, wherein the machining the crystal based on the first surface parameter and the second surface parameter further comprises:in response to the machining quality not satisfying a preset condition, stopping machining or outputting prompt information.
11. The method according to claim 1, wherein the machining the crystal based on the first surface parameter and the second surface parameter comprises:obtaining a vibration response and a reference vibration response of at least one of a polishing plate or a grinding plate during machining;in response to a difference of at least one frequency response parameter between the vibration response and the reference vibration response not satisfying a preset response condition, stopping machining or outputting prompt information.
12. The method according to claim 1, wherein the machining the crystal based on the first surface parameter and the second surface parameter comprises:machining the at least one non-light-output surface based on a first grinding pressure, and obtaining a first grinding feedback;in response to the first grinding feedback satisfying a first pressurization condition, machining the at least one non-light-output surface based on a second grinding pressure, and obtaining a second grinding feedback; andin response to the second grinding feedback satisfying the first pressurization condition, machining the at least one non-light-output surface based on a third grinding pressure;wherein the first grinding pressure is less than the second grinding pressure, and the second grinding pressure is less than the third grinding pressure.
13. The method according to claim 12, wherein each of the first grinding feedback and the second grinding feedback comprises a torque of a motor providing drive for a grinding plate.
14. The method according to claim 12, wherein each of the first grinding feedback and the second grinding feedback comprise a grinding efficiency, the grinding efficiency is determined based on at least one of a grinding amount or a number of grinding revolutions, and the grinding amount is obtained by a laser rangefinder disposed above a grinding plate.
15. The method according to claim 1, wherein the machining the crystal based on the first surface parameter and the second surface parameter comprises:machining the at least one light output surface based on a first polishing pressure, and obtaining a first polishing feedback;in response to the first polishing feedback satisfying a second pressurization condition, machining the at least one light output surface based on a second polishing pressure, and obtaining a second polishing feedback;in response to the second polishing feedback satisfying the second pressurization condition, machining the at least one light output surface based on a third polishing pressure;wherein the first polishing pressure is less than the second polishing pressure, and the second polishing pressure is less than the third polishing pressure.
16. The method according to claim 1, wherein the method further comprises:preprocessing one or more initial blanks to obtain the crystal.
17. A system for crystal machining, comprising:a control host and a machining device, wherein the control host is configured to:obtain at least one of a target optical yield or a target optical output value of the crystal;determine at least one light output surface and at least one non-light-output surface of the crystal;determine a first surface parameter of the at least one light output surface and a second surface parameter of the at least one non-light-output surface based on at least one of the target optical yield or the target optical output value;control the machining device to machine the crystal based on the first surface parameter and the second surface parameter.
18. The system according to claim 17, wherein the machining device comprises: an upper grinding plate and a lower grinding plate, wherein the crystal is placed between the upper grinding plate and the lower grinding plate;the upper grinding plate is connected to a rotating shaft;at least one vibration sensor is arranged on an upper side surface of the upper grinding plate, and the at least one vibration sensor is configured to obtain a vibration response of the upper grinding plate.
19. The system according to claim 17, further comprising a roughness detection device, wherein the roughness detection device is configured to detect a surface roughness of the crystal.
20. The system according to claim 17, wherein the machining the crystal based on the first surface parameter and the second surface parameter comprises:determining a polishing parameter based on the first surface parameter;machining the at least one light output surface based on the polishing parameter;or,determining a grinding parameter based on the second surface parameter;machining the at least one non-light-output surface based on the grinding parameter.
