Cosmetic glass bottle surface polishing method and system

Abstract

The invention discloses a cosmetic glass bottle surface polishing method and system, and the method comprises the steps: carrying out gradient mesh abrasive composite modification on a sand leather base material, and carrying out three-dimensional cutting according to the surface curve contour of a glass bottle through a laser cutting technology to form a composite sand leather polishing piece; the composite sand leather grinding piece is assembled at the executing end of a six-axis linkage mechanical arm, partition differentiated grinding is carried out according to the surface pollutant distribution density image recognition result, and a glass bottle rough blank obtained after preliminary grinding is obtained; performing synchronous purging on the surface of the glass bottle rough blank by adopting inert gas flow with a micro-bubble generation device, and controlling the composite sand leather polishing piece to perform micro-vibration at the same time, so as to obtain a glass bottle semi-finished blank; and the glass bottle semi-finished blank is placed in a sealed cavity to be sprayed, final polishing is carried out in cooperation with pulse type ultrasonic vibration, and a cosmetic glass bottle finished product is obtained. According to the embodiment of the invention, high-precision grinding of the curved surface of the glass bottle can be realized, and the surface cleanliness and the gloss uniformity are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of cosmetics, and in particular to a method and system for polishing the surface of a cosmetic glass bottle. Background Art

[0002] Traditional surface polishing processes for cosmetic glass bottles are generally plagued by low efficiency and insufficient precision, especially when working with complex curved surfaces. Poor fit between the polishing tool and the bottle often leads to scratches and uneven gloss. Existing technologies often rely on manual polishing or fixed-grit grinding wheels, making it difficult to achieve dynamic pressure regulation and precise contaminant removal, and are prone to residual debris and secondary contamination. Furthermore, the separation of conventional cleaning steps from the polishing process increases process complexity. Summary of the Invention

[0003] The purpose of the present invention is to provide a surface polishing method and system for cosmetic glass bottles to address the deficiencies in the prior art, enabling high-precision adaptive polishing of the curved surface of glass bottles and improving surface cleanliness and gloss uniformity.

[0004] One embodiment of the present application provides a method for polishing the surface of a cosmetic glass bottle, the method comprising: The sand leather base material is modified with a gradient abrasive composite, and three-dimensionally cut according to the curved surface contour of the glass bottle through laser cutting technology to form a composite sand leather polished part that precisely fits the curved surface of the bottle body; The composite sand cortex polishing piece is assembled on the execution end of a six-axis linkage robot arm. The three-dimensional morphology data of the glass bottle surface is obtained through infrared scanning. The speed of the servo motor and the contact pressure of the polishing piece are dynamically adjusted based on the morphology data. Combined with the surface contaminant distribution density image recognition results, zone-by-zone differentiated polishing is implemented to obtain the rough glass bottle after preliminary polishing. On the surface of the rough glass bottle after preliminary grinding, an inert gas flow with a micro-bubble generator is used for synchronous blowing, while the composite sand cortex grinding piece is controlled to perform micro-vibration, and the residual grinding debris is removed through the gas-solid coupling effect to obtain a finely ground semi-finished glass bottle blank; The finely polished semi-finished glass bottle is placed in a sealed cavity, sprayed with a suspension of nano-scale silica flexible abrasive particles, and finally polished with pulsed ultrasonic vibration. The Brownian motion of the particles and the ultrasonic cavitation effect synergistically remove residual micron-level pollutants, resulting in a finished cosmetic glass bottle with a clean surface and a preset glossiness.

[0005] Optionally, the sand cortex base material is subjected to composite modification with gradient mesh abrasives, and three-dimensionally cut according to the curved surface contour of the glass bottle by laser cutting technology to form a composite sand cortex polished piece that precisely fits the curved surface of the bottle body, including: The natural sand cortex material is placed in a supercritical carbon dioxide fluid. By adjusting the pressure and temperature to control the fluid density, the nanoscale pore structure of the sand cortex fiber is regulated to form a porous carrier structure, providing anchor points for subsequent abrasive attachment. Using electrophoretic deposition technology, 800 mesh, 1500 mesh, and 3000 mesh zirconium oxide abrasives are sequentially deposited on the surface of the porous carrier structure. The abrasive deposition rate is controlled by pulse current density to form a gradient abrasive layer, forming a gradient abrasive composite substrate. Using a femtosecond laser micro-nano processing system, the gradient abrasive composite substrate is contour-tracked and cut based on the 3D model data of the cosmetic glass bottle. A micro-serrated joint is generated on the cutting edge to complete the preparation of the 3D pre-cut polished part. The three-dimensional pre-cut polished part is immersed in a silane coupling agent solution for surface activation, and then hot-pressed and composited with a shape memory polymer substrate. The shape memory effect of the substrate gives the polished part adaptive fitting ability to form a composite sand leather polished part.

[0006] Optionally, the composite sand cortex polishing piece is assembled on the execution end of a six-axis linkage robot arm, three-dimensional topography data of the glass bottle surface is obtained by infrared scanning, the rotation speed of the servo motor and the contact pressure of the polishing piece are dynamically adjusted based on the topography data, and zone-by-zone differentiated polishing is performed in combination with the surface contaminant distribution density image recognition result to obtain a rough glass bottle after preliminary polishing, including: An infrared thermal imager and a structured light scanner are integrated into the robotic arm's execution end to perform multimodal scanning of the glass bottle surface. Thermal wave imaging technology is used to identify sub-surface defects, and the scan data is fused to generate a 3D point cloud dataset containing morphology and defect information. Based on a 3D point cloud dataset, a dynamic time warping algorithm is used to match a standard bottle model, calculate the local curvature deviation map, and combine it with a deep learning semantic segmentation model to identify the type of surface contaminants and their distribution density, thereby constructing a differentiated polishing strategy map. Taking the curvature deviation map and contaminant distribution density as input, the servo motor speed and contact pressure parameters are deduced in real time through the fuzzy control rule library to form a dynamic control parameter set; According to the dynamic control parameter set, the six-axis robotic arm is driven to move along the preset trajectory, and the partitioned grinding operation is performed synchronously. The dynamic error of the robotic arm is corrected through the iterative learning control algorithm to complete the preliminary grinding to obtain the rough glass bottle blank.

[0007] Optionally, the surface of the rough glass bottle blank after preliminary polishing is synchronously purged with an inert gas flow with a micro-bubble generating device, and the composite sand cortex polishing piece is controlled to perform micro-vibration, and residual polishing debris is removed through gas-solid coupling to obtain a finely polished semi-finished glass bottle blank, including: Argon is introduced into the venturi tube and mixed with deionized water. Microbubbles of a specific diameter are generated at the gas-liquid interface through the ultrasonic cavitation effect, forming a gas-liquid two-phase flow, which serves as a medium for debris removal. The distribution of debris on the surface of the glass bottle is monitored in real time by a high-speed camera, and the trajectory of the debris is calculated using particle image velocimetry technology. The injection angle and flow rate of the gas-liquid two-phase flow are dynamically adjusted based on the trajectory data to form precise purge parameters. Based on the purge parameters, the composite sand cortex polishing piece is controlled to perform elliptical micro-vibration at a specific micro-vibration frequency. The amplitude and phase difference are optimized through vibration modal analysis to enhance the debris resonance separation effect. Ultraviolet radiation is introduced during the purging process to decompose organic residues through photocatalytic oxidation, and an electrostatic adsorption device is used to collect suspended particles to remove residual debris and obtain semi-finished glass bottles.

[0008] Optionally, the finely polished semi-finished glass bottle blank is placed in a sealed cavity, sprayed with a suspension of nano-scale silica flexible abrasive particles, and finally polished in combination with pulsed ultrasonic vibration, thereby removing residual micron-sized contaminants through the synergistic effect of particle Brownian motion and ultrasonic cavitation effect to obtain a finished cosmetic glass bottle with a clean surface and a preset glossiness, including: Silica particles of a specific particle size are dispersed in deionized water, an appropriate amount of polyvinyl pyrrolidone is added as a dispersant, and a stable suspension is prepared by ultrasonic homogenization to serve as the grinding medium for the final polishing; A pressure atomizing nozzle is used to spray the nano-abrasive medium onto the surface of the glass bottle in a fan-shaped spray. The spray particle size distribution is monitored in real time by a laser particle size analyzer, and the spray pressure is dynamically adjusted to form a uniform liquid film. Applying pulsed ultrasonic vibrations at a specific pulse frequency to a glass bottle covered with a uniform liquid film, the generation and collapse cycles of cavitation bubbles are controlled by duty cycle modulation. Fluid dynamics simulation is combined to optimize the transducer position and enhance the selective polishing effect. The polishing effect is monitored in real time through an online gloss meter, and the ultrasonic power and nanoparticle concentration are dynamically adjusted based on the monitoring data. When the glossiness reaches the preset standard, the cleaning program is triggered, and the grinding medium is recovered through the circulating filtration system to obtain a finished cosmetic glass bottle with a clean surface and a preset glossiness.

[0009] Another embodiment of the present application provides a cosmetic glass bottle surface polishing system, the system comprising: The cutting module is used to modify the sand leather base material with a gradient abrasive composite, and to perform three-dimensional cutting according to the curved surface contour of the glass bottle through laser cutting technology to form a composite sand leather polished part that precisely fits the curved surface of the bottle body; The polishing module is used to assemble the composite sand cortex polishing piece on the execution end of the six-axis linkage robot arm, obtain the three-dimensional topography data of the glass bottle surface through infrared scanning, dynamically adjust the speed of the servo motor and the contact pressure of the polishing piece based on the topography data, and implement zone-by-zone differentiated polishing based on the image recognition results of the surface contaminant distribution density to obtain the rough glass bottle after preliminary polishing; The purge module is used to synchronously purge the surface of the rough glass bottle blank after preliminary polishing with an inert gas flow equipped with a micro-bubble generator, while controlling the composite sand cortex polishing piece to perform micro-vibration, thereby removing the residual polishing debris through gas-solid coupling to obtain a finely polished semi-finished glass bottle blank; The spray module is used to place the finely polished semi-finished glass bottle in a sealed cavity, spray it with a suspension of nano-scale silica flexible abrasive particles, and perform final polishing in conjunction with pulsed ultrasonic vibration. The Brownian motion of the particles and the ultrasonic cavitation effect synergistically remove residual micron-level pollutants, resulting in a finished cosmetic glass bottle with a clean surface and a preset glossiness.

[0010] Yet another embodiment of the present application provides a storage medium, wherein the storage medium stores a computer program, wherein the computer program is configured to execute any of the above methods when run.

[0011] Yet another embodiment of the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute any of the above methods.

[0012] Compared with the prior art, the present invention provides a method for polishing the surface of a cosmetic glass bottle, which performs composite modification of a sand cortex base material with a gradient mesh abrasive, and uses laser cutting technology to perform three-dimensional cutting according to the curved surface contour of the glass bottle to form a composite sand cortex polished part; the composite sand cortex polished part is assembled on the execution end of a six-axis linkage robotic arm, and zoned differentiated polishing is implemented in combination with the surface pollutant distribution density image recognition result to obtain a glass bottle rough blank after preliminary polishing; on the surface of the glass bottle rough blank, an inert gas flow with a microbubble generating device is used for synchronous blowing, and at the same time, the composite sand cortex polishing part is controlled to perform micro-vibration to obtain a semi-finished glass bottle blank; the semi-finished glass bottle blank is placed in a sealed cavity for spraying, and is finally polished in combination with pulsed ultrasonic vibration to obtain a finished cosmetic glass bottle, thereby achieving high-precision adaptive polishing of the curved surface of the glass bottle and improving the surface cleanliness and gloss uniformity. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A hardware structure block diagram of a computer terminal for a cosmetic glass bottle surface polishing method provided by an embodiment of the present invention; Figure 2A schematic flow chart of a method for polishing the surface of a cosmetic glass bottle provided by an embodiment of the present invention; Figure 3 This is a schematic structural diagram of a cosmetic glass bottle surface polishing system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0014] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limiting the present invention.

[0015] The embodiment of the present invention first provides a method for polishing the surface of a cosmetic glass bottle. The method can be applied to electronic devices, such as computer terminals, specifically ordinary computers, etc.

[0016] The following describes it in detail by taking running on a computer terminal as an example. Figure 1 The hardware structure block diagram of a computer terminal for a cosmetic glass bottle surface polishing method provided by an embodiment of the present invention. Figure 1 As shown, the computer device includes a processor, a memory, and a network interface connected via a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0017] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, which, when executed, can cause the processor to execute any one of the cosmetic glass bottle surface polishing methods.

[0018] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0019] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any method for polishing the surface of a cosmetic glass bottle.

[0020] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 1 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0021] It should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0022] See also Figure 2 The embodiment of the present invention provides a method for polishing the surface of a cosmetic glass bottle, which may include the following steps: S201: The sand leather base material is modified with a gradient abrasive composite. Laser cutting technology is used to perform three-dimensional cutting according to the curved surface contour of the glass bottle to form a composite sand leather polished part that precisely fits the curved surface of the bottle. Specifically, the natural sand cortex material can be placed in a supercritical carbon dioxide fluid, and the fluid density can be controlled by adjusting the pressure and temperature to achieve the regulation of the nano-scale pore structure of the sand cortex fiber, forming a porous carrier structure, which provides an anchoring point for the subsequent abrasive attachment; Construction of supercritical fluid processing system Natural chamois leather material (such as tanned leather substrate) is first cut into standard-sized billets and loaded into an autoclave equipped with a precision temperature control system (±0.5°C accuracy) and a pressure sensor (range 0-50 MPa). Liquid carbon dioxide is injected into the reactor, and a heater raises the temperature to above the critical temperature of 31.1°C (typical operating temperature is 45°C). Simultaneously, a hydraulic pump raises the pressure to above the critical pressure of 7.38 MPa (typical operating pressure is 15 MPa). In this state, the carbon dioxide enters a supercritical state (SC-ST), with a density that can be continuously adjusted between 300 and 900 kg / m³. By adjusting the pressure and temperature in real time (e.g., increasing the pressure in steps from 12 MPa to 18 MPa and the temperature gradually from 40°C to 50°C), the supercritical carbon dioxide fluid density is controlled at a target value (e.g., 600 kg / m³), which directly affects the fluid's ability to swell and penetrate chamois leather fibers.

[0023] Nanopore regulation mechanism Under the osmotic effect of supercritical carbon dioxide, the collagen molecular chains in the sand cortical fibers undergo directional relaxation. The high diffusivity of the fluid (diffusion coefficient of about 10 -7 The fluid, at a pressure of 15 MPa (1500 sq m / s), penetrates the fiber microstructure, preferentially dissolving and removing lipid impurities and low-molecular-weight non-collagenous proteins. By maintaining high pressure (e.g., 15 MPa) for 30 minutes, the natural pores within the fiber expand under the action of fluid tension, forming uniformly distributed nanoscale channels (NCs) with pore sizes ranging from 50 to 200 nm. The channel density and size are precisely controlled by the fluid density: at a density of 600 kg / m³, dense channels with an average pore size of 80 nm are generated; at a density of 800 kg / m³, the pore size expands to 150 nm, but the density decreases. This process simultaneously degreases and purifies the material, ultimately forming a porous support structure with an open three-dimensional network (porosity > 65%).

[0024] Enhanced anchor point function After exiting the autoclave, the porous support structure undergoes a gradient depressurization process (at a rate of 0.5 MPa / min) to prevent pore collapse. After verifying the pore morphology using scanning electron microscopy (SEM), the structure is immersed in a 0.5 mol / L sodium hydroxide solution for surface hydroxylation, generating a large number of active hydroxyl groups (-OH) on the pore walls. These hydroxyl groups act as anchoring sites (AS), significantly enhancing the chemical bonding strength of the subsequent abrasive particles. The pore depth is controlled by the treatment time: a 30-minute treatment yields an active pore layer approximately 5 microns deep, while a 60-minute treatment extends this to 12 microns, ensuring mechanical locking of the abrasive particles once embedded.

[0025] Using electrophoretic deposition technology, 800 mesh, 1500 mesh, and 3000 mesh zirconium oxide abrasives are sequentially deposited on the surface of the porous carrier structure. The abrasive deposition rate is controlled by pulse current density to form a gradient abrasive layer, forming a gradient abrasive composite substrate. Electrophoretic deposition system configuration The porous support structure is fixed to the cathode electrode plate and placed in an electrophoresis tank. The tank is filled with a suspension consisting of zirconium oxide abrasive particles (particle size: 800 mesh approximately 21 microns, 1500 mesh approximately 10 microns, 3000 mesh approximately 5 microns), isopropyl alcohol (dispersion medium), and cerium nitrate (conductive additive). The suspension solids content is controlled at 20% by weight, and the pH is adjusted to 8.5 with ammonia water to ensure a positive surface charge (zeta potential +35 mV) on the abrasive particles. The deposition process is carried out in three stages, with the suspension being replaced with the corresponding mesh size at each stage.

[0026] Pulsed current deposition control In the first stage, 800-mesh abrasive particles were deposited by applying a pulsed current (PC) with parameters set at a peak current density of 15 mA / cm², a pulse width of 50 milliseconds, and an intermission time of 100 milliseconds. The high current density facilitated the rapid migration of large-size abrasive particles, forming a bottom layer of coarse abrasive approximately 30 microns thick within 50 seconds. In the second stage, a 1500-mesh suspension was switched to the 1500-mesh suspension. The pulse parameters were adjusted to a peak current density of 8 mA / cm², a pulse width of 80 milliseconds, and an intermission time of 50 milliseconds. At this medium current density, the abrasive particles stacked in an orderly manner, forming a 20-micron-thick intermediate transition layer. In the third stage, a 3000-mesh suspension was used. A low peak current density of 5 mA / cm², a pulse width of 100 milliseconds, and an intermission time of 30 milliseconds were used to achieve dense packing of fine abrasive particles by extending the pulse time, resulting in a fine surface layer 10 microns thick.

[0027] Gradient structure solidification After each deposition stage, the process was paused to remove any unbound abrasive particles using a microvibration table (50 micron amplitude, 100 Hz frequency). After the three-layer deposition was complete, the entire structure was immersed in a silica sol binder and heat-treated at 120°C for one hour. The silica sol dehydrated to form a siloxane network, which covalently bonded the abrasive particles to the carrier fibers, ultimately forming a graded abrasive composite substrate (GACS). The abrasive particle size of the surface layer to the bottom layer continuously transitioned from 5 microns to 10 microns to 21 microns, and the interlayer bond strength was tested to reach 15 MPa.

[0028] Using a femtosecond laser micro-nano processing system, the gradient abrasive composite substrate is contour-tracked and cut based on the 3D model data of the cosmetic glass bottle. A micro-serrated joint is generated on the cutting edge to complete the preparation of the 3D pre-cut polished part. Laser processing system calibration The CAD surface model of a cosmetic glass bottle was imported into a femtosecond laser processing system (FLS), equipped with a 1030 nm laser source (pulse width 350 fs, repetition rate 100 kHz). Coordinate calibration was first performed: a fluorescent positioning layer was sprayed onto the surface of a gradient abrasive composite substrate. An image of the substrate was captured using a CCD vision system and spatially aligned with the glass bottle model (positioning accuracy ±5 μm). The processing area was divided according to the surface curvature distribution. High-curvature areas (radius <20 mm) were treated with a fine path spacing of 0.05 mm, while low-curvature areas (radius >50 mm) had a path spacing of 0.2 mm.

[0029] Contour tracing and micro-serration processing A dynamic focusing module maintains the laser focus at a constant position of 10 microns below the substrate surface. A contour tracking algorithm (CTA) adjusts the laser scanning speed in real time as the cut follows the contours of the 3D model: 800 mm / s for straight segments and 400 mm / s for circular segments. Edge modification is performed simultaneously with the cut: a 1 MHz oscillating scan frequency is superimposed on both sides of the cutting path, increasing the laser energy density to 8 joules per square centimeter. This selectively ablates the material, forming a micro-serrated joint (MSJ) with a depth of approximately 50 microns and a pitch of 20 microns. This structure increases the contact area during subsequent lamination.

[0030] 3D pre-cut parts forming After cutting, the substrate is segmented into multiple curved units that perfectly match the bottle's curvature (e.g., shoulder, belly, and base). The cut pieces are separated using a negative pressure transfer device, and white light interferometry is used to inspect the edge serrations (tooth height tolerance ±2 microns) and contour conformity (deviation <25 microns). The resulting 3D pre-cut abrading component (3D-PAC) retains the gradient abrasive structure, with a maximum curvature deviation of no more than 0.5% from the target glass bottle.

[0031] The three-dimensional pre-cut polished part is immersed in a silane coupling agent solution for surface activation, and then hot-pressed and composited with a shape memory polymer substrate. The shape memory effect of the substrate gives the polished part adaptive fitting ability to form a composite sand leather polished part.

[0032] Surface activation treatment A 3% (volume) solution of KH-550 silane coupling agent (aminopropyltriethoxysilane) in ethanol was prepared. Three-dimensional pre-cut abrasive parts were immersed in the solution and treated with ultrasound (300 watts, 40 kHz) for 20 minutes to allow the coupling agent molecules to penetrate the gaps between the abrasive particles. After removal, the parts were pre-cured at 80°C for 15 minutes to form a chemical activation layer. This activation layer contains two functional groups: siloxane (-Si-O-) groups that bond to the abrasive particles / carrier, and amino (-NH2) groups that face outward and provide reactive sites.

[0033] Shape memory polymer composites A 0.5 mm thick polyurethane-based shape memory polymer substrate (SMPS) with a glass transition temperature (Tg) of 55°C was cut. The activated polished part was placed on the SMPS and fed into a hot press. In the first stage, the temperature was raised to 70°C (15°C above the Tg), and a pressure of 0.3 MPa was applied for 5 minutes to soften the SMPS and embed it into the micro-serrated structure of the polished part. In the second stage, the temperature was lowered to 40°C (15°C below the Tg), and pressure was applied to 0.8 MPa for 10 minutes to achieve mechanical interlocking. At this point, the SMPS entered its temporary shape state (TSS).

[0034] Adaptive fit capability The composite part undergoes shape programming: it is reheated to 70°C, deformed into a flat state under the constraints of an external mold, and then cooled to a fixed state, forming a preset strain (PS). When bonded to a glass bottle at 40°C, the SMPS partially recovers due to its temperature approaching Tg, generating a progressive bonding pressure of approximately 0.5 Newtons per square centimeter. This shape memory effect (SME) enables the abrasive tool to automatically adapt to the changing curvature of the bottle (with a curvature radius ranging from 15-200 mm), ultimately creating a composite leather abrasive tool (CLAT) with both gradient grinding capabilities and surface adaptability.

[0035] S202: Assemble the composite sand cortex polishing piece on the execution end of the six-axis linkage robot arm, obtain three-dimensional surface morphology data of the glass bottle through infrared scanning, dynamically adjust the speed of the servo motor and the contact pressure of the polishing piece based on the morphology data, and implement zone-by-zone differentiated polishing in combination with the surface contaminant distribution density image recognition results to obtain a rough glass bottle after preliminary polishing; Specifically, an infrared thermal imager and a structured light scanner can be integrated into the robotic arm's execution end to perform multimodal scanning of the glass bottle surface. Thermal wave imaging technology can be used to identify sub-surface defects, and the scan data can be fused to generate a 3D point cloud dataset containing morphology and defect information. The end effector of a six-axis robotic arm is simultaneously equipped with a composite sand-cored polishing tool, an infrared thermal imager (ITI), and a structured light scanner (SLS) through a customized fixture. The ITI uses a mid-wave infrared detector (wavelength range 3-5 microns) with a HgCdTe focal plane array (FPA) at its core. Its pixel size is 15 microns, and its thermal sensitivity reaches 0.03 degrees Celsius. The SLS uses a blue LED light source (wavelength 450 nanometers) to project a combined Gray code and phase-shifted fringe pattern, paired with a 5-megapixel industrial camera to capture the deformation fringes. When scanning is initiated, the robotic arm orbits the glass bottle along a pre-set path, and the ITI captures the surface temperature distribution at a rate of 30 frames per second. Subsurface cracks or bubbles in the glass bottle exhibit different heat conduction characteristics than normal areas, manifesting as localized temperature differences in the ITI image (e.g., a 0.5-degree Celsius drop in temperature at the crack). The SLS simultaneously projects the grating pattern and captures the deformation fringes, reconstructing the surface geometry with millimeter-level accuracy using a phase unwrapping algorithm.

[0036] Thermal Wave Imaging (TWI) relies on pulsed thermal stimulation: Before scanning, the robotic arm controls a halogen lamp array (1000 watts) integrated into the fixture to apply a short 0.5-second heat pulse to the surface of the glass bottle. The ITI then records the temperature decay curve (sampling interval 0.1 seconds) during the surface cooling process. Subsurface defects can alter the thermal diffusion rate. For example, a microcrack 0.2 mm in depth can cause a 15% increase in the cooling rate in the corresponding area. By extracting the second-order derivative features of the temperature decay curve and combining them with a pre-trained defect recognition model (based on a convolutional neural network (CNN)), the thermal image is converted into a subsurface defect probability map (DPM). Simultaneously, the SLS 3D point cloud data (with an accuracy of ±0.05 mm) is spatially registered with the DPM. Using the robotic arm's base coordinate system as a reference, the point clouds are fused using an iterative closest point (ICP) algorithm, ultimately generating an enhanced 3D point cloud dataset containing 500,000 to 1 million data points. Each data point contains spatial coordinates (X, Y, Z), normal vector (Nx, Ny, Nz), local curvature (C) and defect mark (D value 0-1 indicates defect probability).

[0037] The data fusion process utilizes a multi-resolution pyramid architecture: the original SLS point cloud is first downsampled to 1mm resolution using a voxel grid as the base layer. The ITI defect probability map is then upsampled to the same resolution and mapped onto the base layer point cloud. The high-resolution (0.1mm) SLS detail layer is then overlaid with the defect information layer using a Laplacian pyramid fusion algorithm, preserving geometric details while embedding defect attributes. The fused point cloud dataset is stored in an octree structure, supporting real-time spatial queries. For example, if a point cloud normal vector in a certain area suddenly deviates by more than 15 degrees and the D value is greater than 0.8, the system automatically labels it as a "high-probability subsurface defect area." This designation directly influences the subsequent polishing strategy. The entire process is completed within a single orbital motion of the robotic arm (approximately 12 seconds), providing the data foundation for real-time control.

[0038] Based on a 3D point cloud dataset, a dynamic time warping algorithm is used to match a standard bottle model, calculate the local curvature deviation map, and combine it with a deep learning semantic segmentation model to identify the type of surface contaminants and their distribution density, thereby constructing a differentiated polishing strategy map. The Standard Bottle Model (SBM) is stored in the database as a NURBS surface with a control point density of 5 points per square centimeter. The matching process begins with an initial alignment of the enhanced point cloud dataset with the SBM: seven feature points, such as the bottle mouth and base, are selected and the initial transformation matrix is ​​calculated using least-squares fitting. Dynamic Time Warping (DTW) is then used to address non-rigid deformations: the point cloud is divided into 120 generatrixes along the meridian direction, each containing 200 points. DTW is applied to each generatrix, constructing a cumulative cost matrix (CostMatrix Size 200×200) to find the optimal nonlinear matching path between the point cloud and the corresponding SBM generatrix. This algorithm allows for local time axis scaling, effectively compensating for glass bottle manufacturing tolerances (e.g., ±0.3 mm diameter deviation). After matching, the position deviation vector (ΔX, ΔY, ΔZ) of each point is output, and then the local curvature deviation (CD) is calculated: 3×3 neighborhood points are extracted from the point cloud, and the principal curvatures K1 and K2 are obtained by least squares surface fitting. The deviation value δK (unit: meter) is compared with the theoretical curvature of SBM. -1 ).

[0039] Surface contaminant identification utilizes an end-to-end deep learning framework. First, a ResNet-50 backbone network is initialized through transfer learning, using as input RGB texture maps (resolution 2048×2048) collected by the SLS. The training dataset contains five typical contaminant classes: oil stains (labeled Class 1), fingerprints (Class 2), glue residue (Class 3), dust (Class 4), and water stains (Class 5). The network output layer is connected to a Conditional Random Field (CRF) to optimize spatial continuity, ultimately generating a pixel-level semantic segmentation map (SSM). Contaminant distribution density (CDD) calculation is based on sliding window statistics: the surface is divided into 2 mm × 2 mm grids, and the pixel percentage of each contaminant type within each grid is counted. For example, if the proportion of Class 2 pixels in a grid exceeds 60%, it is labeled as a "high-density fingerprint contamination area."

[0040] The construction of the Differentiated Polishing Strategy Map (DPSM) is a three-dimensional spatial decision-making process: using point cloud data as a carrier, each point is assigned a four-dimensional decision vector (R, P, T, M). R value (speed coefficient): set based on curvature deviation δK, when |δK|>0.5m -1 When (high curvature change area), R=1.2 (corresponding to the servo motor reference speed of 8000 rpm×1.2), and in the flat area, R=0.8; P value (contact pressure): dynamically adjusted according to the type of pollutant. Oil stains (Class 1) require high pressure (P = 1.5 Newtons), and dust (Class 4) requires low pressure (P = 0.7 Newtons). T value (polishing time): determined by CDD, density> 40% area T = 2 times the benchmark time; M value (movement mode): Subsurface defect areas (D>0.8) trigger avoidance mode (M=0).

[0041] The map is spatially continuous through three-dimensional Kriging Interpolation, and ultimately generates a digital strategy grid with a resolution of 0.5 mm, which serves as the direct basis for the execution of the robotic arm.

[0042] Taking the curvature deviation map and contaminant distribution density as input, the servo motor speed and contact pressure parameters are deduced in real time through the fuzzy control rule library to form a dynamic control parameter set; The fuzzy rule base (FRB) uses a Mamdani-type inference architecture and includes two input variables (curvature deviation δK and pollutant density CDD) and two output variables (speed correction coefficient ΔR and pressure correction coefficient ΔP). Input variable fuzzification settings: δK (unit: m -1 ): divided into 5 fuzzy sets {negative big (NB), negative small (NS), zero (ZE), positive small (PS), positive big (PB)}, domain [-1.0, 1.0], triangular membership function; CDD (unit: percentage): divided into three fuzzy sets {low (L), medium (M), high (H)}, domain [0,100%], trapezoidal membership function; The domain of the output variables ΔR and ΔP is [-0.5, 0.5], which is divided into 7 fuzzy sets, and the membership function is Gaussian.

[0043] The rule base contains 15 expert experience rules, such as: Rule 7: If δK is PB (positive large deviation) and CDD is L (low pollution) → then ΔR = PS (positive small speed increase), ΔP = NS (negative small pressure reduction); Rule 12: If δK is ZE (zero deflection) and CDD is H (high pollution) → then ΔR = ZE (zero speed regulation), ΔP = PB (positive pressure increase); During real-time inference, parallel computation is performed on 2000 grid points per second. -1 , CDD = 75%) as an example: δK activates PS (membership 0.7) and PB (0.3), while CDD activates H (1.0). This triggers Rule 10 (PS & H → ΔR = ZE, ΔP = PM) and Rule 11 (PB & H → ΔR = NS, ΔP = PB). Through weighted average defuzzification, the final outputs are ΔR = -0.12 and ΔP = 0.41. Combined with the baseline parameters (R0 = 8000 rpm, P0 = 1.0 N), the real-time commands are: speed 7040 rpm, pressure 1.41 N.

[0044] The Dynamic Control Parameter Set (DCPS) is stored as a four-dimensional space-time matrix (X, Y, Z, Time). The spatial dimension is aligned with the strategy map grid, and the temporal dimension is sampled at a 10 Hz frequency. Parameter transmission utilizes the real-time Ethernet protocol (EtherCAT) with a 1-millisecond cycle. A key innovation lies in the introduction of an environmental compensation module: a temperature sensor monitors the workshop ambient temperature (T_env) and automatically applies compensation rules when T_env exceeds 30°C (e.g., if the temperature rise causes the abrasive to soften, a 10% pressure increase is required). After each bottle polishing operation, the system automatically records the parameter optimization results and updates the fuzzy rule weights through reinforcement learning. For example, if a high-pressure strategy for Class 3 contamination results in a scratch, the corresponding rule confidence is reduced by 15%.

[0045] According to the dynamic control parameter set, the six-axis robotic arm is driven to move along the preset trajectory, and the partitioned grinding operation is performed synchronously. The dynamic error of the robotic arm is corrected through the iterative learning control algorithm to complete the preliminary grinding to obtain the rough glass bottle blank.

[0046] The six-axis robotic arm (5 kg payload, ±0.02 mm repeatability) utilizes B-spline interpolation for trajectory planning. The standard trajectory library includes 300 basic paths for 12 bottle types, such as "shoulder spiral (pitch = 2 mm)" and "bottle meridian reciprocating (amplitude ±15 degrees)." During execution, the spatial coordinates of the Dynamic Control Parameter Set (DCPS) are matched to the trajectory points in real time. When the end of the robotic arm reaches a specific spatial grid (e.g., coordinates X120, Y50, Z200), the (R, P, T, M) parameters at that location are extracted from the DCPS and converted into execution instructions. Contact pressure control is implemented in a closed-loop manner using a six-axis force sensor (range ±200 N, accuracy 0.1 N). If the measured pressure deviates from the target value by more than 0.3 N, an impedance controller adjusts the robotic arm's Z-axis offset within 50 milliseconds (adjustment range ±0.5 mm).

[0047] The core of zoned differentiated polishing is the allocation of time and space resources: the bottle surface is divided into 20 working areas (such as the bottom ring area, the main area of ​​the bottle body, and the bottleneck transition area). High curvature area (|δK|>0.4 meters -1A "high-frequency, small-step" strategy (line speed 40 mm / s, step size 0.8 mm) is used for polishing in flat areas, while a "low-frequency, large-step" strategy (line speed 80 mm / s, step size 2 mm) is used for polishing in flat areas. For highly contaminated areas (CDD > 60%), a three-stage polishing process is implemented: the first stage is high-pressure rough grinding (P = 1.5 N), the second stage is medium-pressure fine grinding (P = 1.0 N), and the third stage is low-pressure polishing (P = 0.5 N). The duration of each stage is determined by the T value. The robot arm avoids sub-surface defect areas (M = 0) in real time during execution: when a defect mark is detected, an avoidance path with a radius of 5 mm is automatically generated, simultaneously triggering compensatory polishing in adjacent areas.

[0048] Iterative Learning Control (ILC) is used to correct system errors: Using the polishing trajectory of the first bottle as a benchmark, the position error e_k(t) (where t is the time variable) at each trajectory point is recorded. After polishing, the actual material removal amount is measured using a laser tracker (with an accuracy of ±0.005 mm) and compared with the target material removal amount to obtain the material removal error ε_k(t). According to the error model: u_{k+1}(t) = u_k(t) + L * e_k(t) + Q * dε_k / dt, where L = 0.3 is the learning gain and Q = 0.15 is the differential gain.

[0049] The control input u_{k+1}(t) for the next bottle is updated. For example, if the neck region is detected to have been removed less than 5 microns, an additional velocity pulse (8% increment) is injected at that point in the trajectory in the next cycle. After 3-5 iterations, the position error is reduced to ±0.01 mm, and the material removal uniformity is improved to 95%. The surface roughness Ra of the final output glass bottle blank is reduced from the initial 3.2 microns to 0.8 microns, laying the foundation for subsequent fine polishing.

[0050] S203, the surface of the rough glass bottle blank after preliminary polishing is synchronously purged with an inert gas flow with a micro-bubble generator, while the composite sand cortex polishing piece is controlled to micro-vibrate, and residual polishing debris is removed through gas-solid coupling to obtain a finely polished semi-finished glass bottle blank; Specifically, argon gas can be introduced into a venturi tube and mixed with deionized water to generate microbubbles of a specific diameter at the gas-liquid interface through the ultrasonic cavitation effect, forming a gas-liquid two-phase flow, which serves as a medium for debris removal; After initial polishing, the surface of the rough glass bottle is covered with glass debris of varying sizes, abrasive residue, and trace amounts of oil. The cleaning process first requires the preparation of a highly efficient cleaning medium—a gas-liquid two-phase flow. This system uses high-purity argon (Ar) as an inert carrier gas. The purity must reach 99.999% (5N grade) to prevent oxygen from oxidizing the glass surface or forming watermarks. Argon is supplied from a high-pressure cylinder, stabilized at 0.5 MPa by a precision pressure reducing valve, and then introduced into the inlet of a Venturi tube. A Venturi tube is a converging and diverging pipe with a variable cross-section and an annular gap at the throat. Deionized water (DIW) (resistivity ≥ 18 megohm·cm) is injected into the inlet of the Venturi tube by a centrifugal pump at a flow rate of 3 liters / minute (L / min). When a high-speed argon gas flow (approximately 15 m / s) flows through the throat, it generates a local negative pressure (approximately -0.2 MPa) based on the Bernoulli principle, forcefully drawing in the deionized water and atomizing it into micron-sized droplets. The gas-liquid mixture undergoes thorough turbulent mixing in the diffuser section, forming a preliminary gas-liquid mixed flow.

[0051] The preliminarily mixed gas-liquid flow then enters the ultrasonic cavitation reaction chamber. This chamber is constructed of titanium alloy, with its inner wall coated with a corrosion-resistant ceramic coating. A piezoelectric ceramic ultrasonic transducer array (UTA) is integrated into its base. The transducer operates at a frequency of 40 kilohertz (kHz) and is driven by a high-frequency power supply to generate ultrasonic waves with a peak power of 800 watts (W). As the sound waves propagate through the fluid, they trigger the ultrasonic cavitation effect at the gas-liquid interface: the negative pressure phase of the sound wave causes the local pressure of the fluid to fall below the saturated vapor pressure, forming tiny cavities (cavitation nuclei). The positive pressure phase causes the cavities to collapse rapidly, instantaneously generating local high pressures exceeding 1000 atmospheres and high temperatures exceeding 5000 Kelvin (K). These extreme physical conditions cause the argon gas to be efficiently sheared and fragmented, combining with water molecules to form microbubbles of highly uniform diameter. By adjusting the ultrasonic power (adjustable from 300-800 W) and the cavitation chamber pressure (0.1-0.3 MPa), the bubble diameter can be precisely controlled within a specific range of 50 ± 5 microns (μm). This size of bubble has optimal surface adsorption capacity and rising stability, effectively enveloping and carrying debris.

[0052] The resulting gas-liquid two-phase flow (GLTPF) consists of 80% microbubbles by volume and a 20% deionized water continuous phase. This fluid is delivered via a pressure-resistant pipe to a porous nozzle mounted at the end of a robotic arm. The nozzle features an aperture of 200 microns and a density of 50 holes per square centimeter, ensuring a uniform curtain-like coverage of the glass bottle surface. Deionized water provides wetting and buffering, preventing scratches from high-hardness glass debris. Argon microbubbles, with their enormous surface area (approximately 12,000 square centimeters per milliliter), attract nano- to micron-sized debris and lift them off the surface through buoyancy. The entire media generation process is monitored by a programmable logic controller (PLC). Key parameters such as argon flow rate (standard: 5 liters / minute), water flow rate, ultrasonic power, and outlet bubble size (monitored online via a laser diffractometer) are all controlled through closed-loop feedback.

[0053] The distribution of debris on the surface of the glass bottle is monitored in real time by a high-speed camera, and the trajectory of the debris is calculated using particle image velocimetry technology. The injection angle and flow rate of the gas-liquid two-phase flow are dynamically adjusted based on the trajectory data to form precise purge parameters. To achieve targeted removal, the system uses a high-speed camera (HSC) mounted above the polishing station. This camera utilizes a CMOS sensor with a resolution of 1920×1080 pixels and a frame rate of up to 2000 frames per second (fps). A ring-shaped LED cold light source (5600K color temperature) provides shadowless illumination. The camera is tilted at a 60-degree angle toward the glass bottle surface, capturing the surface microtopography in real time. The captured images are transmitted to the image processing unit (IPU), where they undergo preprocessing, including Gaussian filtering for noise reduction, adaptive threshold segmentation for debris feature extraction, and morphological opening for artifact removal. In the resulting binary image, each connected domain represents a debris particle, and its coordinates are recorded using a centroid localization algorithm. The system completes a full-area scan every 0.1 seconds, generating a real-time, updated debris distribution heatmap (DDH), which annotates the debris' location, size (equivalent diameter), and density (number of particles per unit area).

[0054] To predict debris detachment, the system applies particle image velocimetry (PIV). This technology is based on an optical flow algorithm: the displacement vector Δd of a single debris in two consecutive image frames (with a time interval of Δt = 0.5 milliseconds) divided by Δt gives the instantaneous velocity vector v (unit: mm / s). By calculating the velocity field of tens of thousands of debris fragments across the entire field, a debris trajectory model (DTM) is constructed. This model predicts three key types of motion: 1) downward tendency due to gravity; 2) radial migration due to air drag; and 3) random deflection caused by interparticle collisions. For example, a 10-micron diameter glass fragment in a 0.2 m / s airflow is expected to move 8 mm in 0.3 seconds, while a 50-micron fragment, due to its greater inertia, will only move 2 mm. The model output is a dynamic trajectory probability cloud (TPC), which identifies the likely location of the debris within the next second.

[0055] Based on the debris distribution heat map (DDH) and trajectory probability cloud map (TPC), the Dynamic Blowing Control System (DBCS) generates precision blowing parameters (PBP). This system utilizes a fuzzy rule base to determine the following: 1) In high-debris density areas (>50 particles / mm²), the nozzle is tilted 45 degrees and the flow rate is increased to 0.8 m / s for impact removal. 2) In low-density areas (<10 particles / mm²), the nozzle is positioned vertically and the flow rate is reduced to 0.3 m / s to prevent overwetting. 3) Based on trajectory prediction, a pre-interception mode is activated for debris about to enter the bottle mouth: the nozzle is turned 0.5 seconds upstream of the trajectory to intercept it. All parameters are controlled by a servo motor to adjust the nozzle pitch angle (continuously adjustable from 0-90 degrees) and a proportional valve to adjust the gas and liquid flow rate (accuracy ±1%). The commands are updated every 0.2 seconds to ensure that the blowing strategy always matches the real-time operating conditions.

[0056] Based on the purge parameters, the composite sand cortex polishing piece is controlled to perform elliptical micro-vibration at a specific micro-vibration frequency. The amplitude and phase difference are optimized through vibration modal analysis to enhance the debris resonance separation effect. While the gas-liquid two-phase flow is being purged, a composite sandpaper polishing tool (CSPT) is endowed with micro-vibration to enhance synergy. The polishing tool is mounted on the end flange of a six-axis robotic arm using a quick-change fixture. A piezoelectric actuator array (PAA) is integrated on its back. The actuator consists of four lead zirconate titanate (PZT-8) piezoelectric ceramic discs arranged in a cross pattern, each independently retractable. When sinusoidal voltage signals with a 90-degree phase difference are applied to the two ceramic discs in the X-axis direction, an elliptical trajectory motion (ETM) is generated within the plane. The vibration parameters are dynamically triggered by the purge parameter (PBP): high-frequency vibration (800 Hz) corresponds to high purge flow rates (>0.6 m / s), while low-frequency vibration (200 Hz) corresponds to low flow rates. The vibration energy complements the kinetic energy of the fluid.

[0057] Key parameters of the elliptical trajectory—major axis amplitude A (range 1-10 microns), minor axis amplitude B (0.5-5 microns), and phase difference Φ (0-180 degrees)—were optimized through vibration modal analysis (VMA). The specific process involved: 1) A micro accelerometer (±50g range, 0.001g resolution) was attached to the surface of the polished workpiece to acquire vibration signals in real time; 2) the spectrum was decomposed using a fast Fourier transform (FFT) to identify the modal shapes at the dominant frequencies; 3) the optimal parameter combination was calculated based on a chip detachment efficiency model. For highly adherent metal abrasives (approximately 100 micronewtons), a "flat ellipse" with a large amplitude (A=8 microns) and high ellipticity (A / B=2.5) was used to induce shear debonding; for lightweight glass powder (adhesion 10 micronewtons), a "slender ellipse" with a small amplitude (A=2 microns) and high frequency (600 Hz) was used to induce resonant detachment. The optimized parameter set is called the Vibration Enhancement Factor (VEF).

[0058] Microvibration enhances debris removal through two mechanisms: 1) Resonance: When the vibration frequency approaches the natural frequency of the debris-matrix system (approximately 300-1000 Hz), the debris resonates, overcoming van der Waals forces. For example, a 20-micron diameter glass shaving can experience an acceleration of 200 g at 500 Hz, generating a detachment force exceeding 100 times its own weight. 2) Interfacial shear: The elliptical trajectory creates a tangential component of motion on the surface of the polished workpiece, exerting a continuous shear force on the debris. Combined with the flotation effect of the gas-liquid two-phase flow, the debris is more easily absorbed and carried away by the fluid. Field tests have shown that microvibration can increase debris removal efficiency by 55% while reducing purge fluid usage by 40%. During vibration, an accelerometer continuously monitors the actual motion trajectory. A PID controller (proportional-integral-differential controller) dynamically compensates for the hysteresis of the piezoelectric ceramic, ensuring vibration accuracy better than ±0.2 microns.

[0059] Ultraviolet radiation is introduced during the purging process to decompose organic residues through photocatalytic oxidation, and an electrostatic adsorption device is used to collect suspended particles to remove residual debris and obtain semi-finished glass bottles.

[0060] To prevent organic pollutants (such as lubricating oil and fingerprint grease) that may remain during the polishing process, the system integrates an ultraviolet irradiation module (UVIM) in the purge station. The module uses a low-pressure mercury lamp as a light source with a main wavelength of 254 nanometers (nm) and a radiation intensity of 30 milliwatts / square centimeter (mW / cm 2 After being focused by a quartz lens, the ultraviolet light covers the entire surface of the glass bottle at a 45-degree angle. Under the excitation of ultraviolet photons, the nano-titanium dioxide photocatalyst (particle size 20 nanometers, concentration 0.1%) pre-sprayed on the surface of the bottle generates electron-hole pairs. Holes (h + ) has strong oxidizing properties, directly breaking down the C-C bonds of organic matter. It also reacts with water molecules in the air to generate hydroxyl radicals (·OH), which have an oxidation potential of 2.8 volts (V) and can indiscriminately degrade pollutants such as grease and resin. A typical action time of 3-5 seconds is sufficient to break down the carbon chains of oily matter into CO2 and H2O.

[0061] Suspended particles carried away from the bottle by the gas and liquid flow must be effectively collected to prevent secondary sedimentation. The system deploys an electrostatic adsorption device (EAD) at the top of the workstation. This device consists of two parallel electrode plates: the upper electrode is a tungsten wire mesh, supplied with a +15 kilovolt (kV) high-voltage DC current; the lower electrode is a grounded, cooling water-cooled stainless steel dust collector. When the gas and liquid flow laden with debris passes through the gap between the electrodes, the particles are ionized and charged (primarily negatively) in the high-voltage electric field. Coulomb forces propel them toward the dust collector. The electric field strength is designed to be 5 kilovolts / cm (kV / cm), ensuring a capture efficiency of >99% for particles larger than 1 micron. The dust collector is coated with Teflon, and dust is automatically removed by a mechanical scraper (period of 60 seconds), which then falls into a sealed waste bin.

[0062] After the cleaning process is complete, the system initiates a final assessment: 1) Surface roughness is randomly inspected using a laser confocal microscope (Ra value <0.05 microns); 2) Water contact angle is measured using a contact angle meter (a value <5 degrees is required to demonstrate the absence of oil contamination); and 3) White glove wiping tests confirm no visible residue. Once these standards are met, the robotic arm transfers the glass bottle to the next process. The workpiece is now called a semi-finished glass bottle blank (SFGB). The surface of this semi-finished blank is free of macroscopic scratches, debris, or organic contamination, and its microscopic roughness is 90% lower than that of the rough blank, creating an ideal substrate for subsequent nano-level polishing. All process parameters (such as UV dose, electrostatic voltage, and cleaning time) are recorded in the MES (Manufacturing Execution System) for full process quality traceability.

[0063] S204: Place the finely polished semi-finished glass bottle in a sealed cavity, spray it with a suspension of nano-scale silica flexible abrasive particles, and perform final polishing in conjunction with pulsed ultrasonic vibration. The Brownian motion of the particles and the ultrasonic cavitation effect synergistically remove residual micron-level pollutants, resulting in a finished cosmetic glass bottle with a clean surface and a preset glossiness.

[0064] Specifically, silica particles of a specific particle size can be dispersed in deionized water, an appropriate amount of polyvinyl pyrrolidone can be added as a dispersant, and a stable suspension can be prepared by ultrasonic homogenization to serve as the grinding medium for the final polishing; The core of this step is to prepare a polishing medium with a precise nanoscale particle size distribution and high stability. First, silica particles are carefully screened for a specific particle size, with an average particle size within the range of 50 to 100 nanometers (e.g., 80 nanometers) and a coefficient of variation (CV) of less than 15 percent. This particle size selection is based on two considerations: first, it must be smaller than the wavelength of visible light (approximately 400-700 nanometers) to prevent optical scattering during the polishing process from affecting monitoring; second, it must possess sufficient Brownian motion activity to penetrate into microscopic recessed areas. The selected silica particles undergo a hydroxylation surface treatment, enriching their surface with silanol groups and enhancing their hydrophilicity. The resistivity of deionized water must be above 18 megohm·cm to ensure that no electrolyte interferes with dispersion stability. Polyvinylpyrrolidone (PVP) is used as a dispersant, with a molecular weight range of 40,000 to 60,000 Daltons (e.g., 50,000 Daltons), and the added concentration is precisely controlled between 0.5 and 0.8 mass fractions (e.g., 0.65 mass fractions). The pyrrolidone ring of PVP is adsorbed on the silica surface through van der Waals force, and its long molecular chain stretches in water to form a steric hindrance layer, which effectively resists particle agglomeration.

[0065] Ultrasonic homogenization treatment adopts a staged energy input strategy. In the first stage, the silica particles, PVP and deionized water are pre-mixed and placed in a titanium alloy variable amplitude ultrasonic processor for five minutes of strong dispersion at a frequency of 20 kHz and a power of 1500 watts. The high-intensity cavitation effect in this stage can break up primary agglomerates, and the acoustic streaming effect promotes macroscopic mixing. In the second stage, the frequency is switched to 40 kHz and the power is 800 watts for thirty minutes of gentle homogenization. In this stage, the cavitation intensity is controlled by adjusting the ultrasonic duty cycle to 50% (0.5 seconds on / 0.5 seconds off), focusing on optimizing the adsorption layer structure on the particle surface and eliminating microbubbles. The entire treatment process is carried out in a constant temperature water bath (25 degrees Celsius ± 1 degrees Celsius) to avoid local overheating that may cause degradation of the PVP molecular chain. The resulting stable suspension must meet three criteria: a zeta potential greater than 35 mV (e.g., -42 mV), indicating electrostatic stability; a particle size polydispersity index less than 0.15 as measured by dynamic light scattering, indicating uniformity; and a sedimentation volume ratio less than 5% after 24 hours of stasis, indicating resistance to sedimentation. This suspension, when stored in a sealed container under nitrogen, has a shelf life of 72 hours.

[0066] To ensure traceability and batch consistency of grinding media, an online quality monitoring system has been established. A UV-Vis spectrophotometer (measuring absorbance at 600 nm, correlating it to particle concentration) and a conductivity meter (monitoring ionic contamination) are integrated into the suspension delivery pipeline. If the absorbance fluctuates by more than 3% of the baseline value or the conductivity exceeds 5 microsiemens / cm, the system automatically triggers an alarm and halts feed. Each batch of suspension is randomly sampled using a scanning electron microscope to verify particle morphology (sphericity greater than 0.9 required) and the absence of hard agglomerates. This strictly controlled nano-grinding media performs three functions: flexible particles achieve micro-cutting through rolling friction; Brownian motion drives particles into submicron-sized defects; and surface active sites adsorb residual contaminants.

[0067] A pressure atomizing nozzle is used to spray the nano-abrasive medium onto the surface of the glass bottle in a fan-shaped spray. The spray particle size distribution is monitored in real time by a laser particle size analyzer, and the spray pressure is dynamically adjusted to form a uniform liquid film. The spray system utilizes a multi-stage precision control architecture. The pressure-type atomizing nozzle features a fan-shaped, wide-angle design, with a spray angle precisely set to 80 degrees ± 2 degrees, ensuring coverage of the curved surface of the glass bottle. The nozzle incorporates a vortex chamber and a precision slit (200 micron width) to control the liquid film breakup pattern by adjusting the inlet pressure. The grinding suspension is delivered by a high-pressure plunger pump, with an operating pressure range of 0.5 MPa to 3 MPa (initial pressure 1.2 MPa). A key innovation lies in dual-fluid assisted atomization: a primary channel delivers the grinding slurry, while an annular secondary channel introduces dry nitrogen (99.999% purity) at 80% of the primary pressure (e.g., 0.96 MPa). The laminar flow of nitrogen envelopes the atomizing cone, suppressing spray dispersion and preventing condensation from affecting the polishing environment. The nozzle array is evenly spaced according to the three-dimensional contour of the glass bottle, with spacing dynamically adjusted based on bottle diameter (e.g., a 60 mm diameter bottle corresponds to a 25 mm spacing).

[0068] A real-time monitoring system, the Laser Diffraction Analyzer (LDA), is crucial for ensuring liquid film uniformity. This instrument utilizes a helium-neon laser (wavelength, 632.8 nm) and a detector array covering a measurement range of 0.1 μm to 1000 μm. A detection window is located 150 mm from the nozzle outlet, collecting spray data ten times per second. Analysis software calculates key parameters in real time: volume average particle size (D50), span (D90-D10) / D50), and the proportion of coarse droplets larger than 5 μm. Dynamic adjustments are triggered when D50 deviates by ±5% from the target value (e.g., 75 μm) or the proportion of coarse droplets exceeds 0.1%. This adjustment logic is based on fuzzy proportional-integral control. If D50 is too high, indicating insufficient atomization, the system increases the plunger pump pressure in steps of 0.05 MPa per second. If the span increases, indicating a broadening of the particle size distribution, the nitrogen assist pressure is simultaneously increased by 3% to 5%.

[0069] Forming a uniform liquid film requires meeting three key criteria: film thickness uniformity error less than ±10% (target thickness 15 microns), coverage greater than 99.8%, and no dry spots or pooling areas. This implementation strategy includes: First, by collaboratively controlling the glass bottle rotation speed (e.g., 10 rpm) and nozzle movement speed (e.g., 20 mm / s), a motion controller calculates the spatial trajectory to ensure a 30% overlap rate. Second, hydrophobic micro-area positioning markers are pre-placed on the bottle surface. A high-speed camera (1000 frames / s) captures the liquid film spreading process. If a contact angle exceeding 90 degrees is detected, the nozzle pressure in that area is increased by 15%. Third, an infrared thermal imager monitors the liquid film evaporation status (temperature resolution 0.1°C), triggering additional spraying when the local temperature difference exceeds 0.5°C. This dynamic closed-loop system enables the nanoparticles to form a molecular-level lubricating film with self-healing properties on the glass surface.

[0070] Applying pulsed ultrasonic vibrations at a specific pulse frequency to a glass bottle covered with a uniform liquid film, the generation and collapse cycles of cavitation bubbles are controlled by duty cycle modulation. Fluid dynamics simulation is combined to optimize the transducer position and enhance the selective polishing effect. The pulsed ultrasonic vibration system consists of three main modules: a piezoelectric ceramic transducer array (made of PZT-8), a pulse power amplifier, and a digital signal generator. The specific pulse vibration frequency was selected to be 28 kHz ± 500 Hz. Modal analysis determined this frequency to avoid the natural resonant frequencies of the glass bottle (typically 18-22 kHz) to prevent structural damage. The pulse waveform utilizes rectangular modulation. Key parameters include a pulse width (Ton) range of 50 to 500 microseconds, a pulse interval (Toff) range of 100 to 1000 microseconds, and a peak acoustic power density of 0.5 to 3 W / cm² (initial value 1.8 W / cm²). The transducer transmits vibrations to the polishing tank floor via a hydraulic coupling rod, generating longitudinal waves in the liquid film. Frequency stability is controlled by a phase-locked loop circuit, with phase jitter less than 0.1 degrees.

[0071] Duty cycle modulation is the core technology for controlling cavitation behavior. The duty cycle is defined as Ton / (Ton + Toff), with an initial value of 40%. By adjusting the ratio of Ton to Toff, the following is achieved: when Ton is extended to 300 microseconds, it promotes the stable growth of cavitation bubble clusters and enhances the impact force of the microjet, making it suitable for removing deeply embedded contaminants. When Toff is extended to 800 microseconds, the recovery period after bubble collapse is prolonged, facilitating the redistribution of abrasive particles. A more advanced modulation employs a chaotic sequence algorithm: a non-periodic duty cycle variation sequence generated using the Lorentz attractor model ensures a random distribution of cavitation bubbles in space and time, eliminating uneven polishing caused by standing waves. Cavitation intensity is monitored in real time by a hydrophone array (center frequency 28 kHz, sensitivity -180 dB). When the sound pressure level exceeds 0.15 MPa, the duty cycle is automatically reduced by 5% to prevent micro-pitting on the glass surface.

[0072] Fluid dynamics optimization simulations were performed using 3D transient simulations using finite element analysis software. The model included the glass bottle geometry, the liquid film boundary layer, the nanoparticle two-phase flow, and the ultrasonic pressure field. Key optimization objectives were to achieve acoustic pressure focusing (enhancement of more than 2.5 times) in the surface convex regions while maintaining laminar flow protection in the concave regions. Simulations revealed that the transducers should be arranged in a spiral array (e.g., six groups spaced 60 degrees apart), with an optimal distance from the bottle wall of one-quarter wavelength (approximately 13 mm in water). A 15-degree tilt angle suppressed surface wave interference. Experimental validation employed particle image velocimetry: fluorescent tracer particles (5 microns) were added to the liquid film, illuminated with a dual-pulsed laser sheet, and the flow field captured with a high-speed camera. Post-optimization data revealed a threefold increase in vorticity and a shear stress of 12 Pascals in the micro-convex regions, while the flow velocity in the concave regions was kept below 0.2 m / s. This selective polishing effect reduced the surface profile arithmetic mean deviation (Ra) from 0.3 microns to 0.02 microns.

[0073] The polishing effect is monitored in real time through an online gloss meter, and the ultrasonic power and nanoparticle concentration are dynamically adjusted based on the monitoring data. When the glossiness reaches the preset standard, the cleaning program is triggered, and the grinding medium is recovered through the circulating filtration system to obtain a finished cosmetic glass bottle with a clean surface and a preset glossiness.

[0074] The online gloss meter uses a 60-degree measurement geometry (ISO 2813), scanning at 20 points per second. The optical system includes a halogen light source (color temperature 2856K), an interference filter (peak wavelength 560 nm), and a silicon photodetector. A pneumatic lift mechanism contacts the bottle surface during the polishing interval (Toff period), with the measuring head applying a constant contact force of 0.5 Newtons to ensure repeatability. The measured data is de-noised using a Kalman filter, and two key values ​​are calculated: specular reflectance (Gs, measured in GU) and haze (indicating the proportion of scattered light). Preset gloss standards are set based on customer requirements (for example, high-end perfume bottles require Gs ≥ 95 GU and Haze ≤ 2%). The system establishes a gloss-roughness mapping database: when Ra = 0.02 microns, Gs = 97 GU, and every 0.01 micron increase in Ra results in a 6 GU decrease in Gs.

[0075] The dynamic adjustment strategy utilizes multivariable decoupling control. First, when the real-time Gs value falls below the target trajectory (for example, the actual value is 75 GU, but should reach 80 GU in the third minute), the ultrasonic power is increased in steps of 0.2 W / cm². The cavitation noise spectrum is simultaneously monitored. If a sudden increase in energy occurs in the high-frequency band (>100 kHz) within the 1 / 3 octave band, the boost is stopped. Second, when the haze value exceeds the target, indicating the presence of micro-scratches or residual particles on the surface, the system increases the nanoparticle concentration from the initial 5% by mass to 7%, and simultaneously increases the PVP dispersant ratio to 0.8%. Third, a polishing efficiency prediction model is established: if the Gs increase rate per unit time is less than 0.3 GU per second, the ultrasonic duty cycle and spray flow rate are simultaneously increased by 10% and 15%. All adjustment commands are transmitted to the actuator via the OPC protocol, with a response latency of less than 100 milliseconds.

[0076] The circulating filtration and finished product delivery process involves four precise steps. First, when the online gloss meter measures within the acceptable range for three consecutive times (e.g., Gs = 96 GU ± 0.5 GU), an audible and visual alarm is triggered, initiating a cleaning process. Second, the circulating filtration system immediately switches valves, allowing the polishing fluid to flow through a multi-stage, serially connected filter. The first stage, a stainless steel sintered filter element (5 micron pore size), intercepts large particles; the second stage, a hollow fiber ultrafiltration membrane (100,000 Dalton molecular weight cutoff), removes aggregates; and the final electrodialysis desalination module maintains water quality. The recovery rate exceeds 95%, and the filtrate is replenished with fresh polishing fluid for reuse. Third, the glass bottles undergo a three-stage cleaning process: ultrasonic rinsing with deionized water at 80°C (40 kHz, 3 minutes); vapor degreasing with anhydrous ethanol (50°C, 2 minutes); and negative ion air knife drying (30 m / s wind speed). Fourth, the finished product passes through an automated sorting line, where a machine vision system rechecks for surface defects (2 micron resolution); a laser interferometer spot-checks for surface accuracy (λ / 10). Acceptable products are then coded and packaged. The surface cleanliness of the final cosmetic glass bottle reaches ISO 14644-1 Class 5, with a gloss fluctuation range of ±1GU, achieving an optical surface with nanometer-level precision.

[0077] It can be seen that the sand cortex base material is modified with a gradient mesh abrasive composite, and three-dimensional cutting is performed according to the curved surface contour of the glass bottle through laser cutting technology to form a composite sand cortex polished part; the composite sand cortex polished part is assembled on the execution end of the six-axis linkage robot arm, and the partitioned differentiated polishing is implemented in combination with the surface pollutant distribution density image recognition result to obtain the glass bottle rough blank after preliminary polishing; on the surface of the glass bottle rough blank, an inert gas flow with a microbubble generating device is used for synchronous blowing, and the composite sand cortex polishing part is controlled to perform micro-vibration at the same time to obtain a semi-finished glass bottle blank; the semi-finished glass bottle blank is placed in a sealed cavity for spraying, and finally polished with pulsed ultrasonic vibration to obtain a finished cosmetic glass bottle, thereby achieving high-precision adaptive polishing of the glass bottle curved surface and improving surface cleanliness and gloss uniformity.

[0078] Another embodiment of the present invention provides a cosmetic glass bottle surface polishing system, see Figure 3 , the system may include: The cutting module 301 is used to perform a gradient abrasive composite modification on the sand cortex base material, and to perform three-dimensional cutting according to the curved surface contour of the glass bottle through laser cutting technology to form a composite sand cortex polished piece that precisely fits the curved surface of the bottle body; Polishing module 302 is used to assemble the composite sand cortex polishing piece on the execution end of the six-axis linkage robot arm, obtain three-dimensional surface morphology data of the glass bottle through infrared scanning, dynamically adjust the speed of the servo motor and the contact pressure of the polishing piece based on the morphology data, and implement zone-by-zone differentiated polishing based on the surface contaminant distribution density image recognition results to obtain the rough glass bottle after preliminary polishing; The purge module 303 is used to synchronously purge the surface of the glass bottle rough blank after preliminary polishing with an inert gas flow with a micro-bubble generator, while controlling the composite sand cortex polishing piece to perform micro-vibration, thereby removing the residual polishing debris through gas-solid coupling to obtain a finely polished glass bottle semi-finished blank; The spray module 304 is used to place the finely polished semi-finished glass bottle in a sealed cavity, spray it with a suspension of nano-scale silica flexible abrasive particles, and perform final polishing in conjunction with pulsed ultrasonic vibration. The Brownian motion of the particles and the ultrasonic cavitation effect synergistically remove residual micron-level pollutants to obtain a finished cosmetic glass bottle with a clean surface and a preset glossiness.

[0079] An embodiment of the present invention further provides a storage medium storing a computer program, wherein the computer program is configured to execute the steps of any one of the above method embodiments when running.

[0080] Specifically, in this embodiment, the above-mentioned storage medium may be configured to store a computer program for performing the following steps: S201: The sand leather base material is modified with a gradient abrasive composite. Laser cutting technology is used to perform three-dimensional cutting according to the curved surface contour of the glass bottle to form a composite sand leather polished part that precisely fits the curved surface of the bottle. S202: Assemble the composite sand cortex polishing piece on the execution end of the six-axis linkage robot arm, obtain three-dimensional surface morphology data of the glass bottle through infrared scanning, dynamically adjust the speed of the servo motor and the contact pressure of the polishing piece based on the morphology data, and implement zone-by-zone differentiated polishing in combination with the surface contaminant distribution density image recognition results to obtain a rough glass bottle after preliminary polishing; S203, the surface of the rough glass bottle blank after preliminary polishing is synchronously purged with an inert gas flow with a micro-bubble generator, while the composite sand cortex polishing piece is controlled to micro-vibrate, and residual polishing debris is removed through gas-solid coupling to obtain a finely polished semi-finished glass bottle blank; S204: Place the finely polished semi-finished glass bottle in a sealed cavity, spray it with a suspension of nano-scale silica flexible abrasive particles, and perform final polishing in conjunction with pulsed ultrasonic vibration. The Brownian motion of the particles and the ultrasonic cavitation effect synergistically remove residual micron-level pollutants, resulting in a finished cosmetic glass bottle with a clean surface and a preset glossiness.

[0081] An embodiment of the present invention further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any one of the above method embodiments.

[0082] Specifically, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0083] Specifically, in this embodiment, the processor may be configured to execute the following steps through a computer program: S201: The sand leather base material is modified with a gradient abrasive composite. Laser cutting technology is used to perform three-dimensional cutting according to the curved surface contour of the glass bottle to form a composite sand leather polished part that precisely fits the curved surface of the bottle. S202: Assemble the composite sand cortex polishing piece on the execution end of the six-axis linkage robot arm, obtain three-dimensional surface morphology data of the glass bottle through infrared scanning, dynamically adjust the speed of the servo motor and the contact pressure of the polishing piece based on the morphology data, and implement zone-by-zone differentiated polishing in combination with the surface contaminant distribution density image recognition results to obtain a rough glass bottle after preliminary polishing; S203, the surface of the rough glass bottle blank after preliminary polishing is synchronously purged with an inert gas flow with a micro-bubble generator, while the composite sand cortex polishing piece is controlled to micro-vibrate, and residual polishing debris is removed through gas-solid coupling to obtain a finely polished semi-finished glass bottle blank; S204: Place the finely polished semi-finished glass bottle in a sealed cavity, spray it with a suspension of nano-scale silica flexible abrasive particles, and perform final polishing in conjunction with pulsed ultrasonic vibration. The Brownian motion of the particles and the ultrasonic cavitation effect synergistically remove residual micron-level pollutants, resulting in a finished cosmetic glass bottle with a clean surface and a preset glossiness.

[0084] The above describes in detail the structure, features and effects of the present invention based on the embodiments shown in the drawings. The above is only a preferred embodiment of the present invention, but the scope of implementation of the present invention is not limited to what is shown in the drawings. Any changes made in accordance with the concept of the present invention, or modifications to equivalent embodiments with equivalent changes, which do not exceed the spirit covered by the description and drawings, should be within the scope of protection of the present invention.

Claims

1. A method for polishing the surface of a cosmetic glass bottle, characterized in that: The method comprises: The sand leather base material is modified with a gradient abrasive composite, and three-dimensionally cut according to the curved surface contour of the glass bottle through laser cutting technology to form a composite sand leather polished part that precisely fits the curved surface of the bottle body; The composite sand cortex polishing piece is assembled on the execution end of a six-axis linkage robot arm. The three-dimensional morphology data of the glass bottle surface is obtained through infrared scanning. The speed of the servo motor and the contact pressure of the polishing piece are dynamically adjusted based on the morphology data. Combined with the surface contaminant distribution density image recognition results, zone-by-zone differentiated polishing is implemented to obtain the rough glass bottle after preliminary polishing. On the surface of the rough glass bottle after preliminary grinding, an inert gas flow with a micro-bubble generator is used for synchronous blowing, while the composite sand cortex grinding piece is controlled to perform micro-vibration, and the residual grinding debris is removed through the gas-solid coupling effect to obtain a finely ground semi-finished glass bottle blank; The finely polished semi-finished glass bottle is placed in a sealed cavity, sprayed with a suspension of nano-scale silica flexible abrasive particles, and finally polished with pulsed ultrasonic vibration. The Brownian motion of the particles and the ultrasonic cavitation effect synergistically remove residual micron-level pollutants, resulting in a finished cosmetic glass bottle with a clean surface and a preset glossiness.

2. The method according to claim 1, characterized in that The method comprises the following steps: performing a composite modification of the sand leather base material with a gradient mesh abrasive, and performing three-dimensional cutting according to the curved surface contour of the glass bottle by laser cutting technology to form a composite sand leather polished piece that precisely fits the curved surface of the bottle body. The method comprises: The natural sand cortex material is placed in a supercritical carbon dioxide fluid. By adjusting the pressure and temperature to control the fluid density, the nanoscale pore structure of the sand cortex fiber is regulated to form a porous carrier structure, providing anchor points for subsequent abrasive attachment. Using electrophoretic deposition technology, 800 mesh, 1500 mesh, and 3000 mesh zirconium oxide abrasives are sequentially deposited on the surface of the porous carrier structure. The abrasive deposition rate is controlled by pulse current density to form a gradient abrasive layer, forming a gradient abrasive composite substrate. Using a femtosecond laser micro-nano processing system, the gradient abrasive composite substrate is contour-tracked and cut based on the 3D model data of the cosmetic glass bottle. A micro-serrated joint is generated on the cutting edge to complete the preparation of the 3D pre-cut polished part. The three-dimensional pre-cut polished part is immersed in a silane coupling agent solution for surface activation, and then hot-pressed and composited with a shape memory polymer substrate. The shape memory effect of the substrate gives the polished part adaptive fitting ability to form a composite sand leather polished part.

3. The method according to claim 2, characterized in that The composite sand cortex polishing piece is assembled on the execution end of the six-axis linkage robot arm, and three-dimensional morphological data of the glass bottle surface is obtained by infrared scanning. The rotation speed of the servo motor and the contact pressure of the polishing piece are dynamically adjusted based on the morphological data. In combination with the surface contaminant distribution density image recognition result, zone-differentiated polishing is performed to obtain a rough glass bottle after preliminary polishing, including: An infrared thermal imager and a structured light scanner are integrated into the robotic arm's execution end to perform multimodal scanning of the glass bottle surface. Thermal wave imaging technology is used to identify sub-surface defects, and the scan data is fused to generate a 3D point cloud dataset containing morphology and defect information. Based on a 3D point cloud dataset, a dynamic time warping algorithm is used to match a standard bottle model, calculate the local curvature deviation map, and combine it with a deep learning semantic segmentation model to identify the type of surface contaminants and their distribution density, thereby constructing a differentiated polishing strategy map. Taking the curvature deviation map and contaminant distribution density as input, the servo motor speed and contact pressure parameters are deduced in real time through the fuzzy control rule library to form a dynamic control parameter set; According to the dynamic control parameter set, the six-axis robotic arm is driven to move along the preset trajectory, and the partitioned grinding operation is performed synchronously. The dynamic error of the robotic arm is corrected through the iterative learning control algorithm to complete the preliminary grinding to obtain the rough glass bottle blank.

4. The method according to claim 3, characterized in that The surface of the rough glass bottle blank after preliminary polishing is synchronously purged with an inert gas flow with a micro-bubble generating device, and the composite sand cortex polishing piece is controlled to perform micro-vibration, and residual polishing debris is removed through gas-solid coupling to obtain a finely polished semi-finished glass bottle blank, including: Argon is introduced into the venturi tube and mixed with deionized water. Microbubbles of a specific diameter are generated at the gas-liquid interface through the ultrasonic cavitation effect, forming a gas-liquid two-phase flow, which serves as a medium for debris removal. The distribution of debris on the surface of the glass bottle is monitored in real time by a high-speed camera, and the trajectory of the debris is calculated using particle image velocimetry technology. The injection angle and flow rate of the gas-liquid two-phase flow are dynamically adjusted based on the trajectory data to form precise purge parameters. Based on the purge parameters, the composite sand cortex polishing piece is controlled to perform elliptical micro-vibration at a specific micro-vibration frequency. The amplitude and phase difference are optimized through vibration modal analysis to enhance the debris resonance separation effect. Ultraviolet radiation is introduced during the purging process to decompose organic residues through photocatalytic oxidation, and an electrostatic adsorption device is used to collect suspended particles to remove residual debris and obtain semi-finished glass bottles.

5. The method according to claim 4, characterized in that The finely polished semi-finished glass bottle blank is placed in a sealed cavity, sprayed with a suspension of nano-scale silica flexible abrasive particles, and finally polished in conjunction with pulsed ultrasonic vibration. The Brownian motion of the particles and the ultrasonic cavitation effect synergistically remove residual micron-sized pollutants to obtain a finished cosmetic glass bottle with a clean surface and a preset glossiness, including: Silica particles of a specific particle size are dispersed in deionized water, an appropriate amount of polyvinyl pyrrolidone is added as a dispersant, and a stable suspension is prepared by ultrasonic homogenization to serve as the grinding medium for the final polishing; A pressure atomizing nozzle is used to spray the nano-abrasive medium onto the surface of the glass bottle in a fan-shaped spray. The spray particle size distribution is monitored in real time by a laser particle size analyzer, and the spray pressure is dynamically adjusted to form a uniform liquid film. Applying pulsed ultrasonic vibrations at a specific pulse frequency to a glass bottle covered with a uniform liquid film, the generation and collapse cycles of cavitation bubbles are controlled by duty cycle modulation. Fluid dynamics simulation is combined to optimize the transducer position and enhance the selective polishing effect. The polishing effect is monitored in real time through an online gloss meter, and the ultrasonic power and nanoparticle concentration are dynamically adjusted based on the monitoring data. When the glossiness reaches the preset standard, the cleaning program is triggered, and the grinding medium is recovered through the circulating filtration system to obtain a finished cosmetic glass bottle with a clean surface and a preset glossiness.

6. A cosmetic glass bottle surface polishing system, characterized in that: The system comprises: The cutting module is used to modify the sand leather base material with a gradient abrasive composite, and to perform three-dimensional cutting according to the curved surface contour of the glass bottle through laser cutting technology to form a composite sand leather polished part that precisely fits the curved surface of the bottle body; The polishing module is used to assemble the composite sand cortex polishing piece on the execution end of the six-axis linkage robot arm, obtain the three-dimensional topography data of the glass bottle surface through infrared scanning, dynamically adjust the speed of the servo motor and the contact pressure of the polishing piece based on the topography data, and implement zone-by-zone differentiated polishing based on the image recognition results of the surface contaminant distribution density to obtain the rough glass bottle after preliminary polishing; The purge module is used to synchronously purge the surface of the rough glass bottle blank after preliminary polishing with an inert gas flow equipped with a micro-bubble generator, while controlling the composite sand cortex polishing piece to perform micro-vibration, thereby removing the residual polishing debris through gas-solid coupling to obtain a finely polished semi-finished glass bottle blank; The spray module is used to place the finely polished semi-finished glass bottle in a sealed cavity, spray it with a suspension of nano-scale silica flexible abrasive particles, and perform final polishing in conjunction with pulsed ultrasonic vibration. The Brownian motion of the particles and the ultrasonic cavitation effect synergistically remove residual micron-level pollutants, resulting in a finished cosmetic glass bottle with a clean surface and a preset glossiness.

7. The system according to claim 6, characterized in that The cutting module is specifically used for: The natural sand cortex material is placed in a supercritical carbon dioxide fluid. By adjusting the pressure and temperature to control the fluid density, the nanoscale pore structure of the sand cortex fiber is regulated to form a porous carrier structure, providing anchor points for subsequent abrasive attachment. Using electrophoretic deposition technology, 800 mesh, 1500 mesh, and 3000 mesh zirconium oxide abrasives are sequentially deposited on the surface of the porous carrier structure. The abrasive deposition rate is controlled by pulse current density to form a gradient abrasive layer, forming a gradient abrasive composite substrate. Using a femtosecond laser micro-nano processing system, the gradient abrasive composite substrate is contour-tracked and cut based on the 3D model data of the cosmetic glass bottle. A micro-serrated joint is generated on the cutting edge to complete the preparation of the 3D pre-cut polished part. The three-dimensional pre-cut polished part is immersed in a silane coupling agent solution for surface activation, and then hot-pressed and composited with a shape memory polymer substrate. The shape memory effect of the substrate gives the polished part adaptive fitting ability to form a composite sand leather polished part.

8. The system according to claim 7, characterized in that The polishing module is specifically used for: An infrared thermal imager and a structured light scanner are integrated into the robotic arm's execution end to perform multimodal scanning of the glass bottle surface. Thermal wave imaging technology is used to identify sub-surface defects, and the scan data is fused to generate a 3D point cloud dataset containing morphology and defect information. Based on a 3D point cloud dataset, a dynamic time warping algorithm is used to match a standard bottle model, calculate the local curvature deviation map, and combine it with a deep learning semantic segmentation model to identify the type of surface contaminants and their distribution density, thereby constructing a differentiated polishing strategy map. Taking the curvature deviation map and contaminant distribution density as input, the servo motor speed and contact pressure parameters are deduced in real time through the fuzzy control rule library to form a dynamic control parameter set; According to the dynamic control parameter set, the six-axis robotic arm is driven to move along the preset trajectory, and the partitioned grinding operation is performed synchronously. The dynamic error of the robotic arm is corrected through the iterative learning control algorithm to complete the preliminary grinding to obtain the rough glass bottle blank.

9. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 5 when executed.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the method according to any one of claims 1 to 5.

Images