Method and system for digitally tracing quality of light-weight glass bottle

Abstract

The invention relates to the technical field of digital quality management, in particular to a light-weight glass bottle quality digital traceability method and system. The method comprises the following steps: acquiring glass bottle design data, and constructing a lightweight glass bottle model; detecting the structural strength of the bottle body based on the lightweight glass bottle model; carrying out bottle body surface defect detection based on the bottle body structural strength to generate bottle body surface defect data; performing defect tracing based on the bottle body surface defect data to obtain bottle body production defect process data; carrying out wall thickness non-uniformity anomaly detection based on the structural strength of the bottle body to obtain wall thickness non-uniformity data; the alignment offset of the mold core and the mold cavity is evaluated based on the wall thickness non-uniformity data; and carrying out positioning pin fracture detection according to the alignment offset of the mold core and the mold cavity to obtain positioning pin fracture data. The glass bottle production process is optimized based on the digital quality management technology, and the quality recognition rate, the abnormal response speed and the production process transparency are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital quality management, and particularly to a method and system for digital traceability of the quality of lightweight glass bottles. Background Art

[0002] Traditional digital traceability of glass bottle quality usually only relies on data from a single production link (such as the forming process or cold-end inspection) for quality records, failing to cover the multi-source data involved in the entire life cycle of glass bottles, resulting in missing or incomplete traceability information; in addition, this system generally uses static barcodes or batch numbers as identification means, unable to uniquely identify each glass bottle, so that when quality anomalies occur, it can only be located to a certain batch rather than a specific single product, affecting the efficiency of accurate backtracking and liability traceability; at the same time, traditional systems lack sufficient guarantee for the real-time data collection and the continuity of transmission, and data is prone to lag, omission or breakpoints, seriously affecting the reliability of the traceability system; in terms of data processing methods, most systems still mainly rely on manual judgment or single-dimensional index analysis, lacking the ability of multi-dimensional fusion analysis and trend prediction of key process parameters such as temperature, pressure, and thickness, restricting the early perception and proactive intervention of quality hazards; moreover, in terms of system integration, traditional solutions are often limited to data sharing within local equipment or local area networks, lacking a cross-section, cross-system collaborative linkage mechanism, and it is difficult to achieve end-to-end full-process quality traceability, affecting the transparency and intelligent level of the entire manufacturing chain, and ultimately being unfavorable for lightweight glass bottle production enterprises to efficiently identify, quickly respond to and pursue full liability for quality risks. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide a method and system for digital traceability of the quality of lightweight glass bottles to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for digital traceability of the quality of lightweight glass bottles includes the following steps: Step S1: Obtain the design data of the glass bottle and construct a lightweight glass bottle model; detect the structural strength of the bottle body based on the lightweight glass bottle model; Step S2: Detect the surface defects of the bottle body based on the structural strength of the bottle body to generate bottle body surface defect data; perform defect traceability based on the bottle body surface defect data to obtain bottle body production defect process data; Step S3: Detect the wall thickness unevenness anomaly based on the structural strength of the bottle body to obtain wall thickness unevenness data; evaluate the alignment offset of the core and cavity based on the wall thickness unevenness data; perform positioning pin fracture detection according to the alignment offset of the core and cavity to obtain positioning pin fracture data; Step S4: Generate a unique glass bottle number based on the lightweight glass bottle model; identify abnormal batches of lightweight glass bottles according to the process data of bottle body production defects and the data of positioning pin fractures to obtain abnormal batch data; perform abnormal glass number mapping on the abnormal batch data according to the unique glass number to obtain abnormal glass numbers, and transmit them to the production equipment system to execute the glass quality traceability task.

[0005] Through the realization of the digital traceability of the quality of lightweight glass bottles, the present invention solves the problems existing in the traditional system, such as data loss, insufficient unique identification, lack of real-time performance, and limitations in data analysis. In the design stage of the glass bottle, through lightweight design, the bottle body structure is optimized, which not only improves the strength of the bottle body but also reduces the consumption of raw materials, thus reducing production costs and meeting the requirements of environmental protection. Through the detection of the structural strength of the bottle body based on the lightweight model, structural weaknesses are identified in advance, improving the overall safety and durability of the glass bottle and providing strong data support for subsequent quality control. During the detection of surface defects on the bottle body, by generating defect data, not only can surface problems be accurately identified, but also defect traceability can be carried out based on these data, thereby clarifying the process problems existing in the production process and promoting the optimization and improvement of the production process. In addition, the abnormal detection of uneven wall thickness can effectively identify the problem of uneven wall thickness in the production of the bottle body, and evaluate the alignment accuracy of the core and cavity according to the data of uneven wall thickness, thereby further detecting and predicting the fracture problem of the positioning pin, avoiding the damage caused by insufficient strength of the positioning pin, and ensuring the stability of the entire production process. Through this series of steps, a unique number can be generated for each bottle through the lightweight glass bottle model, ensuring that each glass bottle has a clear identity identification, providing an accurate basis for subsequent quality traceability. In the process of abnormal batch identification and unique glass number mapping, by combining the process data of production defects and the data of positioning pin fractures, each abnormal product can be accurately located and transmitted to the production equipment system for processing in a timely manner. This not only improves the accuracy of quality traceability but also strengthens the ability to quickly respond to and process abnormal products, thereby improving the intelligent level of the entire production system. This method can ultimately ensure the quality traceability and effective management of the entire process, thereby realizing the efficient identification, rapid response, and full-process accountability of quality risks in the whole life cycle of lightweight glass bottle production enterprises.

[0006] Preferably, this specification also provides a system for digital traceability of the quality of lightweight glass bottles, which is used to execute the method for digital traceability of the quality of lightweight glass bottles as described above. The system for digital traceability of the quality of lightweight glass bottles includes: A bottle body structural strength detection module, which is used to obtain the design data of the glass bottle and construct a lightweight glass bottle model; detect the structural strength of the bottle body based on the lightweight glass bottle model; A defect traceability module is used to detect surface defects of the bottle body based on the structural strength of the bottle body, generate surface defect data of the bottle body; and perform defect traceability based on the surface defect data of the bottle body to obtain process data of production defects of the bottle body. A locating pin fracture detection module is used to detect abnormal wall thickness non-uniformity based on the structural strength of the bottle body to obtain wall thickness non-uniformity data; evaluate the alignment offset of the core and cavity based on the wall thickness non-uniformity data; and perform locating pin fracture detection according to the alignment offset of the core and cavity to obtain locating pin fracture data. A glass quality traceability module is used to generate a unique glass number based on a lightweight glass bottle model; identify abnormal batches of lightweight glass bottles according to the process data of production defects of the bottle body and the locating pin fracture data to obtain abnormal batch data; perform mapping of abnormal glass numbers on the abnormal batch data according to the unique glass number to obtain abnormal glass numbers, and transmit them to the production equipment system to execute the glass quality traceability task.

[0007] The lightweight glass bottle quality digital traceability system of the present invention can implement any one of the lightweight glass bottle quality digital traceability methods of the present invention, and is a medium for coordinating operations and signal transmission between various modules to complete the lightweight glass bottle quality digital traceability method. The internal modules of the system cooperate with each other, optimize the glass bottle production process, and effectively improve the quality recognition rate, abnormal response speed, and production process transparency. Description of the Drawings

[0008] By reading the detailed description of the non-restrictive embodiments with reference to the following drawings, other features, objects, and advantages of the present invention will become more apparent: Figure 1 It is a schematic diagram of the step flow of a lightweight glass bottle quality digital traceability method of the present invention; Figure 2 It is a detailed schematic diagram of the step flow of step S1 in the present invention; The realization, functional features, and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiments

[0009] The technical method of the present invention patent will be clearly and completely described below with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0010] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0011] It should be understood that although terms such as "first" and "second" may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0012] To achieve the above object, please refer to Figures 1 to 2 , the present invention provides a method for digital traceability of the quality of lightweight glass bottles, and the method includes the following steps: Step S1: Obtain the design data of the glass bottle and construct a lightweight glass bottle model; detect the structural strength of the bottle body based on the lightweight glass bottle model; In this embodiment, through the "parametric modeling" module in the three-dimensional mechanical design software SolidWorks, the historical structural design drawings of the target glass bottle are obtained. After importing the design drawings, the initial parameters are set, including the bottle body diameter of 70 mm, the inner diameter of the bottle mouth of 28 mm, the bottle height of 200 mm, and the bottle bottom thickness of 5.5 mm. For the lightweight target, combined with the existing forming process limits, the bottle bottom thickness is set to 4.0 mm, the bottle wall thickness is set to 1.6 mm, and the target bottle body weight is controlled within 240 g. After the parameter update, a new lightweight structural design drawing is generated. Subsequently, the ANSYS structural analysis module is used to perform a static strength analysis on the new structure. The input parameters include the internal air pressure value of 0.5 MPa, the Young's modulus of the glass material of 62 GPa, the Poisson's ratio of 0.23, and the maximum allowable stress of 18 MPa. The entire bottle body is subjected to finite element mesh division under the internal pressure loading condition, using tetrahedral elements, and the mesh density is set to 50 elements per cubic millimeter. After the loading and solution, the stress nephogram and displacement results are extracted. If the equivalent stress value at any position in the structure exceeds 18 MPa, then return to readjust the geometric parameters of the bottle body until the stress of all parts is less than the set threshold.

[0013] Step S2: Based on the structural strength of the bottle body, detect surface defects of the bottle body to generate bottle body surface defect data; based on the bottle body surface defect data, conduct defect traceability to obtain bottle body production defect process data; In this embodiment, the lightweight glass bottle sample after strength verification is sent into the cold-end vision detection device for surface defect detection of the bottle body. The detection equipment uses a double-sided ring LED lighting system, and cooperates with a Basler acA1920-40gm industrial camera for image acquisition. The resolution is 1920×1200 pixels, the exposure time is set to 800 microseconds, and the frame rate is set to 25 frames per second. After image acquisition, gray normalization processing is performed through the image enhancement module in the Halcon vision software, and edge extraction is performed using the edge gradient method. The edge gradient threshold is set to 28 gray units during the extraction process. Abnormal areas such as depressions, bubbles, and cracks extracted from the image are classified and identified using the area threshold method: the judgment criterion for cracks is that the minimum curvature radius is less than 0.3 mm and the length is greater than 2.0 mm; the bubble diameter is greater than 1.5 mm; the depression depth is greater than 0.2 mm. The identified defect type, position coordinates, and size information constitute the bottle body surface defect data. Subsequently, by comparing the production process records corresponding to the surface defect area of the bottle body, including the pressing temperature (for example, the forming die temperature corresponding to the bottom area of the bottle is recorded as 1180°C), the cooling rate, the mold opening and closing time, etc., mapping comparison is performed using the time synchronization mechanism and the bottle body positioning code. If the defect position shows abnormalities in the process parameters during a specific time period (such as the cooling rate is lower than 40°C / s), then mark the production process parameters of this section as the bottle body production defect process data and form a process traceability chain.

[0014] Step S3: Based on the structural strength of the bottle body, detect wall thickness unevenness anomalies to obtain wall thickness unevenness data; based on the wall thickness unevenness data, evaluate the alignment offset of the core and cavity; according to the alignment offset of the core and cavity, conduct positioning pin fracture detection to obtain positioning pin fracture data; In this embodiment, a non-contact ultrasonic wall thickness detection device (such as GE CL5 type) is used to perform a 360-degree rotation measurement on the entire circumference of the glass bottle sample that has passed the strength test. The detection resolution is set to 0.05 mm, and the measurement path covers the vertical axis of the bottle body from the bottle shoulder to the bottle bottom area. One measurement is taken every 10 mm, and 360 angular points are collected in each circle. The obtained wall thickness data is processed by a Matlab data processing script to calculate the standard deviation σ and the maximum deviation value Δt of each measurement point. If σ exceeds 0.15 mm or Δt exceeds 0.25 mm, it is considered that there is wall thickness unevenness. Record the azimuth difference between the maximum wall thickness value and the minimum wall thickness value in this area. Combining the die structure data, calculate the alignment offset Δθ between the core and the cavity, using the formula Δθ = atan[(t_max - t_min) / r], where r is the radius of the bottle body, and t_max and t_min are the maximum and minimum wall thickness values of this cross-section. If Δθ is greater than 3.5 degrees, it is marked as abnormal alignment of the core and the cavity. When the alignment offset is abnormal, further perform a fracture detection on the positioning pin of the pressing die. Use a structured light three-dimensional imaging device (such as Gocator 2400 series) to obtain the topography image of the positioning pin area during die pressing, and use the contour break point method to judge the integrity of the positioning pin. If the continuous interruption length of the scanned cross-section exceeds 1.0 mm or the depth of the positioning pin is less than 50% of the designed depth (such as 8.0 mm), it is determined that the positioning pin is fractured. Record its station number, failure time, and the corresponding die number to generate positioning pin fracture data.

[0015] Step S4: Generate a unique glass number based on the lightweight glass bottle model; identify abnormal batches of lightweight glass bottles according to the process data of bottle body production defects and the positioning pin fracture data to obtain abnormal batch data; perform abnormal glass number mapping on the abnormal batch data according to the unique glass number to obtain abnormal glass numbers, and transmit them to the production equipment system to execute the glass quality traceability task.

[0016] In this embodiment, immediately after the lightweight glass bottle is removed from the mold, a laser coding device (such as the KEYENCE ML-Z series) is used to etch a 12-digit unique identification number in the outer ring area of the bottle bottom. The number format is YYMMDD + mold number + serial number. For example, "250514A1327" represents the 1327th bottle of mold A on May 14, 2025. The etching depth of the number is set to 0.05 mm, the character width is 0.4 mm, the scanning laser power is set to 20 W, and the scanning speed is 180 mm / s to ensure that the coding is completed within 1.5 seconds after the mold is removed. All the coded numbers are registered and archived in real time in the MES system. Subsequently, combining the process data of the bottle body production defects in step S2 and the data of the positioning pin fracture in step S3, the glass bottle numbers within the same time window are cross-screened. If more than 90% of the bottles in a batch are associated with the same production defect parameter group or the positioning pin fracture data of the same work station, then this batch is marked as an abnormal batch. The number range of the abnormal batch generates a set of abnormal glass numbers through the data indexing module. For example, the number range "250514A1300–250514A1350", and this set of numbers is transmitted to the hot-end annealing equipment, packaging equipment, and final inspection and sorting equipment through the industrial Ethernet to identify and remove the glass bottles in this batch and perform digital quality traceability records. This numbered data is also synchronously uploaded to the enterprise cloud traceability platform for subsequent quality event backtracking.

[0017] Particularly importantly, step S41 includes the following steps: Step S41: Generate a unique glass number based on the lightweight glass bottle model; In this embodiment, the real-time data of the equipment during the glass bottle forming process is used as the basic data source. Based on the time stamp from the start to the end of the cooling of each lightweight glass bottle forming, combined with the mold cavity number, mold core number, positioning pin number, team number, production line number, and specific machine number, the field splicing and encryption processing are carried out by calling the hash coding algorithm (such as SHA-256). The field splicing order is fixedly set as "machine number + mold cavity number + mold core number + time stamp + positioning pin number", and the time stamp is accurate to the millisecond level to ensure no duplicate coding. In this way, a globally unique number for each glass bottle is generated, which is stored in the production database in hexadecimal form and laser-coded in the form of a two-dimensional code on the outer ring of the bottle bottom without interfering with the functional structure of the bottle body. The number generation process is integrated into the cold-end detection equipment control system, and the PLC controller and the industrial vision recognition module are synchronously triggered to enable the bottle body to obtain a unique number as soon as it flows out of the cooling process.

[0018] Step S42: Perform normalization processing on the process data of the bottle body production defects to obtain the normalized production defect process data; In this embodiment, surface defect data such as bottle body surface cracks, bubbles, weld lines, cold spots, and wrinkles are extracted from the hot-end infrared detection system and the cold-end vision detection system respectively. The data structure includes fields such as defect type, defect location (based on the bottle body coordinate axis), defect area, defect gray value range, corresponding cavity number, and timestamp. The defect area in the data is normalized using a value with the unit of mm². The maximum value is set to 50 mm², and the minimum value is 0.5 mm². The min-max normalization method is used for mapping. The defect location is standardized using the bottle body coordinate system (the Z-axis is the vertical height, and θ is the angular direction). The Z-axis takes the bottom of the bottle as the origin, and the maximum height value is set to 300 mm, and it is discretized at intervals of 10 mm; the θ angular direction is mapped by equally dividing 360° into 24 segments (each segment is 15°). The normalized defect process data is structured and sorted in chronological order and stored in the structured table of the database for subsequent time series analysis and processing.

[0019] Step S43: Mark the cavity number based on the positioning pin fracture data; In this embodiment, the positioning pin fracture data comes from the bottle mold maintenance workstation and the mold monitoring system. During the replacement of the bottle mold core or the maintenance of the cavity, the cavity identification instrument equipped on the maintenance engineering platform records the fracture or replacement event, and identifies the cavity number through the sensor. After each identification, the built-in data processor registers three fields: the cavity number, the fracture occurrence time, and the fracture location. The cavity number has a fixed structure. For example, "M-12" represents cavity No. 12, and the fracture timestamp is accurate to the second level. This data is uploaded to the quality control master station server through the Modbus TCP protocol and associated with the bottle number and the production line station number. The cavity number and the unique number of the glass bottle are calibrated through timestamp comparison and production sequence number, realizing the unique identification and binding of the fracture data and the specific bottle body data, and forming a data chain available for anomaly analysis.

[0020] Step S44: Conduct time correlation analysis based on the normalized production defect process data and the cavity number to obtain defect process - cavity number data; In this embodiment, the time correlation analysis is realized by constructing a time window algorithm. First, the timestamp field in the normalized production defect process data is extracted and grouped and clustered with the cavity number field. A fixed time window is defined as 30 minutes, and each group of data is divided into subsets according to the cavity number and the timestamp window. In each subset, if there are 3 or more consecutive numbered glass bottles with the same type of defect, and the proportion of the defect severity (with the defect area exceeding 20 mm² as the standard) exceeds 70%, then this window is defined as an abnormal time window. The cavity number corresponding to this time window and the unique glass number are synchronously extracted to form a defect process - cavity number data pair. The size of the time window can be obtained by fitting historical abnormal bottle data, and a multiple that can cover the average bottle forming cycle is selected. For example, when the average bottle forming cycle is 30 seconds, the window length is set to 60 times the cycle, that is, 30 minutes, to ensure that the data correlation coverage is complete.

[0021] Step S45: Identify the abnormal production batches according to the defect process - cavity number data to obtain the abnormal batch data; In this embodiment, an abnormal identification algorithm is constructed based on the defect process - cavity number data pair. Taking the cavity number as the clustering basis, calculate the proportion of defective glass bottles that appear in a certain time range (such as 2 hours) for each cavity number. If the proportion is greater than 10%, it is considered that there is an abnormality in this cavity during this time period. Further compare the positioning pin fracture record corresponding to this cavity number. If the occurrence time is within 1 hour before the time period when the defects appear concentratedly, it is considered that there is a causal relationship between the fracture and the defects. Sort all the bottles in this time period according to the unique number, and batch them as abnormal batches. The abnormal batch data generates a unique identifier in the structure of "date + machine number + cavity number + time period" and is stored in the abnormal batch database for subsequent mapping processing and production data docking.

[0022] Step S46: Perform abnormal glass number mapping on the abnormal batch data according to the unique glass number to obtain the abnormal glass number, and transmit it to the production equipment system to execute the glass quality traceability task.

[0023] In this embodiment, after the abnormal batch data is generated, the number field in the glass unique number database is called to be matched and mapped with the abnormal batch identifier. The matching method is as follows: according to the cavity number and time period recorded in the abnormal batch data, all number records within the corresponding time range are screened in the unique number table, and an exact match is made according to the time stamp range and the cavity number. The set of numbers that match successfully is the abnormal glass number. This set of numbers is uploaded to the MES (Manufacturing Execution System) platform through the industrial Ethernet and synchronized to the device control terminal through the OPC UA interface for alarm prompt, screening and rejection, and recall control. At the same time, the record of the abnormal glass number is uploaded to the traceability data platform to support subsequent quality audits and determination of liability attribution. The abnormal number transmission uses a JSON structure for data encapsulation, and the fields include "number ID", "defect type", "cavity number", "locating pin status", and "abnormal batch ID" to ensure the integrity and consistency of the data.

[0024] Preferably, step S1 is specifically as follows: Step S11: Obtain the glass bottle design data and extract the three-dimensional structure data of the glass bottle; In this embodiment, the process of obtaining the glass bottle design data is first based on the two-dimensional engineering drawings and the design specifications in the product structure database, and is processed by calling Pro / E or SolidWorks platforms in the standard CAD (Computer-Aided Design) system. The input data includes structural dimension parameters such as the bottle mouth diameter, bottle shoulder slope, bottle body height, bottle bottom thickness, overall weight, axial symmetry of the bottle body, and wall thickness distribution, in accordance with the regulations on bottle body thickness, radius of curvature, etc. in the national glass container design standard GB / T 4544-2008. The "reverse modeling tool" in SolidWorks is used to convert the original point cloud data into the three-dimensional structure data of the glass bottle. A three-dimensional laser scanner (such as FARO Edge ScanArm HD) is used to obtain the point cloud of the glass bottle sample, and the accuracy is set to ±0.01 mm. After the scanned data is registered, Geomagic DesignX is used for point cloud reconstruction operations. Through Boolean operations, boundary repair, and closed surface generation, a full-bottle three-dimensional geometric model is obtained. The final output three-dimensional structure data format is STL and IGES for subsequent geometric feature recognition and optimization calculations.

[0025] Step S12: Identify the key optimization areas according to the three-dimensional structure data of the glass bottle to obtain the bottle body area and the bottle shoulder area; In this embodiment, by scanning the normal vector distribution on the outer surface of the bottle, typical regions such as the bottle mouth, bottle shoulder, bottle body, and bottle bottom are divided according to the change trend. The bottle body region is defined as the part between the starting point of the shoulder and the upper end of the bottom contraction section in the middle of the bottle body. The bottle shoulder region is defined as the transition structure between the lower edge of the bottle mouth and the starting point of the bottle body. The local difference detection of the three-dimensional surface model is carried out by using the curvature analysis algorithm based on the second derivative. The critical curvature threshold is set to 0.002 mm⁻¹, and the bottle shoulder region with prominent curvature change is identified according to this threshold. Then, the middle continuous surface with an axial height in the range of 20 mm to 80 mm is extracted through spatial section analysis and determined as the bottle body region. Through the self-written curvature analysis script in MATLAB, the matrix solutions of the Gaussian curvature and the mean curvature are carried out for each triangular patch in combination with the STL geometric mesh file, and the extraction results are output in CSV format to provide basic data for subsequent structure optimization.

[0026] Step S13: Based on the bottle body region, perform curvature transition optimization design to obtain the optimized curvature of the bottle body; In this embodiment, for the curvature optimization operation of the bottle body region, the Bezier curve fitting method is adopted for continuous transition processing. In the specific operation, the longitudinal section within the bottle body region is selected, the section curve is extracted, and the fifth-order Bezier curve function is used for fitting. The uniform distribution of curvature is achieved by optimizing the positions of the control points. The curvature change rate index K′, that is, dK / ds, where s is the arc length of the curve, is used as the control variable of the optimization objective function, and the maximum value of the target K′ is set not to exceed 0.005 mm⁻². The ANSYS SpaceClaim Geometry Modeling tool is used for curve reconstruction. The curve reconstruction module (Curves → Fit Curve to Edge) is used to process 20 sections in the height direction of the bottle body. The movement range of the control points of each section curve is limited to ±0.3 mm, and the distance difference between the control points is ensured not to exceed 0.8 mm to avoid generating sharp or concave regions. By comparing the mean square error (MSE) of the original curvature and the fitted curvature, it is ensured that the fitting error is lower than 0.02 mm, and each optimized section curve is projected back into the three-dimensional model to generate the updated bottle body surface.

[0027] Step S14: Based on the bottle shoulder region, perform wall thickness reduction design to obtain wall thickness reduction data; In this embodiment, for the wall thickness reduction design of the bottle shoulder region, a method based on the modification of the finite element mesh thickness parameters is adopted. In the analysis, first, the bottle shoulder region is cut out from the overall model and imported into HyperMesh for local mesh division. The target element size of the division is set to a tetrahedral mesh with a size of 1 mm, and the total number of meshes is controlled within 50,000. The initial wall thickness is set to 3 mm. According to the compressive strength σ_c of the glass material (set to 45 MPa) and the safety factor (set to 2), using the relationship between thickness and stress σ = F / A, where A is the stress-bearing area, the limit wall thickness value is deduced to be approximately 1.6 mm. In actual operation, the target value of wall thickness reduction is set to 2.0 mm, and the reduction rate is 33.3%. To avoid stress concentration, a transition region with a length of 5 mm is set in the connection region between the bottle shoulder and the bottle body, and a linear thickness transition strategy is adopted, transitioning from the original 3 mm to 2.0 mm. After updating the wall thickness, Altair OptiStruct is used to perform a single-point compression simulation on the stress-bearing area of the bottle shoulder to verify whether the maximum equivalent stress in the reduced-thickness region is still lower than σ_c. The wall thickness reduction data is exported as a TXT format through the mesh thickness attribute and written into the design file for subsequent modeling use.

[0028] Especially importantly, step S14 includes the following steps: Step S141: Identify the bottle shoulder structure contour based on the bottle shoulder region; In this embodiment, the three-dimensional point cloud data of the glass bottle shoulder is obtained by a high-precision three-dimensional laser scanner. The resolution of the scanner should be set to not less than 0.05 mm to ensure accurate capture of every detail in the bottle shoulder region. During scanning, the bottle shoulder region needs to be covered at multiple angles of 360 degrees to ensure complete three-dimensional data acquisition. After scanning, data processing software is used for noise reduction and screening of the point cloud data to remove irrelevant stray data. After processing the point cloud data, the main structure contour of the bottle shoulder region is identified using a boundary extraction algorithm (such as the RANSAC algorithm), and the edge lines and transition surfaces of the bottle shoulder region are determined. Through these extracted boundary lines, an accurate bottle shoulder geometric model can be constructed, which will serve as the basis for subsequent wall thickness measurement and optimization adjustment.

[0029] Step S142: Perform thickness measurement based on the bottle shoulder structure contour to obtain the bottle shoulder thickness data; In this embodiment, after the contour of the bottle shoulder region is recognized, the thickness of the bottle shoulder region is measured by an automated wall thickness measurement system. This system uses a lidar sensor for precise measurement to ensure that the error at each measurement point does not exceed 0.1 mm. During measurement, a circumferential measurement strategy is adopted to cover the entire geometric surface of the bottle shoulder region. At each circumferential measurement position, the lidar sensor makes a precise scan perpendicular to the bottle shoulder of the bottle shoulder surface to obtain the wall thickness data from the surface to the inside. The obtained thickness data is recorded according to the spatial position of the measurement point (i.e., the distance from the bottom of the bottle) and converted into a two-dimensional or three-dimensional data format. These thickness data provide the detailed thickness distribution of each region of the bottle shoulder, providing a basis for subsequent stress tests and wall thickness optimization.

[0030] Step S143: According to the bottle shoulder thickness data, load a standard simulated internal pressure through an industrial stress test device and connect to a strain bridge array to monitor the real-time stress response, so as to obtain the bottle shoulder stress data; In this embodiment, a standard simulated internal pressure is loaded through an industrial stress test device. The device should be able to simulate a working environment with an internal pressure of 0.8 MPa in the bottle. This pressure value is set according to the design standard of the glass bottle, usually a typical value of the pressure borne by the bottle body. The pressure sensor of the stress test device needs to have a measurement ability with an accuracy not lower than 0.1 MPa to ensure that the loaded pressure meets the design requirements. After loading the internal pressure, the stress response of the bottle shoulder region is monitored in real time through a strain bridge array. The strain bridge array consists of multiple strain sensors, which are evenly distributed in the bottle shoulder region to monitor the stress changes at different positions. Each strain bridge array is connected to a data acquisition system, and the data acquisition system needs to have a data acquisition frequency of at least 100 Hz to ensure that the stress changes can be recorded in time. Through the processing and analysis of the strain signals, the stress distribution data of the bottle shoulder region can be obtained. The stress data can not only reflect the load-bearing capacity of the bottle shoulder region but also reveal the stress concentration problems in some regions of the bottle shoulder, providing an important basis for subsequent wall thickness adjustment.

[0031] Step S144: Based on the bottle shoulder stress data, perform point-by-point wall thickness adjustment, limit the minimum wall thickness to not less than 2.1 mm, and obtain the wall thickness thinning data.

[0032] In this embodiment, based on the stress data analysis, the stress concentration conditions in different regions of the bottle shoulder are determined to identify the key positions of stress concentration. For example, relatively high stress values exceeding the safety threshold of the glass material will occur in certain regions of the bottle shoulder under the standard simulated internal pressure. For these regions, the wall thickness is adjusted point by point. During the adjustment process, based on the design requirements, it is ensured that the minimum wall thickness at each adjustment point is not less than 2.1 mm. If the thickness of some regions is insufficient, it can be adjusted through the thickness increase and decrease algorithm, thickening the too-thin regions and thinning the regions that are already thick enough, so that the stress distribution of the bottle shoulder under internal pressure is more uniform. The specific wall thickness change value at each adjustment point is calculated by the difference from the original thickness data to obtain the wall thickness thinning data for each point. The finally obtained wall thickness thinning data will be used to guide the processing or reprocessing of the bottle shoulder mold to ensure that the structure of the bottle shoulder not only meets the lightweight requirements but also ensures the strength and safety of the glass bottle.

[0033] Step S15: Integrate the optimized curvature of the bottle body and the wall thickness thinning data, and construct a lightweight glass bottle model; In this embodiment, the integration process is implemented based on the 3D reconstruction module of the CATIA platform. First, the optimized curvature of the bottle body obtained in step S13 is imported as an independent geometric surface object, and the "Multi-section Surface" tool of CATIA is used to construct a new bottle body surface by setting the continuity between sections to G2 (curvature continuous). Subsequently, the shoulder thinning mesh data obtained in step S14 is imported as the thickness reference, and the original shoulder surface is offset equidistantly through the Surface Offset module to construct the updated solid with the thinned wall thickness. Then, the Join module is used to connect the optimized bottle body surface, the thinned shoulder surface, and the bottle bottom and bottle mouth regions to form a closed body. The Part Design module is used for solidification operation to generate the complete 3D structure of the lightweight glass bottle. The final output format of the model file is the STEP file, which contains complete entity information and wall thickness annotation information for structural strength simulation analysis.

[0034] Step S16: Detect the structural strength of the bottle body based on the lightweight glass bottle model.

[0035] In this embodiment, the structural strength detection is carried out by finite element analysis using ANSYS Workbench. First, the lightweight glass bottle model is imported into ANSYS for mesh generation. The tetrahedral structural element is adopted, and the element size is 0.8 mm. Approximately 180,000 elements are generated for the whole bottle body. The material properties are set as silicate glass, with an elastic modulus of 70 GPa, a Poisson's ratio of 0.23, and a density of 2500 kg / m³. The loading conditions are set as follows: a 5 kg pressure is applied vertically to the bottle body to simulate the stacking load, and a fixed constraint is set at the bottom of the bottle; a 20 N axial impact force is applied to the bottle shoulder area to simulate the bottle cap sealing force; a 10 N lateral pressure is applied to the side wall of the bottle body to simulate the lateral impact during handling. The calculation targets include the maximum principal stress, Von Mises equivalent stress, and deformation amount. The limit judgment criteria are set respectively: the maximum principal stress shall not exceed 45 MPa, the equivalent stress shall not exceed 40 MPa, and the total deformation amount shall not exceed 0.3 mm. After the calculation is completed, the stress distribution diagram (in JPG format) and the stress numerical file (in CSV format) are exported and recorded in the quality traceability system for subsequent defect correlation analysis. All simulation data are archived synchronously with the timestamp and the unique design number.

[0036] Preferably, step S16 is specifically as follows: Step S161: Import the lightweight glass bottle model into the simulation software; In this embodiment, the three-dimensional structure file of the constructed lightweight glass bottle is exported in STEP format, and the "Static Structural" module in ANSYS Workbench 2021 R2 is used for the import operation of structural strength analysis. In the ANSYS environment, select the "Engineering Data" module to load the new material property library and configure the simulation engineering file. Subsequently, in the "Geometry" module, load the glass bottle structure file through the "Import Geometry" command. The default coordinate system is used for three-dimensional position recognition during the import process to ensure that there are no geometric anomalies such as topological gaps and discontinuous surfaces in the model. At the same time, the "Share Topology" tool is used to merge the contact areas to avoid discontinuous boundary conditions during the simulation analysis. After the import, use the "Mesh" module to perform the model mesh generation operation. The mesh type is set as tetrahedral elements (Tetrahedral Element), and the initial mesh size is set to 1.0 mm. Local refinement is performed according to the high-stress areas such as the bottleneck and the bottom of the bottle, controlling the minimum mesh size to be not less than 0.2 mm and the maximum size to be not more than 3 mm. The mesh quality standard is Skewness < 0.9 to ensure the calculation accuracy.

[0037] Step S162: Set the wall thickness range of the bottle body to 0.3 mm - 3 mm, the neck thickness range to 0.2 mm - 1.5 mm, and the bottom thickness range to 0.5 mm - 4 mm in the simulation software; In this embodiment, after importing and completing the grid division of the glass bottle, enter the ANSYS "Model" module, and use the "Named Selections" function to group different regions of the bottle body structure: named Body (bottle body), Neck (bottle neck), and Bottom (bottle bottom) respectively. Use the "Thickness" command to set the initial wall thickness parameters of the three groups of regions. The wall thickness range of the bottle body is controlled to be 0.3 mm - 3 mm, and the setting strategy is to set the wall thickness change according to each 10 mm layer in the axial direction of the bottle body height, and the distribution function Wall_Thickness_Body(z) = 0.3 + 2.7*sin(πz / H) is adopted, where z is the axial height and H is the total height of the bottle body; the neck thickness setting range is 0.2 mm - 1.5 mm, and a conical gradient structure is adopted, and the thickness changes linearly along the axis according to the function Neck_Thickness(z) = 0.2 + 1.3(z / L), where L is the height of the bottle neck; the bottom area is set with a thickness range of 0.5 mm - 4 mm, and the thickness transition is carried out according to the fillet radius r connected to the bottle body, and the thickness function is Bottom_Thickness(r) = 0.5 + 3.5(1 - exp(-r / 5)), where r is the radial distance from the center of the bottle bottom to the edge (unit: mm), which is used to control the thickness change of the bottom thickening area to adapt to the stress difference.

[0038] Step S163: Set the external pressure range to 0.5 MPa - 2 MPa, the internal pressure range to 0.3 MPa - 1.5 MPa, and the transportation impact force range to 10 N - 200 N in the simulation software; In this embodiment, when configuring the load conditions in the "Static Structural" module of ANSYS, various external environmental stresses are applied to the model surface respectively. The external pressure range is set to 0.5 MPa - 2 MPa, the loading mode is constant surface pressure, which acts on the entire outer surface of the bottle. It is set in stages, and four groups of working conditions are set to 0.5 MPa, 1.0 MPa, 1.5 MPa, and 2.0 MPa respectively; the internal pressure range is 0.3 MPa - 1.5 MPa, which is applied to the inner surface of the bottle and the direction is outward. Four groups of pressure values are also set to establish the pressure influence curve; the transportation impact force is set as an instantaneous concentrated force (Point Load), which acts on different key areas of the bottle, including the bottle shoulder, the bottom edge of the bottle, and the waist of the bottle. The impact force at the corresponding points is set to four levels: 10 N, 50 N, 100 N, and 200 N, and the action time is set as a step impact within 0.1 second. The coupled solution simulation is carried out through the "Transient Structural" module, and the impact directions include the axial direction, the radial direction, and the vertical direction. A total of 16 groups of simulation working conditions are set for the combination form of the load and the boundary conditions to cover the pressure and impact conditions encountered in the typical transportation and storage environments.

[0039] Step S164: Set the elastic modulus range of the glass material to 60 GPa - 75 GPa and the Poisson's ratio range to 0.22 - 0.26 in the simulation software. In this embodiment, the glass material properties are configured in the "Engineering Data" module, and silicate glass is selected as the base material. The setting range of the material elastic modulus (Young’s Modulus) is from 60 GPa to 75 GPa, and four values are set to 60 GPa, 65 GPa, 70 GPa, and 75 GPa, and the input unit is Pa; the Poisson's ratio is set in the range of 0.22 to 0.26, and the corresponding values are three levels: 0.22, 0.24, and 0.26; the fracture toughness is set as a non-linear material parameter, and three groups of values of 0.3, 0.5, and 0.8 MPa·m^1 / 2 are respectively input into the ANSYS material non-linear fracture library. The fracture behavior function needs to be loaded through the User Defined Material model, and the LEFM (Linear Elastic Fracture Mechanics) criterion is adopted to analyze the fracture stress distribution under various pressure and impact loading conditions. Each parameter is derived from the national glass product quality standard GB / T2828 and the ASTM C158 - 02 standard, and is calibrated by comparing with the existing physical property test data of the glass bottle to ensure that the parameter settings in the material constitutive model are consistent with the physical properties of the glass.

[0040] Step S165: Run the structural strength analysis program in the simulation software and record the structural strength of the bottle body.

[0041] In this embodiment, after setting the boundary conditions, load conditions, and material properties, select "Equivalent Stress (von-Mises)" and "Maximum Principal Stress" as output variables in the ANSYS "Solution" module, and run the simulation task through the "Solve" command. The simulation results record the location of the structural stress concentration area, the maximum stress value, and its corresponding coordinate points. For different combinations of thickness, pressure, and impact conditions, the output results are exported in tabular form. The stress threshold is controlled between 30 MPa and 90 MPa according to the glass fracture limit strength. If the equivalent stress in a local area exceeds the set material fracture stress value (such as 85 MPa), it is determined that the design does not meet the structural integrity requirements. The stress distribution data output by each simulation is stored as a.csv file according to the node number and spatial coordinates. The data fields include node ID, X / Y / Z coordinates, von-Mises stress value, principal stress value, and its direction angle, etc. All results are automatically classified and summarized through the Post Processing module for subsequent bottle body quality assessment and traceability comparison analysis.

[0042] Preferably, step S2 is specifically as follows: Step S21: Extract the bottle body deformation characteristics and the bottle body pressure resistance characteristics based on the structural strength of the bottle body to obtain the bottle body deformation data and the bottle body pressure resistance data; In this embodiment, after completing the structural strength simulation analysis, perform the extraction operations of the bottle body structural deformation characteristics and pressure resistance characteristics based on the recorded stress distribution and displacement field change data of the bottle body under different working conditions. First, under the extreme load condition of setting the external pressure to 2 MPa, the internal pressure to 1.5 MPa, and the impact force to 200 N, export the total displacement values of each unit node through the finite element simulation results, and use the three-dimensional point cloud interpolation method to extract the displacement mutation interval. The identification of the maximum deformation area of the bottle body is achieved by setting the displacement gradient threshold Δd = 0.4 mm. Any area where the gradient change exceeds this threshold is defined as a high-risk deformation area, and its spatial coordinates are recorded. During the extraction of the bottle body pressure resistance characteristics, the internal pressure of the bottle body is linearly increased from 0.3 MPa to 1.5 MPa, and the equivalent stress distribution is calculated every time it increases by 0.1 MPa. If the local maximum equivalent stress σ_eq exceeds the compressive limit of the glass material σ_c = 120 MPa, the corresponding pressure value is recorded as the instability pressure resistance value. In this way, the maximum stable pressure resistance value of the bottle body structure is recorded to obtain the complete bottle body deformation data (including node coordinates, total deformation value, stress concentration area) and the bottle body pressure resistance data (including critical internal pressure value, stress distribution matrix under each pressure).

[0043] Step S22: Based on the bottle body deformation data, perform bottle body crack detection to obtain bottle body crack data; In this embodiment, based on the bottle body deformation data obtained in step S21, the digital image correlation method (DIC) and the three-dimensional structured light scanning results are used for crack detection. Based on the displacement data extracted in the stress concentration area, three-dimensional microscopic observation is performed on the high-gradient area. A confocal microscopy system with an accuracy of 5 μm is used to scan the surface of the bottle body layer by layer to obtain the crack depth and crack width. The recognition threshold for crack detection is set as: crack depth d_crack ≥ 0.1 mm, crack width w_crack ≥ 0.02 mm. If there is a continuous gray-scale mutation greater than 50 units in the extracted area of the three-dimensional scanned image and the shape extends linearly for more than 1 mm, then this area is judged to have a structural crack. The crack characteristics are stored as crack data in the form of parameters such as structural coordinates, crack length, crack depth, and crack direction. Combining with the initial simulation pressure condition, the crack propagation trend can be further marked to generate a bottle body crack data set with time node identifiers.

[0044] Step S23: Based on the bottle body pressure resistance data, perform transportation simulation and identify the peeling of the surface layer glass during the transportation of the bottle body to obtain the surface layer glass peeling data; In this embodiment, the bottle body pressure resistance data obtained in step S21 is imported into the transportation simulation environment. According to the transportation packaging drop test standard GB / T4857.5-92, the drop height is set to 1.2 meters, and the glass bottle bottom, bottle side, and bottle shoulder are impacted by free fall in three directions, and the surface stress distribution and damage characteristics after the collision are recorded respectively. A high-speed camera (frame rate greater than 5000 fps) is used to record the glass surface layer rupture process at the moment of impact, and an image processing algorithm is used to extract the peeling area. Infrared thermography combined with the surface refractive index measurement method is used for peeling detection. If the change value of the glass surface refractive index Δn > 0.05 or the size of the surface peeling particles is greater than 100 μm, then this area is defined as the glass peeling area. After each simulation, the peeling occurrence coordinates, particle size distribution, glass thickness change, and stress change are recorded together to generate surface layer glass peeling data including the peeling position, area, and glass strength reduction.

[0045] Step S24: Integrate the bottle body crack data and the surface layer glass peeling data to obtain the bottle body surface defect data; In this embodiment, the crack data obtained in step S22 and the surface spalling data obtained in step S23 are subjected to data integration operations. Through spatial coincidence analysis, a unified three-dimensional coordinate system is established to spatially calibrate all crack and spalling positions, and the damage areas with overlapping regions and adjacent boundary distances less than 1 mm are extracted as the core set of the bottle body surface defect data. Define the coding rules for structural defect blocks, and construct a data table according to information such as defect type (crack CR, spalling SP), position coordinates, size, area, direction, etc. Calculate the volume loss rate of the defect area through three-dimensional point cloud comparison. If it exceeds 0.5 cm³, it is marked as a serious defect point. Finally, the bottle body surface defect data is exported in CSV format, and the fields include defect type, position coordinates (x, y, z), defect size (length, width, depth), lost volume, crack direction angle, maximum diameter of spalling particles, etc.

[0046] Step S25: Based on the bottle body surface defect data, defect traceability is carried out to obtain the bottle body production defect process data.

[0047] In this embodiment, after obtaining the bottle body surface defect data, defect traceability analysis is carried out based on the production process parameter records of the glass bottle. Each defect sample is multi-dimensionally matched with parameter data such as mold temperature (set range 1100°C - 1400°C, acquisition interval is 1 second), preform cooling time (set range 2s - 5s), spraying agent concentration (measurement range 0% - 5%), annealing temperature curve (set at 580°C to 620°C), etc. The decision tree classification algorithm is used to construct association rules. If cracks are concentrated in the bottle shoulder area, and the corresponding mold cooling time is less than 2.5 s and the annealing curve deviates by ±10°C, it is determined that the thermal stress concentration cracks are caused by insufficient annealing. If the spalling area is concentrated in the concave part of the bottle bottom, and the corresponding preform thickness is less than 0.8 mm and the spraying agent coverage rate is less than 90%, it is attributed to the spalling caused by uneven cooling stress accumulation due to uneven spraying. Finally, the bottle body production defect process data is output, and the fields include defect code, defect type, traceability process parameters (temperature, time, thickness, etc.), abnormal interval time point, matching probability score, etc., as the historical tracking basis for quality control.

[0048] Preferably, step S22 is specifically: Step S221: Based on the bottle body deformation data, bottle body rotation control is carried out to obtain bottle body rotation data; In this embodiment, the bottle body deformation data extracted in step S21 is used as the basic data and input into the control system. The six-degree-of-freedom rotation execution platform is used to control the bottle body to perform stable rotation operations along the vertical axis and the horizontal axis. The bottle body is fixed in the pneumatic fixture at the center of the rotation platform, and the clamping pressure is set to 0.5 MPa to ensure that the bottle body does not undergo displacement and sliding during rotation. The rotation angular velocity is set to 15° / s, and the rotation angle range is from 0° to 360°. The angular position information and the deformation vector direction are collected every 5°. An angle encoder (resolution 0.01°) and a laser displacement sensor (measurement accuracy 1μm) are used in conjunction to collect the deformation distribution data of the bottle body at any angle during rotation, forming a complete three-dimensional rotation deformation vector field. Finally, the bottle body rotation data is output, including parameters such as angle coordinates, rotation deformation vector values, axial torque distribution, and changes in moment of inertia.

[0049] Step S222: Calculate the stress gradient based on the bottle body rotation data, and extract the high stress gradient values based on the stress gradient; In this embodiment, after the bottle body rotation data obtained in step S221 is imported into the stress calculation system, first, the deformation stress distribution is calculated according to the displacement data at each angle on the rotation path. At this time, the elastic modulus of the glass material is 72 GPa and the Poisson's ratio is 0.22, and these material parameters are used to calculate the stress distribution of the glass bottle body at each rotation angle according to the linear elastic body theory. During the calculation, based on the displacement difference method, the system first obtains the displacement amount corresponding to each rotation angle and calculates the local strain at each angle by the difference method. According to the relationship between strain and stress, the stress value at each angle point is calculated using the basic formula in elasticity mechanics. Next, the system calculates the stress difference between each angle and the adjacent angle. First, the stress values of the adjacent angles are extracted, the difference between them is calculated, and then divided by the angle difference (for example, each rotation is 5°) to obtain the stress gradient at each point. The stress gradient is an index describing the change of stress distribution, and its magnitude represents the stress change rate under the change of unit angle. After the calculation is completed, the system sets a threshold (for example, 3 MPa / °). If the stress gradient at a certain angle point is greater than this threshold, then this point will be marked as a high stress gradient point. Finally, the system records all the stress gradient values higher than the set threshold and their corresponding rotation angle positions, generates a high stress gradient data set, and exports and saves this data set. This data set contains the specific values and positions of each high stress gradient point, providing data support for subsequent crack identification and high stress area positioning.

[0050] Step S223: Identify the high deformation area of the bottle body according to the high stress gradient value; In this embodiment, region recognition is performed on the high stress gradient value data in step S222. The spatial clustering method DBSCAN is used, with the neighborhood radius ε set to 3 mm and the minimum number of samples MinPts set to 6, which is set according to the empirical value of the discrete point distribution density on the bottle surface. The input data includes the spatial coordinates (X, Y, Z) of the points and their angle θ labels. During the clustering process, multiple high-density region clusters are identified, and each cluster contains several continuous stress gradient mutation points. The region surrounded by each clustering cluster is fitted into an ellipsoid, and the center coordinates C(x, y, z), the lengths of the major and minor axes (a, b, c), and the number of clustering points N are recorded. The screening conditions are N ≥ 10 and a / b ≥ 1.5 as the high deformation region criteria. The output data includes the spatial boundary point set of each high deformation region, the deformation field (obtained by projecting the deformation vector field), the region number, and the rotation angle θ range, which are used to guide the subsequent X-ray scanning path planning.

[0051] Step S224: Perform X-ray circular irradiation scanning on the high deformation region of the bottle to obtain the X-ray image of the bottle; In this embodiment, a high-resolution X-ray scanning device (such as GE phoenix v|tome|x m) is used, with a focal spot size of 2 μm, the working voltage set to 100 kV, the current set to 250 μA, and the exposure time of 50 ms. The bottle is fixed on a 360° rotating platform, and the rotation accuracy is set to 0.05°. The scanning path is limited to the angle range of ±10° of the high deformation region in step S223 to ensure obtaining complete crack information. The X-ray detector uses a 2000×2000 pixel flat panel detector, and the single pixel size is 50 μm. One frame of image is captured every 1° of rotation during image acquisition, and the image sequence number corresponds to the angle θ one by one. The generated image file is named "Region_XX_Angle_XXX.tiff". A total of 21 images are collected to cover each high stress region. All X-ray images are imported into the image management system for archiving, including metadata such as the image matrix, exposure parameters, scanning angle, bottle number, and region number.

[0052] Step S225: Extract the significant boundaries based on the X-ray image of the bottle to obtain the significant boundary data; In this embodiment, the X-ray images collected in step S224 are processed frame by frame using the OpenCV image processing library. First, Gaussian filtering (kernel size 5×5, σ = 1.0) is performed on the original image to remove high-frequency noise. Subsequently, the Canny edge detection algorithm is used to extract the boundaries, with the low threshold T1 = 50 and the high threshold T2 = 100 set respectively, based on the dynamic range of the image grayscale and the crack edge gradient. After extracting the boundaries, contour tracking is performed. The findContours function in OpenCV is used to number the closed boundaries and extract the point set of each boundary (the number of points is not less than 100). Curvature analysis is performed on the boundary point set, and the boundary with a curvature change < 5% is defined as the "significant boundary". The output significant boundary data structure includes: boundary number, contour point coordinate sequence, boundary length, closedness flag (Boolean value), curvature change rate, image number, and angle θ label.

[0053] Step S226: Calculate the aspect ratio based on the significant boundary data to obtain the aspect ratio. In this embodiment, for each significant boundary contour point set, the minimum bounding rectangle fitting is performed (using the minAreaRect function). The major axis L and minor axis W of the rectangle are represented in physical units (mm). The pixel points (px) are converted to the actual size using the image calibration parameters, and the calibration coefficient is 1px = 0.05mm. The aspect ratio AR is calculated as L / W. If the number of boundary points ≥ 150 and the contour is approximately a straight-line structure (rectangle angle difference < 10°), then this area is determined to be a crack area. All boundary data record fields include: boundary number, major axis L (mm) of the bounding rectangle, minor axis W (mm), aspect ratio AR (unitless), position of the contour center point, and rotation angle θ.

[0054] Step S227: Identify the crack contour based on the aspect ratio. In this embodiment, based on the aspect ratio data calculated in step S226, the boundaries with AR ≥ 5 are identified for crack contour. Further, the angle φ between the main direction vector of the boundary and the main direction of the local stress of the bottle body is calculated. The projection stress direction algorithm is used to obtain the main stress direction. If the angle φ ≤ 15°, it is calibrated as the crack path. The B-spline interpolation method is used to fit the boundary point sequence, and the fitting step size is set to 0.2mm. The output crack contour is a continuous path line with a direction vector, recording the starting point, ending point, path length, crack number, scanning angle θ, and fitting residual RMSE.

[0055] Step S228: Calculate the crack defect density based on the crack contour to obtain the bottle body crack data.

[0056] In this embodiment, the X-ray image area of the bottle body is projected onto a two-dimensional plane and divided into grid cells with an equal spacing of 25 mm × 25 mm, generating M × N grid cells. Within each grid cell, by analyzing the crack paths in the X-ray image, the total length (unit: mm) of the crack segments and the number of cracks in each grid cell are counted. The path length of each crack within the corresponding grid cell is recorded, and the crack defect density is calculated accordingly. The specific calculation formula for the crack defect density ρ is as follows: for each grid cell (i, j), first calculate the total path length sum L_k of all cracks within the cell, and then according to the formula ρ_i,j = Σ(L_k) / 625, where L_k is the path length of each crack, in mm. Dividing by 625 is because the area of each grid cell is 25 mm × 25 mm, that is, 625 mm², so the calculated density unit is mm / mm². Through this calculation, the crack defect density value within each grid cell is obtained. Finally, the crack defect densities of all grid cells are output in matrix form. The grid cells with density values greater than or equal to 0.4 are marked in red to indicate the high crack risk areas. The final results are output as a crack defect density map (CSV format matrix), a corresponding crack number index table, and a crack structure file (JSON format) corresponding to the bottle body number for subsequent traceability comparison and defect statistical analysis. During the processing, all crack paths and density values are accurately calculated based on image processing techniques and geometric analysis methods to ensure the accuracy and reliability of the data.

[0057] Preferably, step S23 is specifically as follows: Step S231: Screen low-pressure-resistant bottle bodies based on the bottle body pressure resistance data and conduct transportation simulation to obtain low-pressure-resistant bottle body transportation data; In this embodiment, by collecting the bottle body pressure resistance data, low-pressure-resistant bottle bodies are screened. During the screening process, first set the pressure resistance threshold to 30 MPa, and compare the pressure resistance data of the bottle bodies one by one. If the pressure resistance value of a bottle body is lower than this threshold, it is regarded as a low-pressure-resistant bottle body. Next, for the screened low-pressure-resistant bottle bodies, transportation simulation is carried out. The transportation simulation uses a dedicated transportation simulation software, and factors such as the impact force, vibration, and external pressure encountered by the bottle body during transportation are considered in the simulation environment. By setting the maximum impact force during transportation to 200 N, the maximum vibration frequency to 50 Hz, and conducting multiple simulations on the bottle body, record the external force acting on and the deformation of the low-pressure-resistant bottle body in each simulation, and finally obtain the low-pressure-resistant bottle body transportation data.

[0058] Step S232: Conduct a force analysis based on the low-pressure-resistant bottle body transportation data to obtain the bottle body transportation force data; In this embodiment, through the finite element analysis method and combined with the transportation simulation data, the stress conditions of the low-pressure-resistant bottle body are carefully calculated. First, the geometric shape and material properties of the bottle body are set (for example, the elastic modulus is 72 GPa and the Poisson's ratio is 0.22). In the simulation environment, the external forces are distributed to each part of the bottle body, and the stress and strain of each part are calculated. Based on the stress analysis results, a stress distribution diagram is drawn, and the area with the maximum stress borne by the bottle body during transportation is recorded. By comparing the stress values of different parts, the stress data of the bottle body are obtained, including the maximum stress value of each part, the stress concentration situation, and the area where rupture occurs.

[0059] Step S233: Identify the stress concentration area based on the stress data of the bottle body during transportation and record it as the peeling risk area; In this embodiment, stress data during transportation are collected in real time at multiple positions on the glass bottle body through sensors (such as strain gauges, stress sensors, etc.). After the obtained data are processed, a stress distribution diagram of the bottle body can be obtained, showing the stress values at different positions. Next, analysis is carried out according to the set stress concentration threshold, and the threshold is set to 150 MPa. When the stress value at a certain position of the bottle body is greater than or equal to 150 MPa, it indicates that the stress on this part is too large, resulting in damage or peeling risk of the bottle body. By comparing and analyzing the stress distribution diagram of the entire bottle body, the areas where the stress values exceed this threshold are identified and marked as peeling risk areas. These areas are usually located in the parts of the bottle body with irregular shapes and large forces, such as the bottle mouth, bottom, and sharp corners. After determining the peeling risk areas, subsequent inspections and monitoring are carried out, especially strengthening the inspection of these high-risk areas to ensure that the glass bottle does not break or peel during transportation, thereby improving the transportation safety and the overall quality control level of the glass bottle.

[0060] Step S234: Obtain a multi-spectral image of the bottle body according to the peeling risk area to get the multi-spectral image of the bottle body; In this embodiment, first, a multispectral camera capable of providing reflectance data in multiple bands is selected. Such a camera can capture images within a specific wavelength range. The spectral band range is set from 400 nm to 1000 nm to cover the spectral region from ultraviolet to near-infrared, so as to obtain rich spectral information about the bottle surface. During the operation, the bottle is placed in a stable environment to ensure that there is no debris interference on its surface. An appropriate light source is used for illumination to avoid the overexposure problem caused by strong direct illumination. During the shooting process, the multispectral camera captures the reflectance data of the bottle surface at different wavelengths, thereby obtaining images in multiple bands. These images respectively represent the reflectance information at different wavelengths and can provide detailed data about the material and state of the bottle surface. After the images are acquired, data preprocessing is performed. First, the image interference caused by environmental or device noise is eliminated through a denoising algorithm. Common denoising methods include median filtering, mean filtering, etc. These methods help to remove the random noise in the image and make the image clearer. Secondly, illumination equalization processing is performed to ensure that the brightness distribution in the images of each band is uniform and avoid the image quality difference caused by uneven illumination conditions. Common techniques include histogram equalization, etc. After preprocessing, high-resolution images of each band are obtained, which contain the reflectance information of the bottle surface. Through these images, it is possible to further analyze whether there are defects, stress concentration areas, etc. on the bottle surface, providing basic data for subsequent risk assessment and quality analysis.

[0061] Step S235: Extract the spectral feature vectors of pixel points based on the multispectral image of the bottle, and construct a pixel-level spectral feature matrix according to the spectral feature vectors of pixel points; In this embodiment, through pixel-level processing of the multi-spectral image of the bottle body, the spectral feature vector of each pixel is extracted. The spectral feature vector of each pixel contains the reflectance values of the pixel at different bands. During specific operations, for each pixel, first read the reflectance data corresponding to it in each set band (such as multiple bands in the range of 400nm to 1000nm) from the multi-spectral image. Assuming that the image contains reflectance information of multiple bands, by extracting the reflectance values of each pixel at these bands, a multi-dimensional vector can be obtained, representing the spectral characteristics of the pixel. Next, convert the spectral data of each pixel in the image into a vector, and these vectors together form a data set. For each row of pixels in the image, the extracted spectral feature vector is the spectral feature data of the pixel. Organize all the extracted pixel spectral feature vectors to construct a pixel-level spectral feature matrix. Each row of this matrix represents a pixel point in the image, and each column corresponds to the reflectance value of a band. Each element in the matrix represents the reflectance value of the pixel at a specific band. This matrix will contain the spectral information of the entire image, can reflect the reflection characteristics of different regions and different types, and provide a data basis for subsequent analysis. To further process and analyze this matrix, clustering analysis methods such as K-means or hierarchical clustering can be used to classify similar spectral features, so as to identify different regions or feature types. This matrix is not only used for subsequent spectral anomaly detection, but also can provide necessary data support for defect analysis. Through the processing of this matrix, different material or structural problems existing on the surface of the bottle body can be detected, thus providing a scientific basis for subsequent quality control and defect diagnosis.

[0062] Step S236: Identify the spectral anomaly bands based on the pixel-level spectral feature matrix, and calculate the band mutation coefficient; In this embodiment, statistical analysis is first performed on each band in the pixel-level spectral feature matrix. The specific operation is as follows: for each band, calculate the mean and standard deviation of the reflectance values of all its pixel points. The mean reflects the average reflectance of the band, while the standard deviation reflects the fluctuation range of the reflectance of the band. Through these statistics, regions with large changes in reflectance values in different bands can be identified. To identify spectral anomaly bands, a standard deviation threshold is set (for example, a band with a standard deviation greater than 0.1). When the standard deviation of a certain band exceeds this threshold, the band is considered to have a large change in reflectance and is thus regarded as an anomaly band. Next, for each anomaly band, calculate its mutation coefficient. The calculation of the mutation coefficient is achieved by comparing the change degree of reflectance between each pixel point in the band. Specifically, the mutation coefficient can be calculated by comparing the reflectance difference between adjacent pixel points, calculating the ratio of this difference, and thus quantifying the severity of the reflectance change. If the reflectance difference is large, the mutation coefficient is also large, indicating that the reflectance has mutated in this region, which is a manifestation of surface defects. Through the calculation of the mutation coefficient, spectral anomaly bands can be identified, and it can be further determined whether these bands correspond to the defective regions on the bottle body surface, providing a basis for subsequent quality analysis and defect location.

[0063] Step S237: Calculate the reflectance gradient based on the multi-spectral image of the bottle body; determine the peeling of the surface glass according to the band mutation coefficient and the reflectance gradient, and obtain the surface glass peeling data.

[0064] In this embodiment, first calculate the reflectance gradient of each pixel point, that is, the change rate of the reflectance value between adjacent pixel points. Specifically, for each pair of adjacent pixels in the image, calculate the difference in reflectance at the same band and normalize it with the distance between these two pixels to obtain the reflectance gradient. Regions with a large reflectance gradient usually indicate that there are large changes in reflectance in this region, which is a manifestation of uneven surface structure or the presence of defects. Then, by combining the mutation coefficient obtained in the previous steps, further screen out regions with large reflectance changes and meeting the peeling risk characteristics. Specifically, when the reflectance gradient of a certain region is large and its mutation coefficient exceeds the set threshold, it can be considered that this region has the risk of surface glass peeling. Therefore, according to these conditions, these regions are marked as surface glass peeling regions. Finally, record the coordinates of these peeling regions and the corresponding defect information, including the range and depth of peeling, etc., to obtain the surface glass peeling data. These data will provide an important basis for subsequent quality analysis, defect repair, and traceability data, helping to analyze whether there are quality problems in the production process of the bottle body and providing a reference for subsequent improvement measures.

[0065] Preferably, the abnormal detection of wall thickness non-uniformity in step S3 is specifically: Identify the low-strength region of the glass bottle body based on the structural strength of the bottle body; In this embodiment, in this step, it is first necessary to collect the material strength data and the bottle body structure data of the glass bottle. The structural strength threshold of the bottle body is set at 50 MPa, and all areas below this strength value are regarded as low-strength areas. Use finite element analysis (FEA) to model the structure of the glass bottle, set the material properties (such as the elastic modulus of the glass is 72 GPa and the Poisson's ratio is 0.22), and consider the shape and size of the bottle body during the modeling process. By simulating the influence of external forces (such as impact force, pressure, etc.) on the bottle body, calculate the stress value of each area, and compare it with the set strength threshold to identify the low-strength areas. During the calculation, use the geometric dimensions of the glass bottle and the preset material parameters, and based on the stress distribution map, obtain the position of the low-strength area as the basic data for subsequent operations.

[0066] According to the low-strength areas of the glass bottle body, perform grid division on the wall thickness area to obtain the grid area of the bottle body; In this embodiment, scan the surface of the bottle body, use a laser scanner to perform three-dimensional modeling on the glass bottle to generate the three-dimensional point cloud data of the bottle body. Import this point cloud data into the grid division software, and perform grid division according to the geometric structure of the bottle body and the distribution of the low-strength areas. Set the size of the grid unit to 25 mm × 25 mm to ensure that each grid unit can accurately reflect the wall thickness change. Perform uniform meshing on the bottle body. The divided grid area includes all surfaces of the bottle body, and ensure that the low-strength areas are consistent with the grid division for facilitating subsequent wall thickness measurement. Finally, obtain a meshed data including all surfaces and low-strength areas of the bottle body as the basis for wall thickness measurement.

[0067] Based on the grid area of the bottle body, measure the wall thickness of the glass bottle to obtain the wall thickness data of the glass bottle; In this embodiment, use the already divided grid area and adopt X-ray computed tomography (CT) technology or ultrasonic measurement technology to measure the wall thickness of each grid unit. The CT scan will generate cross-sectional images of each grid unit, and through image analysis algorithms, measure the wall thickness of each grid unit. For ultrasonic measurement, place sensors on the surface of the bottle body, and measure the reflection time of the wall thickness through ultrasonic pulses to obtain the wall thickness data of each grid. Measure all grid units one by one, and finally generate the wall thickness data of each grid area and record it as the input data for subsequent wall thickness deviation calculation.

[0068] Calculate the wall thickness deviation according to the wall thickness data of the glass bottle; identify the wall thickness deviation area based on the wall thickness deviation; In this embodiment, first, a standard wall thickness value is set (for example, the designed wall thickness of the bottle body is 4 mm). The actual wall thickness of each grid area is compared with the standard wall thickness, and the wall thickness deviation of each area is calculated. The formula for calculating the wall thickness deviation is: wall thickness deviation = actual wall thickness - standard wall thickness. For each grid cell, the wall thickness deviation data of each grid cell is calculated through the difference from the standard wall thickness. The wall thickness deviations of all grid areas are recorded and summarized to form an overall wall thickness deviation data set, which is convenient for subsequent identification of wall thickness deviation areas. First, a threshold range of the wall thickness deviation is set. Specifically, when the wall thickness deviation is greater than 0.5 mm, this area is regarded as having an obvious wall thickness deviation and needs to be marked. By traversing all grid areas, the wall thickness deviation of each grid is calculated one by one, and the areas with deviation values exceeding the set threshold are identified and marked as wall thickness deviation areas. The identification of wall thickness deviation areas is based on the magnitude of the deviation value and combines the actual wall thickness data of each grid, so as to facilitate more in-depth analysis and classification in the future.

[0069] Divide the wall thickness deviation types according to the wall thickness deviation areas to obtain vertical wall thickness deviation data and circumferential wall thickness deviation data; In this embodiment, by further analyzing the wall thickness deviation areas, based on the geometric structure of the bottle body, the wall thickness deviation is classified according to the vertical direction and circumferential direction of the bottle body. In this step, the vertical wall thickness deviation threshold is set to 0.3 mm, and the circumferential wall thickness deviation threshold is set to 0.3 mm. The wall thickness deviation data in each direction is extracted respectively. By classifying the wall thickness deviations in each direction, the vertical wall thickness deviation data and the circumferential wall thickness deviation data are obtained respectively. For the deviation in the vertical direction, the analysis is carried out based on the vertical section of the bottle body; for the circumferential deviation, the analysis is carried out on the horizontal section. These data will provide a basis for subsequent anomaly identification and eccentricity detection.

[0070] Based on the vertical wall thickness deviation data, identify the gravity casting anomaly to obtain the gravity casting anomaly data; In this embodiment, it is first necessary to set a deviation threshold for gravity casting. This threshold can be adjusted based on experience or historical data to ensure the accuracy of the recognition result. For example, the set threshold is 0.5 mm, which will be used as the standard for determining abnormal gravity casting. When the vertical wall thickness deviation value in a certain area exceeds this threshold and the deviation direction is consistent with the gravity direction of the bottle body, this area is considered to have abnormal gravity casting. During the analysis process, first, the gravity direction of the bottle body needs to be obtained, which can be determined through the geometric model of the bottle body and the gravity direction during the manufacturing process. Then, the vertical wall thickness deviation of each area is compared and analyzed with the gravity direction to identify those areas with larger wall thickness deviation values and the deviation direction consistent with the gravity direction. These areas are more likely to have abnormal gravity casting. In this way, the defective areas in the bottle body manufacturing process can be accurately located. Finally, the areas identified as having abnormal gravity casting are recorded and incorporated into the abnormal data set for subsequent analysis, correction, or optimization of the manufacturing process. These abnormal gravity casting data can provide valuable information for further quality control.

[0071] Based on the circumferential wall thickness deviation data, wall thickness eccentricity detection is performed to obtain the vertical wall thickness eccentricity ratio; In this embodiment, it is first necessary to perform wall thickness eccentricity detection based on the circumferential wall thickness deviation data. The circumferential wall thickness deviation data is obtained by measuring the wall thickness differences at different circumferential positions of the bottle body. Usually, the wall thickness is measured at multiple circumferential positions of the bottle body to obtain the wall thickness data at each position. Next, the calculation formula for the wall thickness eccentricity ratio is set. Specifically, the eccentricity value of each circumferential area can be obtained by calculating the difference between the maximum wall thickness and the minimum wall thickness in the area. The formula can be expressed as: Eccentricity value = Maximum wall thickness - Minimum wall thickness. The key to this step is to accurately extract the wall thickness data of each circumferential area and ensure that the wall thicknesses of multiple points are measured within each circumferential area, so as to calculate an accurate eccentricity value. After calculating the eccentricity value, it is necessary to compare these values with the circumferential uniformity of the bottle body to obtain the eccentricity ratio of each circumferential area. The calculation method of the eccentricity ratio can be carried out through the following formula: Eccentricity ratio = Eccentricity value / Circumferential average wall thickness. The circumferential average wall thickness is the average value of the wall thickness values of all measurement points within each circumferential area. This calculation helps to measure the wall thickness uniformity of each circumferential area. The larger the eccentricity ratio, the more uneven the wall thickness of the circumferential area and the more obvious the eccentricity phenomenon. Finally, based on the eccentricity ratios calculated for each circumferential area, the eccentricity ratios of all circumferential areas are summarized to obtain the overall vertical wall thickness eccentricity ratio. The vertical wall thickness eccentricity ratio can be obtained by weighted averaging the eccentricity ratios of each circumferential area or directly calculating the standard deviation. This index can provide the overall situation of the wall thickness eccentricity of the bottle body, reflecting whether the geometric shape and wall thickness distribution of the bottle body are uniform, and providing a basis for subsequent quality analysis and optimization.

[0072] Integrate the gravity casting anomaly data and the vertical wall thickness eccentricity to obtain the wall thickness non-uniformity data.

[0073] In this embodiment, the gravity casting anomaly data and the vertical wall thickness eccentricity are integrated to obtain the wall thickness non-uniformity data. Finally, the gravity casting anomaly data and the vertical wall thickness eccentricity data are integrated, and by methods such as weighted average, the final wall thickness non-uniformity data is generated. These data demonstrate the wall thickness non-uniformity problem existing in the manufacturing process of the glass bottle, providing a basis for subsequent quality control and defect traceability.

[0074] Preferably, in step S3, the evaluation of the alignment offset of the core and cavity is specifically as follows: Based on the wall thickness non-uniformity data, count the time periods of the wall thickness non-uniformity of the glass bottle. In this embodiment, first, collect and organize the wall thickness data of the glass bottles at each time point during the production process. These data can be obtained through CT scans, ultrasonic measurements, or real-time monitoring by optical sensors. For each glass bottle, record the wall thickness deviation during the production process and calculate the degree of wall thickness non-uniformity of each bottle body. Set a standard threshold for wall thickness non-uniformity. For example, the part with a deviation greater than 0.3 mm is regarded as non-uniform. Using time series analysis methods, based on the wall thickness non-uniformity data at each time point, count the wall thickness non-uniformity situation during the production cycle and identify the time periods with significant wall thickness non-uniformity problems. For example, if the wall thickness deviation of the bottle bodies within a certain time period is generally greater than the set standard threshold, then mark this time period as the "wall thickness non-uniformity time period". By calculating statistical quantities such as the average value and standard deviation of the wall thickness deviation values within each time period, confirm the significance of the non-uniformity phenomenon. Record and output these statistical results as the basis for subsequent analysis.

[0075] Extract the operating parameters of the forming mold according to the time periods of the wall thickness non-uniformity of the glass bottle. In this embodiment, according to the wall thickness non-uniformity time periods obtained in the previous step, extract the mold operating parameters related to the forming process during these time periods. By monitoring the production system, obtain various operating parameters of the mold during these time periods, such as the mold heating temperature, cooling time, injection pressure, injection speed, etc. These parameters are usually collected in real time by sensors connected to the forming equipment and recorded in the production control system. The set parameter ranges include the mold temperature between 100°C and 200°C, the injection pressure between 8 MPa and 15 MPa, and the injection speed between 30 mm / s and 60 mm / s. Extract the mold operating parameters for each time period, record all relevant parameters, and screen them according to the wall thickness non-uniformity time periods. Finally, obtain a set of mold operating parameters related to wall thickness non-uniformity for subsequent thermal expansion analysis.

[0076] Identify the thermal expansion state based on the operating parameters of the forming mold to obtain the mold thermal expansion data. In this embodiment, the thermal expansion state of the forming die is identified by analyzing the operating parameters of the die during the working process. First, a coefficient of thermal expansion is set (for example, the coefficient of thermal expansion of the die material is 12×10^-6 / °C), which can describe the volume expansion of the die during the heating process. According to the extracted die temperature change data, the temperature change range of the die during the heating process is calculated. For example, the temperature of the die rises from 25°C to 180°C during the heating stage. Combining the coefficient of thermal expansion, the expansion amount generated by the temperature change of the die during the heating process is calculated according to the formula ΔL = L0 * α * ΔT, where ΔL is the thermal expansion amount of the die, L0 is the initial length, α is the coefficient of thermal expansion, and ΔT is the temperature change. Through this process, die thermal expansion data can be obtained, reflecting the degree of expansion of the die caused by temperature change. Record these thermal expansion data to provide the necessary input information for subsequent die deformation analysis.

[0077] Calculate the die thermal expansion deformation amount based on the die thermal expansion data; In this embodiment, based on the thermal expansion data obtained from the forming die, the die thermal expansion deformation amount is calculated. First, the geometric dimensions and material properties of the die are set, such as the initial length, width, height of the die, and the elastic modulus of the die. By applying the thermal expansion theory and combining the geometric shapes of the die components, the deformation amount generated by the die during the heating process is calculated. According to the thermal expansion amount of the die and the material properties of the die, the thermal expansion deformation amounts of each part of the die can be calculated. For example, the deformation amount of a certain part of the die along the X-axis direction and the deformation amount of a certain part of the die along the Y-axis direction. The calculation formula is: thermal expansion deformation amount = initial dimension × coefficient of thermal expansion × temperature change. The key parameters in this step include the initial dimension of the die (for example, the die length is 1m and the width is 0.5m) and the coefficient of thermal expansion (such as the coefficient of thermal expansion of the die material is 12×10^-6 / °C). Through these parameters, the specific deformation value of the die during the heating stage is calculated, generating die thermal expansion deformation amount data as the basis for further predicting die deformation.

[0078] Predict the relative deformation probability of the die core and cavity based on the die thermal expansion deformation amount; In this embodiment, according to the thermal expansion deformation amount of the mold, the relative deformation probability of the mold core and the mold cavity is predicted. In this step, first, a geometric model between the mold core and the mold cavity is established, and the relative displacement between the mold core and the mold cavity is calculated according to the thermal expansion data of the mold. A standard deformation value is set. For example, when the relative deformation value between the mold core and the mold cavity exceeds 0.5 mm, it is considered that significant deformation has occurred. Through the thermal expansion deformation amount calculation model, the deformation amounts generated by the mold core and the mold cavity during the heating process are obtained, and according to the distribution of the deformation amounts, the relative deformation probability of the mold core and the mold cavity is predicted using probability theory methods. These probability data reflect the possibility of deformation between the mold core and the mold cavity under specific thermal expansion conditions, providing data support for mold design and adjustment.

[0079] Determine the alignment offset amount of the mold core and the mold cavity based on the relative deformation probability of the mold core and the mold cavity.

[0080] In this embodiment, in combination with the relative deformation probability data of the mold core and the mold cavity, the alignment offset amount between the mold core and the mold cavity is analyzed and calculated. First, according to the relative deformation probability calculated in the previous step, the regions with a greater possibility of deformation are determined, and the actual offset amounts of the mold core and the mold cavity are calculated in these regions. For example, if the relative deformation probability of a certain mold region is greater than 90%, there is a large alignment offset in this region. According to the tolerance requirements in mold design, the maximum allowable offset amount is calculated (for example, the maximum allowable offset is 0.2 mm). By combining the relative deformation probability with the allowable offset amount, the actual alignment offset amount of the mold core and the mold cavity is obtained. These data can be used to adjust the mold design and reduce quality problems caused by deformation.

[0081] Preferably, the detection of the positioning pin fracture in step S3 is specifically: Evaluate the mold assembly accuracy according to the alignment offset amount of the mold core and the mold cavity; In this embodiment, first, the alignment offset amount data of the mold core and the mold cavity at different time points during the production process are collected. This data can be measured by a high-precision displacement sensor or a three-dimensional laser scanner to obtain the relative displacement information of the mold core and the mold cavity. Specifically, measurement points are set and displacement measurements are carried out at key positions of the mold (such as the contact area between the mold cavity and the mold core). Assuming that the set allowable error is 0.1 mm, if the actually measured offset amount exceeds this threshold, it is considered that there is a problem with the mold assembly accuracy. By comparing the offset amount data in different time periods, the errors in the mold assembly process are identified, and the assembly accuracy is calculated. For example, if the maximum alignment offset amount between the mold core and the mold cavity is 0.15 mm, exceeding the standard error range, it indicates that the mold assembly accuracy is unqualified.

[0082] Conduct a positioning pin load detection based on the mold assembly accuracy to obtain positioning pin load data; In this embodiment, according to the assembly accuracy of the mold, the load condition of the locating pin is detected. A load sensor (such as a piezoelectric sensor or a strain gauge) is used and installed on the locating pin to monitor in real time the load borne by the locating pin during operation. A standard load value range is set. For example, the maximum load-bearing capacity of the locating pin during operation is 500 N. By measuring the load data of the locating pin under different working conditions, the load values of the locating pin at different production stages are obtained. This data reflects the load state of the locating pin and provides a basis for subsequent durability analysis. If the load of the locating pin exceeds the set maximum load threshold of 500 N in a certain measurement, it will be recorded as abnormal load data.

[0083] Classify according to the locating pin load data to obtain the static load data and dynamic load data of the locating pin; In this embodiment, according to the time-sequence change of the locating pin load, the load data is analyzed and classified. First, determine the criteria for distinguishing static load and dynamic load. Set the criterion for static load as the load remaining constant within a certain period of time, while dynamic load is manifested as the load changing rapidly with time. Through signal processing techniques such as Fourier transform, the load signal is decomposed into low-frequency (static) and high-frequency (dynamic) parts. Perform frequency-domain analysis on the load data to extract the static load and dynamic load data. For example, if the change in the load of the locating pin does not exceed 10 N within a certain period of time, it is regarded as static load; if the load change exceeds 10 N and the frequency is high, it is regarded as dynamic load. According to this method, the static load data and dynamic load data are obtained respectively.

[0084] Conduct material durability analysis based on the static load data of the locating pin to obtain material durability data; In this embodiment, first, set the durability standard of the material. For example, the material used for the locating pin needs to have a durability of at least 1 million static load cycles. According to the static load data (such as a single load of 200 N and a load duration of 2 hours), use the material fatigue test standard (such as the S-N curve) for analysis and calculate the fatigue life of the material under different loads. For example, if the static load maintains 200 N within 2 hours and the fatigue life of the material under this load is 500,000 times, then the material durability data is 500,000 times. In addition, according to different load levels, the durability of the material can be further divided. For example, the durability under low load is 800,000 times, under medium load is 600,000 times, and under heavy load is 400,000 times.

[0085] Detect the vibration amplitude based on the dynamic load data of the locating pin; In this embodiment, according to the dynamic load data, a vibration sensor (such as an accelerometer or a vibration sensor) is used to detect the vibration amplitude of the positioning pin. First, set the detection range of the vibration amplitude. For example, if the vibration amplitude exceeds 5 mm / s, it is considered abnormal vibration. The vibration sensor is installed near the positioning pin, and the vibration data of the positioning pin in the working state is collected in real time to calculate the vibration amplitude. Set the processing method of the vibration signal, including removing background noise and signal filtering. If the vibration amplitude data of the positioning pin exceeds 5 mm / s during operation, it is considered that the vibration amplitude exceeds the normal range. By statistically analyzing the vibration data, the distribution of the vibration amplitude is obtained and used for subsequent impact tolerance assessment.

[0086] Evaluate the impact tolerance of the positioning pin based on the vibration amplitude to obtain the impact tolerance data of the positioning pin; In this embodiment, according to the vibration amplitude data, the impact tolerance of the positioning pin is evaluated. First, set the standard of the impact tolerance. For example, the maximum impact acceleration that the positioning pin can withstand is 1000 m / s². If the vibration amplitude exceeds the set value of 5 mm / s, further analyze the impact components in the vibration signal. By high-speed sampling and fast Fourier transform, the vibration signal is decomposed into an impact signal and a normal vibration signal. Use the impact energy formula (such as E = 1 / 2 * m * v²) to calculate the impact energy in the vibration signal and further evaluate the impact tolerance of the positioning pin in the working state. If the calculated impact energy exceeds the set maximum impact energy standard, the impact tolerance data of the positioning pin indicates that its tolerance is insufficient.

[0087] Perform positioning pin fracture prediction based on the material durability data and the positioning pin impact tolerance data to obtain the positioning pin fracture data.

[0088] In this embodiment, the fracture prediction of the positioning pin is performed by integrating the material durability data of the static load and the impact tolerance data of the dynamic load. Set a comprehensive index. For example, the fracture probability of the positioning pin is related to factors such as the fatigue life of the material and the impact energy. According to the foregoing analysis, if the durability of the material is less than 800,000 times and the impact energy exceeds the set threshold, the fracture risk of the positioning pin increases. Use a stress analysis method (such as the finite element analysis method) to simulate the stress distribution of the positioning pin under different loads and predict the probability of its fracture occurrence. For example, if the predicted fracture probability is greater than 5%, it is considered that the positioning pin has a high fracture risk. Finally, obtain the fracture data of the positioning pin and use it for optimized design or early warning.

[0089] Preferably, this specification also provides a digital traceability system for the quality of lightweight glass bottles, which is used to execute the digital traceability method for the quality of lightweight glass bottles as described above. The digital traceability system for the quality of lightweight glass bottles includes: The bottle body structure strength detection module is used to obtain the design data of the glass bottle and construct a lightweight glass bottle model; detect the bottle body structure strength based on the lightweight glass bottle model; The defect traceability module is used to detect the surface defects of the bottle body based on the bottle body structure strength and generate the surface defect data of the bottle body; perform defect traceability based on the surface defect data of the bottle body to obtain the process data of the production defects of the bottle body; The positioning pin fracture detection module is used to detect the abnormal wall thickness unevenness based on the bottle body structure strength to obtain the wall thickness unevenness data; evaluate the alignment deviation of the core and cavity based on the wall thickness unevenness data; perform positioning pin fracture detection according to the alignment deviation of the core and cavity to obtain the positioning pin fracture data; The glass quality traceability module is used to generate a unique glass number based on the lightweight glass bottle model; identify the abnormal batches of the lightweight glass bottles according to the process data of the production defects of the bottle body and the positioning pin fracture data to obtain the abnormal batch data; perform abnormal glass number mapping on the abnormal batch data according to the unique glass number to obtain the abnormal glass number, and transmit it to the production equipment system to execute the glass quality traceability task.

[0090] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be included in the present invention.

[0091] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A digital traceability method for the quality of lightweight glass bottles, characterized in that, It includes the following steps: Step S1: Obtain the design data of the glass bottle and construct a lightweight glass bottle model; Detect the structural strength of the bottle body based on the lightweight glass bottle model; Step S2: Conduct surface defect detection of the bottle body based on the structural strength of the bottle body, and generate surface defect data of the bottle body; Trace the defects based on the surface defect data of the bottle body to obtain the process data of production defects of the bottle body; Step S3: Conduct uneven wall thickness anomaly detection based on the structural strength of the bottle body to obtain uneven wall thickness data; Evaluate the alignment offset of the core and cavity based on the uneven wall thickness data; Detect the fracture of the positioning pin according to the alignment offset of the core and cavity to obtain positioning pin fracture data; Step S4: Generate a unique glass number based on the lightweight glass bottle model; Identify abnormal batches of lightweight glass bottles according to the process data of production defects of the bottle body and the positioning pin fracture data to obtain abnormal batch data; Perform abnormal glass number mapping on the abnormal batch data according to the unique glass number to obtain abnormal glass numbers, and transmit them to the production equipment system to execute the glass quality traceability task.

2. The method for digital traceability of the quality of lightweight glass bottles according to claim 1, characterized in that, Specifically, step S1 is as follows: Step S11: Obtain the design data of the glass bottle and extract the three-dimensional structure data of the glass bottle; Step S12: Identify the key optimization areas according to the three-dimensional structure data of the glass bottle to obtain the bottle body area and the bottle shoulder area; Step S13: Conduct curvature transition optimization design based on the bottle body area to obtain the optimized curvature of the bottle body; Step S14: Conduct wall thickness reduction design based on the bottle shoulder area to obtain wall thickness reduction data; Step S15: Integrate the optimized curvature of the bottle body and the wall thickness reduction data, and construct a lightweight glass bottle model; Step S16: Detect the structural strength of the bottle body based on the lightweight glass bottle model.

3. The method for digital traceability of the quality of lightweight glass bottles according to claim 2, characterized in that, Specifically, step S16 is as follows: Step S161: Import the lightweight glass bottle model into the simulation software; Step S162: Set the wall thickness range of the bottle body to 0.3mm - 3mm, the bottleneck thickness range to 0.2mm - 1.5mm, and the bottle bottom thickness range to 0.5mm - 4mm in the simulation software; Step S163: Set the external pressure range to 0.5MPa - 2MPa, the internal pressure range to 0.3MPa - 1.5MPa, and the transportation impact force range to 10N - 200N in the simulation software; Step S164: Set the elastic modulus range of the glass material to 60GPa - 75GPa and the Poisson's ratio range to 0.22 - 0.26 in the simulation software; Step S165: Run the structural strength analysis program in the simulation software and record the structural strength of the bottle body.

4. The method for digital traceability of the quality of lightweight glass bottles according to claim 1, characterized in that, Specifically, step S2 is as follows: Step S21: Extract the deformation characteristics and pressure resistance characteristics of the bottle body according to the structural strength of the bottle body to obtain the bottle body deformation data and the bottle body pressure resistance data; Step S22: Conduct bottle body crack detection based on the bottle body deformation data to obtain bottle body crack data; Step S23: Conduct transportation simulation based on the bottle body pressure resistance data and identify the peeling of the surface glass during the transportation of the bottle body to obtain the surface glass peeling data; Step S24: Integrate the bottle body crack data and the surface glass peeling data to obtain the surface defect data of the bottle body; Step S25: Trace the source of the defects based on the defect data on the bottle surface to obtain the process data of the production defects of the bottle.

5. The method for digital traceability of the quality of lightweight glass bottles according to claim 4, characterized in that, Step S22 specifically includes: Step S221: Control the rotation of the bottle based on the bottle deformation data to obtain the bottle rotation data; Step S222: Calculate the stress gradient based on the bottle rotation data, and extract the high stress gradient value based on the stress gradient; Step S223: Identify the high deformation area of the bottle according to the high stress gradient value; Step S224: Perform circular X-ray irradiation scanning on the high deformation area of the bottle to obtain the bottle X-ray image; Step S225: Extract the significant boundary according to the bottle X-ray image to obtain the significant boundary data; Step S226: Calculate the aspect ratio based on the significant boundary data to obtain the aspect ratio; Step S227: Identify the crack profile according to the aspect ratio; Step S228: Calculate the crack defect density based on the crack profile to obtain the bottle crack data.

6. The method for digital traceability of the quality of lightweight glass bottles according to claim 4, characterized in that, Step S23 specifically includes: Step S231: Screen the low-pressure-resistant bottles based on the bottle pressure resistance data and perform transportation simulation to obtain the transportation data of the low-pressure-resistant bottles; Step S232: Conduct a force analysis according to the transportation data of the low-pressure-resistant bottles to obtain the force data of the bottle during transportation; Step S233: Identify the stress concentration area based on the force data of the bottle during transportation and record it as the peeling risk area; Step S234: Obtain the bottle multispectral image according to the peeling risk area; Step S235: Extract the spectral feature vectors of the pixel points based on the bottle multispectral image, and construct the pixel-level spectral feature matrix according to the spectral feature vectors of the pixel points; Step S236: Identify the spectral abnormal band based on the pixel-level spectral feature matrix and calculate the band mutation coefficient; Step S237: Calculate the reflectivity gradient based on the bottle multispectral image; Determine the peeling of the surface glass based on the band mutation coefficient and the reflectivity gradient to obtain the surface glass peeling data.

7. The method for digital traceability of the quality of lightweight glass bottles according to claim 1, characterized in that The specific detection of wall thickness non-uniformity abnormality in Step S3 is as follows: Identify the low-strength area of the glass bottle based on the bottle structure strength; Perform grid division on the wall thickness area according to the low-strength area of the glass bottle to obtain the bottle grid area; Measure the wall thickness of the glass bottle based on the bottle grid area to obtain the wall thickness data of the glass bottle; Calculate the wall thickness deviation according to the wall thickness data of the glass bottle; Identify the wall thickness deviation area based on the wall thickness deviation; Classify the wall thickness deviation types according to the wall thickness deviation area to obtain the vertical wall thickness deviation data and the circumferential wall thickness deviation data; Identify the gravity casting abnormality based on the vertical wall thickness deviation data to obtain the gravity casting abnormality data; Detect the wall thickness eccentricity based on the circumferential wall thickness deviation data to obtain the vertical wall thickness eccentricity; Integrate the gravity casting abnormality data and the vertical wall thickness eccentricity to obtain the wall thickness non-uniformity data.

8. The method for digital traceability of the quality of lightweight glass bottles according to claim 1, characterized in that The specific evaluation of the alignment offset of the mold core and mold cavity in Step S3 is as follows: Statistically count the time period of the wall thickness non-uniformity of the glass bottle based on the wall thickness non-uniformity data; Extract the operating parameters of the forming mold according to the time period of the wall thickness non-uniformity of the glass bottle; Identify the thermal expansion state based on the operating parameters of the forming mold to obtain the mold thermal expansion data; Calculate the thermal expansion deformation amount of the mold according to the mold thermal expansion data; Predict the relative deformation probability of the mold core and mold cavity based on the thermal expansion deformation amount of the mold; Determine the alignment offset of the core and cavity based on the relative deformation probability of the core and cavity.

9. The method for digital traceability of the quality of lightweight glass bottles according to claim 1, characterized in that The specific detection of the positioning pin fracture in step S3 is as follows: Evaluate the die assembly accuracy according to the alignment offset of the core and cavity; Conduct a positioning pin load test based on the die assembly accuracy to obtain positioning pin load data; Classify according to the positioning pin load data to obtain static positioning pin load data and dynamic positioning pin load data; Conduct a material durability analysis based on the static positioning pin load data to obtain material durability data; Detect the vibration amplitude based on the dynamic positioning pin load data; Evaluate the impact tolerance of the positioning pin based on the vibration amplitude to obtain positioning pin impact tolerance data; Conduct a positioning pin fracture prediction according to the material durability data and the positioning pin impact tolerance data to obtain positioning pin fracture data.

10. A digital traceability system for the quality of lightweight glass bottles, characterized in that, For implementing the method for digital traceability of the quality of lightweight glass bottles as described in claim 1, the system for digital traceability of the quality of lightweight glass bottles includes: A bottle body structure strength detection module, configured to obtain glass bottle design data and construct a lightweight glass bottle model; detect the bottle body structure strength based on the lightweight glass bottle model; A defect traceability module, configured to detect surface defects of the bottle body based on the bottle body structure strength to generate bottle body surface defect data; conduct defect traceability based on the bottle body surface defect data to obtain bottle body production defect process data; A positioning pin fracture detection module, configured to detect wall thickness unevenness anomalies based on the bottle body structure strength to obtain wall thickness unevenness data; evaluate the alignment offset of the core and cavity according to the wall thickness unevenness data; conduct positioning pin fracture detection according to the alignment offset of the core and cavity to obtain positioning pin fracture data; A glass quality traceability module, configured to generate a unique glass number based on the lightweight glass bottle model; identify abnormal batches of lightweight glass bottles according to the bottle body production defect process data and the positioning pin fracture data to obtain abnormal batch data; perform an abnormal glass number mapping on the abnormal batch data according to the unique glass number to obtain an abnormal glass number, and transmit it to the production equipment system to execute the glass quality traceability task.

Images