Digital twinborn control method and system for coring of special optical component

Abstract

The invention relates to the technical field of process control, and discloses a digital twinning control method and system for coring of a special optical component. The method comprises the steps that a digital twin model is constructed through a multi-source sensor, and a machining geometric reference is determined; distortion features are identified, residual stress distribution is analyzed, and classification identifiers are generated; a dynamic simulation prediction technology is adopted to optimize a machining path based on distortion causes, and an adjustment scheme is formed; quantizing the parameter weight and adjusting the stress release regulation and control parameter to generate an optimized execution scheme; processing dynamic evolution is tracked, the distortion cause attenuation trend is analyzed, and geometric residual error distribution is obtained; if the error does not reach the standard, clamping parameters are adjusted, the model is updated, and a final control strategy is determined; and executing the strategy and integrating multi-source data, and outputting a final processing control result. According to the invention, the precision of special optical component coring processing is improved.

Description

Technical Field

[0001] This application relates to the field of process control technology, and in particular to a digital twin control method and system for core extraction of special optical components. Background Technology

[0002] Specialized optical components are core components of high-end optical systems, such as lithography machines, aerospace remote sensing equipment, and high-precision interferometers. The core-taking accuracy directly determines the imaging quality and operational stability of the optical system. However, during the core-taking process, the combined effects of multiple factors, such as residual stress release, clamping tilt, environmental temperature and humidity fluctuations, and uneven film thickness, can easily lead to uncontrollable distortion in the components, severely affecting the processing accuracy.

[0003] Existing core-taking processing control methods mostly rely on experience-based parameter presets or single-sensor feedback regulation, which has three major pain points: First, it is difficult to accurately identify the coupling relationship of multiple distortion sources, resulting in a lack of targeted regulation strategies; second, the processing process is disconnected from virtual simulation, making it impossible to predict processing path deviations in advance, requiring repeated trial and error adjustments, which reduces production efficiency; third, the dynamic evolution of key parameters such as residual stress and clamping posture is not tracked in real time, making it difficult to accurately control residual errors and meet the nanometer-level precision requirements of high-end optical components. Digital twin technology provides new possibilities for virtual-real fusion process control, but its application in core-taking processing of optical components is still in its early stages. It has not yet formed a complete closed-loop control scheme of "data acquisition-simulation prediction-dynamic regulation-precision verification", and cannot effectively solve the high-precision control problem under multi-source distortion coupling.

[0004] Therefore, developing a core-taking processing control method that integrates digital twin technology to achieve multi-factor collaborative regulation and precision assurance has become an urgent need in the field of special optical manufacturing. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides a digital twin control method and system for core taking of special optical components, which enables accurate prediction and real-time control of the core taking process of special optical components, effectively improving the accuracy, stability and yield of core taking.

[0006] In a first aspect, this application provides a digital twin control method for chip extraction of special optical components, the method comprising:

[0007] Step S1: Collect geometric topography data of the surface to be processed of the special optical component using multi-source sensors, construct a digital twin model of the special optical component, and determine the geometric reference for core processing;

[0008] Step S2: Based on the core extraction processing geometric datum, identify the distortion features during the processing, integrate environmental data to analyze the residual stress distribution during the core extraction process, and generate distortion feature classification labels;

[0009] Step S3: Identify the cause of distortion based on the distortion feature classification identifier, and use dynamic simulation prediction technology to predict the core extraction processing path. Combine dynamic working condition data to dynamically adjust the core extraction processing path and determine the core extraction process path adjustment scheme.

[0010] Step S4: According to the core extraction process path adjustment scheme, quantify the influence weight of each core extraction parameter on the core extraction accuracy, adjust the stress release control parameters, and determine whether the adjusted core extraction accuracy exceeds the preset range. If it does, generate an optimized core extraction execution scheme.

[0011] Step S5: Based on the core extraction execution scheme, track the dynamic evolution record of the processing process, analyze the real-time decay trend of multi-source distortion causes, and obtain the geometric residual error distribution after processing;

[0012] Step S6: If the geometric residual error distribution does not meet the preset accuracy standard, adjust the assembly and clamping tilt compensation parameters, update the digital twin model, and determine the final core extraction control strategy.

[0013] Step S7: Based on the final core extraction control strategy, acquire real-time evolution record data during the core extraction process, integrate stress release control effect and working condition dynamic data, and output the final core extraction processing control result.

[0014] Secondly, this application provides a digital twin control system for the core extraction of special optical components, the system comprising:

[0015] The module is used to collect geometric topography data of the surface to be processed of special optical components through multi-source sensors, construct a digital twin model of the special optical components, and determine the geometric reference for core processing.

[0016] The classification module is used to identify distortion features during the core extraction process based on the core extraction geometric reference, analyze the residual stress distribution during the core extraction process by integrating environmental data, and generate distortion feature classification labels.

[0017] The adjustment module is used to identify the cause of distortion based on the distortion feature classification identifier, and to predict the core extraction processing path using dynamic simulation prediction technology. It also integrates dynamic working condition data to dynamically adjust the core extraction processing path and determine the core extraction process path adjustment scheme.

[0018] The optimization module is used to quantify the influence weight of each core extraction parameter on the core extraction accuracy according to the core extraction process path adjustment scheme, adjust the stress release control parameters, and determine whether the adjusted core extraction accuracy exceeds the preset range. If it does, an optimized core extraction execution scheme is generated.

[0019] The analysis module is used to track the dynamic evolution record of the processing process according to the core extraction execution scheme, analyze the real-time decay trend of multi-source distortion causes, and obtain the geometric residual error distribution after processing.

[0020] The update module is used to adjust the assembly and clamping tilt compensation parameters, update the digital twin model, and determine the final core extraction control strategy if the geometric residual error distribution does not meet the preset accuracy standard.

[0021] The output module is used to acquire real-time evolution record data during the core extraction process based on the final core extraction control strategy, integrate stress release control effect and working condition dynamic data, and output the final core extraction processing control result.

[0022] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0023] This application provides a digital twin control method and system for core taking of special optical components. The method deeply integrates environmental and operating condition data collected by multi-source sensors with a digital model. During processing, it intelligently identifies and classifies distortion features and visualizes the residual stress distribution through stress heatmaps, thereby accurately locating the cause of distortion. Furthermore, using dynamic simulation prediction technology, the processing path is simulated and iteratively optimized under the coupling of multiple influencing factors such as residual stress and clamping tilt, generating an adaptive process adjustment scheme. Based on this scheme, the stress release process is dynamically regulated by quantifying parameter weights and implementing closed-loop feedback control, while real-time tracking of processing evolution records and analysis of the attenuation trend of distortion causes are performed, ultimately generating a quantified geometric residual error distribution. If the error does not meet the standard, the clamping compensation parameters can be automatically adjusted and the twin model updated to form the final control strategy, and then a high-precision processing control result is output through a data fusion algorithm.

[0024] This application effectively solves the precision control problem caused by the complex physical state, diverse distortion factors, and fixed processing path in the core-taking process of special optical components. It significantly improves the predictability, adaptability, and controllability of the processing, and ultimately greatly improves the yield and process stability while ensuring processing accuracy. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart of a digital twin control method for chip extraction of special optical components in an embodiment of this application; Figure 2 This is a schematic diagram of a stress heat map according to an embodiment of this application; Figure 3 This is a schematic diagram of the geometric residual error distribution cloud map of an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a digital twin control system for core extraction of special optical components according to an embodiment of this application. Detailed Implementation

[0027] This application provides a digital twin control method and system for the core extraction of special optical components. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0028] Core extraction for specialized optical components is a core and critical process in the precision machining of specialized optical components. Essentially, it involves precisely separating and shaping the core functional area, or "core," from specialized optical blanks, such as laser crystal blanks and optical glass blanks, based on pre-defined optical performance and geometric accuracy requirements. This process requires the removal of non-functional areas from the blank using ultra-precision cutting and grinding techniques. Throughout the machining process, cutting parameters, clamping posture, and machining stress must be strictly controlled to prevent phase distortion, surface damage, or internal cracks in the core area. The final formed core area directly determines the core performance indicators of specialized optical components, such as optical transmission efficiency and imaging accuracy, and is widely used in key optical systems such as high-end laser equipment, high-precision imaging instruments, and fiber optic communication modules.

[0029] Therefore, this application proposes a precise control method for core extraction of special optical components based on digital twins. Through closed-loop management of the entire process of "data acquisition - model construction - scheme optimization - process tracking - parameter adjustment - result verification", it solves the technical pain points in core extraction, such as difficulty in precision control, large multi-source distortion interference, and poor scheme adaptability, and ensures that the core extraction accuracy and optical performance meet the standards.

[0030] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the digital twin control method for chip extraction of special optical components in this application includes:

[0031] Step S1: Collect geometric morphology data of the surface to be processed of special optical components through multi-source sensors, construct a digital twin model of the special optical components, and determine the geometric reference for core taking processing.

[0032] Step S1 includes: collecting geometric topography data, core-taking tool condition data, and clamping and positioning data of the surface to be processed of the special optical component through multi-source sensors, and integrating them to form complete core-taking physical state data; based on the core-taking physical state data, constructing a digital twin model of the special optical component using virtual modeling mapping technology, and realizing real-time synchronization between the collected data and the model state; comparing the synchronized digital twin model with the preset design target model to determine the core-taking machining geometric reference.

[0033] Specifically, a multi-source sensor acquisition system is activated. These sensors include a laser profile sensor, a force sensor, and a visual positioning sensor. The laser profile sensor acquires geometric topography data of the surface to be processed from the special optical components. This data is specifically surface point cloud coordinate information, characterizing the surface's unevenness. The force sensor, integrated into the shank of the core-taking tool, acquires core-taking tool condition data, including cutting force, vibration frequency, and other parameters reflecting the tool's working state. The visual positioning sensor is mounted above the processing platform to acquire clamping and positioning data, specifically the actual positioning coordinates and attitude angles of the components on the clamping fixture. Subsequently, the above three types of acquired data are transmitted to the data integration module. Data splicing and redundancy removal algorithms are used to integrate the data, forming complete core-taking physical state data. The data splicing algorithm uses coordinate system transformation to correlate data from different sources, while the redundancy removal algorithm uses the 3σ criterion to eliminate abnormal data, ensuring data integrity and reliability.

[0034] Based on the integrated core-taken physical state data, a digital twin model of a special optical component is constructed using virtual modeling mapping technology. This digital twin model is a virtual simulation model that can map the physical state of a physical entity in real time. Its input data is the aforementioned core-taken physical state data. The model construction process is as follows: First, the geometric shape data is reconstructed using reverse engineering software to generate a geometric model of the component; then, the tool condition data and clamping positioning data are associated and mapped with the geometric model, giving the geometric model physical and motion attributes; to achieve real-time synchronization between the acquired data and the model state, a real-time data transmission protocol is used to transmit the sensor-acquired data to the digital twin model at a preset frequency, such as 100Hz. The model updates the state parameters of the corresponding virtual module in real time based on the received data. Finally, the synchronized digital twin model is compared with the preset design target model to determine the geometric reference for core processing. The design target model is a pre-stored ideal processing state model of the component, including parameters such as ideal geometric dimensions and surface roughness. The comparison process is implemented using an iterative nearest-point algorithm, specifically: first, the feature point sets of the synchronized digital twin model and the design target model are extracted respectively. The feature point sets are selected from the point cloud data of key geometric feature parts such as the edge contour points and hole centers of the component; then, the optimal matching transformation matrix is ​​calculated through iterative optimization to map the feature point set of the digital twin model to the feature point set coordinate system of the design target model, completing the spatial alignment of the two models; subsequently, the geometric deviation of the corresponding feature points of the two models after alignment is calculated, including geometric dimensions. Deviations include the difference between the actual size and the ideal size, and surface profile deviations include the normal distance between the actual surface point cloud and the ideal curved surface. Weighting coefficients are set, and the setting of weighting coefficients is combined with the accuracy requirements of core machining. Geometric deviations directly determine the assembly compatibility of components and have a more critical impact on machining accuracy. Therefore, the weight of geometric deviations can be set to be greater than that of surface profile deviations. For example, the weight of geometric deviations is 0.6 and the weight of surface profile deviations is 0.4. For each feature point, the comprehensive deviation of a single feature point is calculated according to the formula: comprehensive deviation = geometric deviation × corresponding weighting coefficient + surface profile deviation × corresponding weighting coefficient. Then, the comprehensive deviations of all feature points are mapped to the model coordinate system according to their spatial distribution to form a comprehensive geometric deviation distribution covering the entire surface of the component.

[0035] Finally, the continuous region with the smallest deviation is selected from the comprehensive geometric deviation distribution. The actual state of the region with the smallest deviation is closest to the ideal design state. Using this as a benchmark can minimize the transmission effect of benchmark deviation on subsequent processing. The geometric benchmark parameters for core taking in this region are defined, including benchmark planes such as the mounting and mating surfaces of components, benchmark axes such as the central axis of holes, the rotation axis of cylindrical surfaces, benchmark points such as corner vertices, the center of feature holes, etc., as well as the corresponding benchmark dimension parameters such as the flatness tolerance of the benchmark plane, the straightness tolerance of the benchmark axis, the distance between benchmark points, and the perpendicularity between the benchmark plane and the benchmark axis, etc., thus completing the determination of the geometric benchmark for core taking.

[0036] By integrating multi-source data and using digital twin modeling, a precise mapping between physical entities and virtual models was achieved. The determined geometric datum for core extraction provides a reliable reference for subsequent precision control, effectively solving the problems of inaccurate core extraction datum and the disconnect between physical and virtual states that lead to processing deviations. This improves the accuracy and controllability of core extraction and lays the foundation for subsequent distortion cause identification and processing path optimization.

[0037] Step S2: Based on the geometric datum of core extraction, identify the distortion features in the processing, integrate environmental data to analyze the residual stress distribution in the core extraction process, and generate distortion feature classification labels.

[0038] Step S2 includes: generating a preliminary phase distribution reference map based on the core extraction geometric reference; scanning the phase distribution reference map using a phase gradient calculation method; calculating the phase change rate of each pixel; if the phase change rate exceeds a preset threshold, defining the corresponding pixel as a distortion feature and generating a distortion key region; collecting environmental data during the core extraction process; fusing the distortion key region information with the environmental data; performing stress tensor calculation on the fused data to obtain a stress heat map of the residual stress distribution state; establishing a mapping relationship between distortion features and distortion causes based on the stress heat map; classifying distortion features into multiple categories based on the mapping relationship; assigning an identification code to each type of distortion feature; and generating a distortion feature classification identifier that represents the association between distortion features and their causes.

[0039] Specifically, during the core-taking process, environmental fluctuations such as temperature and humidity changes, processing vibrations, and residual stress from clamping can easily trigger the release of residual stress, leading to distortion in special optical components. This distortion directly damages the optical performance and processing accuracy of the components. Therefore, after determining the geometric datum for core-taking, the next step is to identify distortion features and correlate their causes. First, a preliminary phase distribution datum map is generated based on the core-taking geometric datum. The phase distribution datum map is a two-dimensional image representing the phase distribution of the component under its current processing state. Its generation process is as follows: using parameters such as the reference plane and reference axis in the core-taking geometric datum as references, a unified coordinate system is established. A digital twin model is retrieved, and the actual phase values ​​of each position of the component in the model are extracted. These actual phase values ​​are then filled into the corresponding pixels according to the coordinate mapping relationship, forming a preliminary phase distribution datum map where pixel values ​​and phase values ​​correspond one-to-one. The coordinates of the pixels in the map are precisely matched with the actual surface position of the component, and the phase value represents the current actual optical phase state at that position. Subsequently, the phase distribution baseline image is scanned using a phase gradient calculation method to calculate the phase change rate of each pixel. Specifically, the Sobel operator is used to perform a convolution operation on the phase distribution baseline image, and the phase partial derivatives of each pixel along the x and y axes are calculated. Then, the phase change rate is calculated using the formula: Calculate the phase gradient magnitude of this pixel, i.e., the phase change rate, where, Let be the phase partial derivative in the x-axis direction. The phase partial derivative in the y-axis direction represents the phase change rate. A larger amplitude indicates a greater rate of phase change, suggesting a more drastic phase change at that location and a higher likelihood of distortion. Therefore, a preset phase change rate threshold is set. This threshold is determined through multiple core-taking and machining experiments. During the scanning of the phase distribution baseline map, if the phase change rate of a pixel exceeds this preset threshold, the surface of the corresponding component is considered to have a distortion risk, and this is defined as a distortion feature. Subsequently, adjacent distortion feature pixels are analyzed and merged using connected component analysis to generate a continuous distortion critical region. The coordinate range, area, and pixel distribution of this region are output as distortion critical region information.

[0040] Next, environmental data is collected during the core extraction process. This environmental data specifically includes temperature, humidity, vibration, and air pressure data of the processing environment. These data are collected in real-time at preset frequencies by temperature, humidity, vibration, and air pressure sensors installed around the processing platform. After preprocessing, the data is transmitted to the data fusion module. The previously generated information on the key distortion region is fused with the preprocessed environmental data. The fusion process employs a data association mapping algorithm, using the coordinates of the key distortion region as an index to bind environmental data within the same time dimension and spatial range to the key distortion region information, forming fused data of distortion location and environmental parameters. Subsequently, based on the fused data and the digital twin model, the residual stress distribution is obtained. The finite element analysis module in the digital twin model is then invoked, using environmental data from the fused data, such as temperature field changes and vibration acceleration fields, as physical boundary conditions and loads, applied to the three-dimensional simulation model of the component. The simulation model has pre-set physical parameters of the component material, including elastic modulus (e.g., fused silica, Poisson's ratio, coefficient of thermal expansion, and density). By solving the elastoplastic constitutive equations coupled with thermo-mechanical effects, simulations were performed to derive the predicted three-dimensional residual stress field of the component under current environmental loads and historical processing loads. The equivalent stress corresponding to the distortion-critical regions on the component surface was extracted, such as the Mises stress distribution, and mapped onto a two-dimensional predicted stress distribution map. The calibrated predicted surface equivalent stress distribution data was visualized using a color gradient mapping method to generate a stress heatmap, as shown below. Figure 2 As shown, this is a schematic diagram of a stress heatmap. The diagram clearly shows the distribution boundaries and isostress lines of the high-stress, medium-stress, and low-stress regions. The X-axis and Y-axis correspond to the actual coordinates of the surface of the special optical component, respectively. The magnitude of residual stress is represented by different color gradients, providing an intuitive and visual basis for subsequent correlation analysis between distortion characteristics and stress distribution.

[0041] Finally, based on the stress heatmap and fused data, a mapping relationship between distortion features and distortion causes is established through quantitative rules, and classification labels are generated. Specifically: First, the spatial overlap between the key distortion region and the high-stress region in the stress heatmap is calculated. The high-stress region refers to the region in the stress heatmap where the equivalent stress value exceeds a preset threshold, such as 50 MPa. The area of ​​the key distortion region and the area of ​​the high-stress region are calculated using image processing algorithms, and the intersection area of ​​the two regions is obtained. The spatial overlap is defined as the intersection area divided by the area of ​​the smaller of the two regions. Next, automatic decision-making is performed based on quantitative thresholds calibrated through process experiments. For example, the threshold standard for "complete overlap" is a spatial overlap of not less than 0.9, and the threshold standard for "partial overlap" is a spatial overlap between 0.3 and 0.9. At the same time, judgment thresholds for environmental parameter fluctuations are set: for example, the threshold for temperature change is ±0.5 degrees Celsius, the threshold for vibration acceleration is 0.1 times the gravitational acceleration, and the threshold for humidity change is ±5% of relative humidity. Then, the automatic classification decision-making process is executed. For each critical distortion region, the following judgments are made in sequence: If the spatial overlap between this region and the high-stress region is 0.9 or higher, and the absolute value of the maximum temperature change monitored in the corresponding fused data exceeds 0.5 degrees Celsius, then the distortion feature is determined to be "temperature stress type". If the spatial overlap is between 0.3 and 0.9, and the maximum vibration acceleration monitored in the corresponding fused data exceeds 0.1 times the gravitational acceleration, then it is determined to be "vibration-stress coupling type". If the spatial overlap is less than 0.3, but the absolute value of the maximum humidity change monitored in the corresponding fused data exceeds 5%RH, then it is determined to be "humidity stress type". If none of the above conditions are met, but the distance between the geometric center of the critical distortion region and the clamping point recorded in the digital twin model, such as the center distance of the contact surface of the clamping jaws, is less than 2 mm, then it is determined to be "clamping stress type". Finally, based on the above decision results, a structured distortion feature classification label is generated for each critical distortion region. This identifier is stored as a database record, and its content includes at least: region number, center coordinates, region area, determined cause type (one of the four types mentioned above), the triggered rule number (e.g., "Rule 1"), and the measured value of the main environmental parameter that led to the determination (e.g., "Temperature change: +0.6°C"). This identifier is used in subsequent step S3 to guide targeted processing path optimization.

[0042] Through the above steps, accurate identification of distortion features and precise correlation with their causes are achieved, solving the problems of ambiguous distortion feature positioning and inaccurate cause judgment. The generated distortion feature classification labels provide a precise basis for targeted optimization of subsequent processing paths, ensuring the effectiveness and targeted nature of core extraction processing precision control.

[0043] Step S3: Identify the causes of distortion based on the distortion feature classification labels, and use dynamic simulation prediction technology to predict the core extraction processing path. Combine dynamic working condition data to dynamically adjust the core extraction processing path and determine the core extraction process path adjustment scheme.

[0044] Step S3 includes: identifying distortion causes based on distortion feature classification identifiers; determining whether the dominant influencing factor of the core extraction processing state is the coupling effect of residual stress and clamping tilt based on the distortion causes; if so, triggering the dynamic simulation prediction process; simulating the core extraction processing path using dynamic simulation prediction technology based on the determination result of coupling dominance, and iteratively calculating the phase distribution evolution trend and stress release trend of the core extraction processing path under the coupling dominance state by combining fused data; collecting dynamic working condition data in real time during the core extraction process, fusing the dynamic working condition data into the simulation process, adjusting the weights of various parameters in the simulation process through data calibration algorithms, and correcting the simulation results of the core extraction path; comparing the corrected simulation path with the preset core extraction path benchmark, calculating the path deviation value, and generating a core extraction process path adjustment scheme if the path deviation value exceeds the preset threshold.

[0045] Specifically, different distortion causes affect core extraction through varying mechanisms. In particular, the coupling effect of residual stress and clamping tilt significantly alters stress distribution and material removal characteristics during processing. Directly using a preset path can easily exacerbate distortion or result in substandard processing accuracy. Therefore, after generating distortion feature classification labels, it is necessary to identify distortion causes and optimize the core extraction path. By identifying the core causes through distortion feature classification labels, a targeted dynamic simulation prediction process can be triggered. Combined with working condition data, the processing path can be dynamically corrected, achieving precise adaptation of the processing path and providing scientific path guidance for subsequent high-precision core extraction. Specifically, the distortion cause is first identified based on the distortion feature classification identifier. By parsing the cause field and corresponding identifier code in the identifier, the core inducing factor of each type of distortion feature can be directly located. This further determines whether the dominant influencing factor of the core-taking processing state is the coupling effect of residual stress and clamping tilt. The specific judgment criteria are as follows: when the overlap area of ​​the distortion feature identified as clamping stress type exceeds a preset threshold (e.g., 50%) with other residual stress type distortion features, and the attitude angle deviation in the corresponding clamping positioning data exceeds a preset threshold (e.g., 0.02°), the dominant influencing factor is determined to be the coupling effect of residual stress and clamping tilt. The calculation method for the overlap area ratio is as follows: extract the area of ​​the "clamping stress type" distortion feature, the distortion areas identified as "temperature stress type" and "vibration-stress coupling type," and calculate the area of ​​the merged overlapping area among the three. If the proportion of the overlapping area to the total area of ​​all identified distortion areas exceeds a preset threshold (e.g., 50%), it means that the action area of ​​clamping stress and other types of residual stress on the component surface highly overlaps, and the two have formed a... The coupling effect, rather than independent action, causes the local stress amplitude to superimpose, significantly increasing the risk of aggravated distortion. In the clamping and positioning data, the attitude angle deviation specifically refers to the angular difference between the actual clamping posture and the ideal clamping posture of the special optical component. It includes three dimensions: pitch angle deviation, yaw angle deviation, and roll angle deviation. The ideal clamping posture refers to the standard clamping posture defined in advance in the digital twin model and machining process file to meet the core machining accuracy requirements of the special optical component. It is the optimal spatial pose state that the component should maintain during the machining process. The attitude angle deviation parameter is obtained by comparing the actual coordinates of the feature points on the surface of the component collected by the visual positioning sensor with the coordinates of the feature points of the ideal clamping posture, and then calculating it through the spatial posture solution algorithm. When the attitude angle deviation exceeds the preset threshold, it means that the component clamping state is significantly tilted and not in the ideal machining posture. This tilt will change the contact angle between the tool and the component, resulting in uneven distribution of cutting force, which in turn induces new residual stress and aggravates the release of the original residual stress, forming a two-way coupling between stress and clamping tilt.

[0046] When both judgment conditions are met, clamping tilt and residual stress form a vicious cycle of "two-way coupling". Clamping tilt leads to uneven distribution of cutting force and aggravates local stress concentration. The release of residual stress causes small deformation of components, further increasing the degree of clamping tilt. The effect of this coupling on machining accuracy is nonlinear and strongly correlated. Conventional simulations do not include coupling variables, and the generated paths will deviate significantly from the actual machining state. Direct use will lead to aggravated distortion and substandard accuracy. Therefore, this application triggers the dynamic simulation prediction process at this time. If the judgment criterion is not met, the clamping posture deviation can be ignored, and the residual stress acts in a dispersed and independent state. The effect on machining accuracy is linear and controllable. Conventional simulation can generate a path that meets the accuracy requirements, while avoiding the need to start multi-physics coupling simulation, which has higher computational costs, thus achieving the optimal balance between machining accuracy and simulation efficiency.

[0047] The dynamic simulation prediction process includes the following steps: Based on the judgment result of coupling dominance, dynamic simulation prediction technology is used to simulate the core extraction path. Dynamic simulation prediction technology is a real-time feedback simulation method based on digital twin models. Its core is to call the digital twin model of special optical components, using the identified residual stress distribution data and clamping tilt parameters such as attitude angle deviation as initial input conditions. Combined with the basic parameters of core extraction, such as tool speed, feed rate, and depth of cut, a "stress-clamping-machining" coupled simulation model is constructed. The dynamic release law of stress during material removal is simulated using finite element analysis. Simultaneously, the influence of clamping tilt on the machining trajectory is simulated by combining geometric constraint relationships. Finally, the phase fraction of the core extraction path under the coupling dominance state is iteratively calculated using fused data. The evolution trend and stress release trend of the fabric are analyzed through the following iterative process: First, an initial machining path is obtained. This initial machining path is an ideal machining path pre-planned and generated in the digital twin system based on the design drawings of special optical components, core extraction processing technical requirements, and historical process data. Its trajectory planning aims to achieve precise material removal and ensure optical performance, including the tool's starting point, ending point, the motion trajectory of each machining segment, and the corresponding cutting parameters. Based on the preset initial machining path, machining nodes are divided according to a preset step size, such as 0.1 mm. The residual stress release trend and phase distribution evolution trend after cutting are calculated node by node. The stress difference before and after cutting is obtained by solving the elastoplastic constitutive equation using the finite element method, and the stress release trend is traced. Finally, the normal displacement Δd generated by stress deformation is combined with... Calculate the phase deviation and analyze the phase evolution trend node by node. λ refers to the design working wavelength of the special optical component, that is, the incident light wavelength corresponding to the actual application scenario of the component, such as 532nm in the visible light band and 1064nm in the infrared band.

[0048] Next, dynamic data of the working conditions during the core extraction process are collected in real time. This dynamic data includes real-time cutting force of the tool, spindle vibration frequency, tool wear, and real-time temperature of the machining area. These data are collected in real time at preset frequencies by sensors integrated into the tool, spindle, and machining platform, and then transmitted to the simulation system after preprocessing. The dynamic data is then integrated into the simulation process, and the weights of various parameters are adjusted using a data calibration algorithm, such as weighted least squares. The parameters to be calibrated include residual stress weight, clamping tilt weight, and tool parameter weight. The core logic of the weight setting is to dynamically adjust the weights based on the deviation between the working condition data and the simulation results. For example, when the deviation between the real-time cutting force and the simulated cutting force exceeds a preset threshold, such as 10%, it indicates a significant change in the material cutting resistance in the residual stress concentration area, which is the dominant factor causing the deviation. In this case, the weight of the residual stress parameter is increased, for example, from an initial 0.3. Adjusting the value to 0.4 allows the simulation model to more accurately reflect the influence of stress on cutting force. At the same time, reducing the weight of tool parameters, for example, from the initial 0.4 to 0.3, can prevent excessive interference of tool parameters with the simulation results. When the spindle vibration frequency deviation exceeds the preset threshold, such as 5Hz, it means that the clamping tilt causes the coaxiality deviation between the components and the tool to exceed the standard, resulting in periodic impact vibration during the cutting process. At this time, it is necessary to increase the weight of the clamping tilt parameter, for example, from the initial 0.2 to 0.3, which can strengthen the influence mechanism of clamping posture on vibration and make the simulation model more accurately match the actual vibration state. After the parameter weights are adjusted, the calculation equations of the "stress-clamping-machining" coupled simulation model are reconstructed based on the adjusted weights. The adjusted weights are used as the influence coefficients of each parameter and integrated into the calculation process of core simulation indicators such as cutting force, stress release, and phase deviation. Simulation calculations are then performed again for each node of the entire machining path: For each machining node, the combined influence of residual stress, clamping tilt, and tool parameters on the cutting state of the node is calculated using the adjusted weights. The simulated cutting force value, stress release amount, and phase deviation value of the node are then re-output, thereby generating the corrected node coordinates and cutting parameters. After all nodes are recalculated, they are integrated to form the corrected complete core-taking machining path. At the same time, the deviation between the simulation indicators of each node and the real-time working condition data is checked. If the deviation is not controlled within the preset range, such as 5%, the above "weight adjustment-re-simulation calculation" process is repeated until the deviation between the simulation results and the actual working conditions meets the preset requirements, thus completing the correction of the core-taking path simulation results.

[0049] After the simulation results are corrected, the corrected simulation path is compared with the preset core extraction path benchmark. The preset core extraction path benchmark is an ideal machining path generated based on the design target model, including standard data such as the tool's motion trajectory and cutting parameters of each node. The comparison process uses the trajectory overlap analysis method. First, the coordinate data and cutting parameter data of each corresponding machining node in the corrected simulation path and the path benchmark are extracted, and then the results are compared using the formula... Calculate the spatial coordinate deviation values ​​of each node, where, For the coordinates of the simulated path nodes, Using the coordinates of the reference path nodes, the deviation percentage of cutting parameters is statistically analyzed. Assuming a preset coordinate deviation of 0.01mm and a parameter deviation percentage threshold of 5%, if the calculated path deviation value exceeds the preset threshold, it indicates that the current corrected simulation path still cannot meet the machining accuracy requirements. At this point, a core extraction process path adjustment scheme is generated based on the deviation data of each node. The scheme specifically includes path correction parameters for each deviation node, such as coordinate compensation value, cutting speed adjustment amount, and feed rate adjustment amount; stress release control parameters, clamping optimization parameters, and key areas for working condition monitoring after path adjustment; and clarifies the specific values ​​and implementation steps of each adjustment parameter to complete the determination of the core extraction process path adjustment scheme.

[0050] Through the above steps, accurate prediction and dynamic optimization of the processing path based on distortion cause were achieved. In particular, a special simulation model was built for the coupling effect of residual stress and clamping tilt, which effectively solved the problems of mismatch between processing path and actual processing state and large blindness in path optimization. It provides a more adaptable path scheme for accurate processing and precision control in subsequent steps, and further ensures the core-taking processing quality of special optical components.

[0051] Step S4: Based on the core extraction process path adjustment plan, quantify the influence weight of each core extraction parameter on the core extraction accuracy, adjust the stress release control parameters, and determine whether the adjusted core extraction accuracy exceeds the preset range. If it does, generate an optimized core extraction execution plan.

[0052] Step S4 includes: extracting core extraction parameters related to core extraction accuracy from the core extraction process path adjustment scheme; combining real-time operating data during the core extraction process; using a weighted regression algorithm to calculate the influence weight of each core extraction parameter on the core extraction accuracy; generating a parameter influence weight mapping map; inputting the parameter influence weight mapping map into a closed-loop feedback control mechanism; using a proportional-integral-derivative control algorithm; dynamically adjusting the stress release control parameters based on the real-time deviation of the core extraction accuracy; synchronously collecting the adjusted stress distribution data and core extraction accuracy data; feeding the data back to the control algorithm for iterative correction; comparing the real-time monitored core extraction accuracy data with a preset threshold to determine whether the core extraction accuracy exceeds the allowable range; if the core extraction accuracy exceeds the preset range, then integrating the real-time operating data and accuracy deviation data to optimize and correct the core extraction process path adjustment scheme, ultimately generating an optimized core extraction execution scheme that can be directly executed.

[0053] Specifically, after determining the core extraction process path adjustment scheme, the next step is to optimize the closed-loop control and execution scheme for core extraction machining accuracy. The core objective of this step is to transform the generated path adjustment scheme into an optimized core extraction execution scheme that can be directly executed through parameter weight analysis, closed-loop feedback control, and accuracy verification optimization, ensuring that the core extraction machining accuracy of special optical components meets the standards. First, core extraction parameters are extracted and their influence weights on core extraction accuracy are calculated, generating a parameter influence weight mapping diagram. The specific operation is as follows: From the core extraction process path adjustment scheme generated in step S3, core core extraction parameters directly related to core extraction machining accuracy are selected and extracted. Core extraction parameters include path correction parameters, such as X / Y / Z axis coordinate compensation values, cutting speed adjustment, and feed rate adjustment; stress release control parameters, such as cutting depth gradient adjustment values ​​and segmented machining interval time; and clamping optimization parameters, such as clamping pressure adjustment values ​​and attitude angle compensation values. Subsequently, a data acquisition and analysis system was built to synchronously collect real-time operating condition data during the core extraction process. Utilizing the sensor acquisition system from step S3, core extraction accuracy detection data was supplemented, with the acquisition frequency maintained at 100Hz, consistent with the operating condition data. The core extraction accuracy detection data was acquired jointly using a laser interferometer and a coordinate measuring machine, specifically including the flatness and contour of the component's processed surface, and the phase deviation value of the core optical surface. Based on the collected real-time operating condition data and core extraction accuracy data, a weighted regression algorithm was used to construct a parameter-accuracy correlation model. The core extraction accuracy deviation value was used as the dependent variable, and each core core extraction parameter was used as the independent variable. The algorithm iteratively calculated the influence weight of each independent variable on the dependent variable. The algorithm presets a priority coefficient for accuracy influence, assigning higher weights (e.g., 0.6) to key accuracy indicators such as flatness and phase deviation, and secondary weights (e.g., 0.4) to auxiliary accuracy indicators. Finally, the accuracy influence weight value corresponding to each core extraction parameter was output. Import the parameter names, parameter value ranges, and corresponding accuracy impact weight values ​​into the data visualization module to generate a parameter impact weight mapping chart. This mapping chart is presented in two-dimensional coordinate form, with the horizontal axis representing the core extraction parameter type and the vertical axis representing the impact weight value. The weight ranking and key impact thresholds of each parameter are marked, intuitively reflecting the control priority of different core extraction parameters on core extraction accuracy.

[0054] Next, a closed-loop feedback control mechanism is constructed to dynamically adjust the stress release control parameters. The parameter influence weight mapping diagram generated above is imported into a preset closed-loop feedback control module. This module has a built-in proportional-integral-derivative (PID) control algorithm. The algorithm parameters are first initialized, and the control cycle is synchronized with the data acquisition cycle. The core sampling accuracy target value is determined, such as flatness error ≤ 0.005 mm and phase deviation ≤ 0.02 rad as a benchmark. The real-time deviation value of the core sampling accuracy is calculated, that is, the difference between the real-time detection value and the target value. This deviation value is used as the input of the PID control algorithm. The algorithm, based on a parameter influence weight mapping, prioritizes the top three stress release control parameters by weight, such as cutting speed adjustment, feed rate adjustment, and depth of cut gradient adjustment, as the control targets. It dynamically outputs parameter adjustment commands based on real-time deviation values: when the real-time accuracy deviation is positive and exceeds a preset warning threshold (e.g., 50% of the target value), the algorithm increases the adjustment amplitude of the stress release control parameter through a proportional loop, eliminates accumulated deviation through an integral loop, and predicts the deviation change trend through a derivative loop, adjusting the parameter change rate in advance; when the deviation value approaches the target value, the algorithm automatically reduces the adjustment amplitude to avoid parameter overshoot. During parameter adjustment, stress sensors collect real-time stress distribution data of the adjusted components, and a precision detection device collects core sampling accuracy data. These two sets of data are fed back to the closed-loop feedback control module in real-time as the basis for iterative correction of the PID control algorithm. The algorithm recalibrates the adjustment step size of each control parameter based on the feedback data, achieving a closed-loop cycle of "deviation detection - parameter adjustment - data feedback - iterative correction," ensuring coordinated optimization of stress release state and core sampling accuracy.

[0055] Finally, the core extraction accuracy verification and path adjustment scheme optimization are performed to generate the final execution scheme. The real-time monitored core extraction accuracy data is compared with the preset allowable accuracy range. If the real-time core extraction accuracy data is within the allowable range, it indicates that the current core extraction process path adjustment scheme meets the processing requirements. This is directly used as the basic framework, supplemented with auxiliary parameters required for processing execution, such as tool change intervals, cleaning timing of the processing area, and detection node settings, to form the initial core extraction execution scheme. If the real-time core extraction accuracy data exceeds the allowable range, the scheme optimization and correction process is initiated: first, the specific indicators exceeding the accuracy standard and the corresponding processing nodes are extracted; then, the real-time working condition data and accuracy deviation data of the node are integrated to pinpoint the cause of the deviation. If it is due to insufficient stress release, the cutting depth gradient adjustment value of the corresponding node is further increased or the processing interval time is extended. If it is due to clamping posture fluctuations, the clamping pressure and posture angle compensation values ​​are adjusted. If it is due to insufficient matching of cutting parameters, the values ​​of the corresponding parameters are fine-tuned based on the parameter influence weight mapping diagram. After correction, a new optimized core extraction process path adjustment plan is generated, and accuracy is verified again through a closed-loop feedback control mechanism until the real-time core extraction accuracy data fully meets the preset allowable range. Finally, the optimized path parameters, control parameters, auxiliary parameters, and accuracy testing standards are integrated to form an optimized core extraction execution plan that can be directly executed. The plan adopts a standardized process document format, clearly defining the parameter values, execution steps, testing requirements, and anomaly handling measures for each processing stage, ensuring that the processing process is traceable and controllable.

[0056] Among them, the stress release control parameters are dynamically adjusted based on the real-time deviation of the core extraction accuracy, including: the real-time deviation of the core extraction accuracy includes size deviation components and phase deviation components; the difference between the two types of deviation components and the preset threshold is calculated to determine the deviation priority; the corresponding stress release control parameter adjustment strategy is matched according to the deviation priority, and the adjustment amount of the output stress release control parameter is calculated through the proportional-integral-derivative control algorithm.

[0057] Specifically, the stress release control parameters for real-time deviation dynamic adjustment based on core-taking accuracy include the following steps: First, clarify the definition standards for the two core components of the real-time deviation of core-taking accuracy: Dimensional deviation refers to the deviation between the actual geometric dimensions of the component and the preset accuracy target value during the core-taking process, mainly covering geometric accuracy indicators such as flatness deviation, contour deviation, and critical dimension tolerance deviation; Phase deviation refers to the deviation between the actual phase distribution of the core optical surface of the component and the preset phase target value, directly affecting optical transmission and imaging performance. Then, a dual-path parallel detection system is built to acquire the two types of deviation data: Dimensional deviation is acquired using a high-precision coordinate measuring machine at a frequency of 100Hz. After filtering and noise reduction preprocessing, the real-time detection values ​​of each geometric indicator are extracted; Phase deviation is acquired using a laser interferometer, synchronously matched to a 100Hz acquisition frequency, to obtain the real-time phase data of each detection point on the core optical surface. Subsequently, the differences between the two types of deviation components and preset thresholds are calculated. Based on the precision requirements of core machining, corresponding preset thresholds for the two types of deviation components are pre-set. For example, the preset thresholds for the dimensional deviation component are: flatness ≤ 0.005 mm and contour ≤ 0.008 mm; the preset threshold for the phase deviation component is: core optical surface phase deviation ≤ 0.02 rad. For the collected real-time deviation data, the differences between the two types of deviation components and their corresponding preset thresholds are calculated respectively. After the difference calculation is completed, the positive and negative attributes of the two types of deviations are marked. Positive values ​​represent deviations exceeding the threshold, and negative values ​​represent deviations within the threshold range.

[0058] Next, deviation priority is determined, following the core principle of "optical performance first," and the determination rules are set based on the absolute value of the deviation difference: when |phase deviation threshold| ≥ 0.01 rad, i.e. 50% of the phase deviation threshold, regardless of the size deviation difference, the phase deviation is determined to be of first priority and needs to be controlled first; when |size deviation threshold| ≥ 0.0025 mm, i.e. 50% of the size deviation threshold and |phase deviation threshold| < 0.01 rad, the size deviation is determined to be of first priority; when the absolute value of the difference between the two types of deviations is lower than 50% of the corresponding threshold, the phase deviation is of second priority and the size deviation is of third priority, and weak control is implemented; when only one type of deviation difference is positive, the deviation is of first priority, and the other type is used as an auxiliary constraint.

[0059] Subsequently, the stress release control parameters are adjusted according to the deviation priority: if the phase deviation is of primary priority, a cutting speed-feed rate coordinated reduction strategy is adopted, with core control of cutting speed and feed rate, supplemented by fine-tuning of the cutting depth gradient to a preset threshold, such as 0.015mm / segment, to smoothly release stress and avoid aggravation of phase distortion; if the dimensional deviation is of primary priority, a cutting depth-clamping posture linkage strategy is adopted, with core control of cutting depth gradient and posture angle compensation value, supplemented by fine-tuning of clamping pressure to ensure consistency between the machining trajectory and the reference to correct geometric deviations. Finally, the output adjustment amount is calculated through a proportional-integral-derivative control algorithm. PID parameters are initialized based on the deviation type. PID parameters include Kp, Ki, and Kd. Kp is the proportional coefficient, used to respond to immediate changes in deviation, outputting an adjustment amount proportional to the deviation to ensure rapid control. Ki is the integral coefficient, used to accumulate deviation and eliminate static error, solving the problem that proportional control cannot completely eliminate residual error. Kd is the derivative coefficient, used to predict the trend of deviation changes and output a reverse adjustment amount in advance to suppress parameter overshoot and improve system stability. For phase deviation control, for example, Kp=0.9, Ki=0.2, Kd=0.1; for size deviation control, Kp=0.8, Ki=0.3, Kd=0.1. The control cycle is synchronized with the acquisition frequency at 10ms. The first-level priority deviation difference is the core input, and the second level is the auxiliary constraint, using the core PID formula: ,in, The adjustment amount of the output control parameter. The first-level priority deviation difference for core input. This is the integral term of the deviation difference, reflecting the cumulative deviation. This is the differential term of the deviation difference, reflecting the rate of change of the deviation. During the calculation process, it is simultaneously checked whether the adjustment amount exceeds the preset maximum range. If it does, the maximum limit is taken as the output value. At the same time, a second check is performed in conjunction with the secondary priority deviation difference of the auxiliary constraint to ensure that the adjustment will not cause the secondary deviation to exceed the threshold. Finally, the qualified adjustment amount, including the specific value and adjustment direction, is transmitted to the machine tool actuator to complete the dynamic adjustment of the stress release control parameters.

[0060] By precisely breaking down deviations and implementing graded control, the precision control is effectively improved; with differentiated strategies, stress is released smoothly, thereby reducing processing distortion; the PID algorithm ensures stable and precise control, reducing the risk of overshoot; and a closed-loop controllable mechanism is formed throughout the entire process, which is not only easy to promote in engineering, but also effectively improves yield and reduces costs.

[0061] Step S5: Based on the core extraction execution scheme, track the dynamic evolution record of the processing process, analyze the real-time decay trend of multi-source distortion causes, and obtain the geometric residual error distribution after processing.

[0062] Step S5 includes: performing core extraction based on the optimized core extraction execution scheme, simultaneously collecting time-series data of the entire core extraction process, including phase distribution data, clamping posture data, cutting force data, and environmental parameter data; extracting feature data corresponding to multi-source distortion causes from the time-series data, calculating the attenuation rate of each distortion cause, and generating a multi-source distortion attenuation trend curve; and calculating the difference between the actual phase value and the reference phase value at each moment based on the core extraction geometric reference and the phase distribution data in the time-series data, thereby generating the geometric residual error distribution after processing.

[0063] Specifically, after generating the core extraction execution plan, to verify its practical applicability, the dynamic evolution of multi-source distortions such as stress release and clamping posture fluctuations during the machining process is further precisely captured. The real-time attenuation trends of various distortion causes are clarified, and the geometric residual error distribution after machining is ultimately quantified. This provides reliable data support for subsequent machining quality assessment, process optimization iteration, and reuse of machining experience for similar products. First, core extraction machining is carried out based on the optimized core extraction execution plan, and a time-series data acquisition system is built simultaneously to achieve continuous data acquisition throughout the machining process. The acquisition system integrates multiple types of sensors and detection equipment. The specific time-series data acquired includes: phase distribution data of the core optical surface, clamping posture data, cutting force data, and environmental parameter data. All acquired data is timestamped and stored in a real-time database to ensure temporal consistency and traceability. Subsequently, feature data corresponding to multi-source distortion causes are extracted from the time-series data, the attenuation rate of each distortion cause is calculated, and attenuation trend curves are generated. First, the scope of the causes of multi-source distortion is defined, including three core causes: stress release distortion, clamping posture distortion, and environmental interference distortion. Corresponding feature data are extracted: for stress release distortion, the feature data selected are the amplitude of cutting force fluctuation and the change in phase deviation; for clamping posture distortion, the feature data selected is the peak value of the posture angle deviation; and for environmental interference distortion, the feature data selected is the temperature fluctuation difference. Based on the extracted feature data, the attenuation rate is calculated using the following formula: ,in, The attenuation rate, The peak value of the characteristic data at the initial moment of processing. For a given characteristic data value at a specific moment, the decay rate of three types of distortion causes is calculated time-by-time. The decay rate data at each moment is imported into the data visualization module, and a multi-source distortion decay trend curve is generated with processing time as the horizontal axis and decay rate as the vertical axis. This curve visually presents the decay pattern of various distortion causes as the processing progresses. If the decay rate of a certain type of distortion is consistently lower than a preset threshold, it is marked as insufficient decay, and the cause needs to be traced later.

[0064] Finally, combining the core extraction processing geometric datum and phase distribution data, the geometric residual error distribution is calculated. Using the core extraction processing geometric datum as a reference, the phase distribution data at the processing completion time in the time-series data is extracted as the actual phase value. Simultaneously, the datum phase value is retrieved from the design target model corresponding to the core extraction execution scheme. For each detection point on the core optical surface, the phase residual error value at each point is calculated, where positive values ​​represent positive residual errors and negative values ​​represent negative residual errors. The phase residual error values ​​of all detection points are associated with their corresponding spatial coordinates and imported into the error distribution cloud map generation module. Using a color gradient mapping method, a geometric residual error distribution cloud map after processing is generated, as shown below. Figure 3 The diagram shown is a schematic of the geometric residual error distribution cloud map. This diagram intuitively shows the spatial distribution of surface phase residual errors after the core processing of special optical components. The distribution pattern of high-incidence areas of errors can be clearly identified from the cloud map, and the error value range can be quantified into specific values. This geometric residual error distribution cloud map provides a key quantitative basis for subsequent accuracy verification, process optimization, and control strategy adjustment.

[0065] By tracking the dynamic evolution of the entire processing, analyzing the multi-source distortion attenuation law and quantifying the geometric residual error distribution, the adaptability of the core extraction execution scheme can be effectively verified, providing an intuitive basis for processing quality assessment. At the same time, it supports the iterative optimization of process schemes and the reuse of processing experience of similar products, thereby improving processing stability and yield, and reducing production costs.

[0066] Step S6: If the geometric residual error distribution does not meet the preset accuracy standard, adjust the assembly and clamping tilt compensation parameters, update the digital twin model, and determine the final core extraction control strategy.

[0067] Step S6 includes: locating the non-compliant areas and corresponding dominant error sources based on the geometric residual error distribution; adapting and optimizing the clamping posture parameters by combining the multi-source distortion attenuation trend curve; the clamping posture parameters include clamping angle compensation values ​​and clamping force distribution parameters; inputting the optimized clamping posture parameters into the digital twin model to update the model; conducting core extraction process simulation verification based on the updated digital twin model until the core extraction residual error meets the standard; integrating the optimized clamping parameters, core extraction execution scheme, and model calibration parameters to determine the final core extraction control strategy.

[0068] Specifically, after generating the geometric residual error distribution cloud map, to accurately rectify situations where residual errors do not meet the preset accuracy standards, further optimization of clamping parameters, updating of the digital twin model, and simulation verification are conducted to determine a more adaptable final core extraction control strategy, ensuring that the final component processing accuracy meets the standards. First, the areas with substandard positioning accuracy and the dominant error sources are identified. Based on the geometric residual error distribution cloud map generated in step S5, areas with residual error values ​​exceeding the preset accuracy standards are selected, marked as substandard areas, and their spatial coordinate range is recorded. Simultaneously, combining the obtained multi-source distortion attenuation trend curves, the attenuation rate data of various distortion causes during the processing period corresponding to the substandard areas are compared. If the attenuation rate of a certain type of distortion cause is lower than a preset threshold (e.g., 30%) during that period, and the peak value of the characteristic data fluctuation is positively correlated with the error value of the substandard area, then this distortion is determined to be the dominant error source of the substandard area. Subsequently, the clamping posture parameters are adapted and optimized. Based on the identified non-compliant areas and dominant error sources, targeted optimization is carried out focusing on clamping attitude parameters: If the dominant error source is clamping attitude distortion, clamping angle compensation values ​​are calculated based on the spatial position deviation of the non-compliant areas, specifically including fine-tuning compensation for pitch and yaw angles; simultaneously, based on the correlation analysis between clamping force distribution and error distribution, clamping force distribution parameters are optimized by adjusting the pressure values ​​at each clamping point within a range of 0.05-0.1 MPa to avoid residual stress errors caused by excessive local clamping force. During the optimization process, the multi-source distortion attenuation law in step S5 is referenced to ensure that the parameter adjustment direction matches the distortion attenuation trend, thereby improving the effectiveness of parameter optimization.

[0069] Next, the digital twin model is updated and simulation verification is conducted. The optimized clamping posture parameters, namely the clamping angle compensation value and clamping force distribution parameters, are input into the preset core extraction digital twin model. Simultaneously, the time-series data from step S5, the geometric residual error distribution, and other actual machining data are imported. The parameters of core modules such as the clamping module and stress relief module are calibrated to complete the model update. Based on the updated digital twin model, the core extraction process of step S5 is reproduced, and the residual error of core extraction is simulated to obtain the simulated residual error distribution. The simulated residual error distribution is compared with the preset accuracy standard. If it does not meet the standard, the process of "fine-tuning clamping posture parameters - updating the model - simulation verification" is repeated; if the simulated residual error meets the standard, it indicates that the optimized parameters and the model have good compatibility. Finally, the final core extraction control strategy is determined. By integrating the optimized clamping posture parameters, the core extraction execution scheme generated in step S4, and the model calibration parameters generated during the digital twin model update process, a complete final core extraction control strategy is formed. This strategy clarifies the clamping parameter values, core extraction execution steps, model calibration standards, and error monitoring thresholds throughout the entire machining process, ensuring that the actual machining process can be accurately controlled. At the same time, the rectification experience and parameter optimization logic of non-compliant areas are included in the strategy appendix to provide a reference for subsequent similar machining scenarios.

[0070] Step S7: Based on the final core extraction control strategy, acquire real-time evolution record data during the core extraction process, integrate stress release control effect and working condition dynamic data, and output the final core extraction processing control result.

[0071] Step S7 includes: executing core extraction processing according to the final core extraction control strategy, and simultaneously acquiring real-time evolution record data of the core extraction process. The real-time evolution record data includes phase distribution timing data, clamping posture dynamic data, and real-time core extraction dimension monitoring data; collecting stress release control effect data and working condition dynamic data, using a Kalman filter fusion algorithm to weight and integrate the real-time evolution record data, stress release control effect data, and working condition dynamic data, calculating the joint influence factor of multi-source data on core extraction accuracy, verifying whether the core extraction accuracy meets the preset benchmark, and outputting the final core extraction processing control result of the special optical component.

[0072] Specifically, after determining the final core extraction control strategy, to accurately execute the core extraction process and comprehensively verify the processing effect, ensuring that the output of special optical components meets the accuracy requirements, real-time evolution data of the processing process is acquired according to the final core extraction control strategy. This data is then integrated with the control effect and operating condition data to complete accuracy verification and output the final control result. First, core extraction is executed while simultaneously acquiring real-time evolution data. The core extraction operation of the special optical components is carried out strictly according to the final core extraction control strategy, while a preset data acquisition system is activated to achieve continuous acquisition of real-time evolution data throughout the entire processing process. The real-time evolution recording data focuses on three core types of data: First, phase distribution time-series data, which acquires dynamic phase values ​​of each detection point on the core optical surface using a laser interferometer at a 10ms acquisition frequency, reflecting the real-time status related to optical performance; second, clamping attitude dynamic data, which uses a spindle attitude sensor to collect pitch, yaw, and roll angle changes throughout the machining process, ensuring that the clamping status meets strategy requirements; and third, core dimension real-time monitoring data, which uses a high-precision coordinate measuring machine to dynamically acquire key geometric dimension data, tracking real-time changes in dimensional accuracy. All acquired data carries precise timestamps and is stored in a dedicated database to ensure temporal consistency and traceability. Subsequently, multi-source data is collected and weighted and integrated using a Kalman filter fusion algorithm. On the one hand, stress release control effect data is collected, which is obtained by comparing stress distribution data at each stage of machining with preset stress thresholds, reflecting the actual effectiveness of stress release control measures; on the other hand, dynamic operating condition data is collected, including real-time data on cutting force, cutting temperature, temperature and humidity of the machining area, and air pressure, comprehensively capturing the machining environment and equipment operating status. The aforementioned real-time evolution record data, stress release control effect data, and operating condition dynamic data are imported into the data fusion module. A Kalman filter fusion algorithm is used for weighted integration. Based on the weights of various data types on core extraction accuracy (e.g., phase distribution time-series data weighted at 0.4, real-time core extraction size monitoring data weighted at 0.3, and other data weighted at a total of 0.3), data noise interference is eliminated through the algorithm to obtain standardized fused data. Finally, the joint influence factor is calculated, accuracy is verified, and control results are output. Based on the fused data, a correlation model between multi-source data and core extraction accuracy is constructed, and the joint influence factor of multi-source data on core extraction accuracy is calculated. This factor quantitatively reflects the comprehensive influence of various data types on the final accuracy under synergistic effects. The actual core extraction accuracy data after processing is compared and verified with a preset benchmark. If the actual core extraction accuracy meets the benchmark, the final core extraction control strategy is effective. The evolution record data, fused data, joint influence factor, and accuracy verification results throughout the processing are integrated, and the final core extraction processing control result for the special optical components is output. If the benchmark is not met, the process returns to step S6 to re-optimize parameters and control strategies.

[0073] By synchronously collecting real-time data throughout the entire processing process and fusing it using Kalman filtering, the reliability and completeness of the data are effectively improved. Combined with the quantitative analysis of the joint impact factor, the comprehensive effect of multi-source data on accuracy is analyzed, providing a scientific basis for accuracy verification. The effectiveness of the final core extraction control strategy is ensured through a closed-loop verification mechanism, guaranteeing that the processing accuracy meets the standards. At the same time, the integrated full-process data can provide detailed references for subsequent process optimization and processing of similar products, improving processing stability and efficiency.

[0074] The above describes a digital twin control method for core extraction of special optical components in the embodiments of this application. The following describes a digital twin control system for core extraction of special optical components in the embodiments of this application. Please refer to [link to relevant documentation]. Figure 4 One embodiment of a digital twin control system for core extraction of special optical components in this application includes:

[0075] The module is used to collect geometric topography data of the surface to be processed of special optical components through multi-source sensors, construct a digital twin model of the special optical components, and determine the geometric reference for core extraction processing.

[0076] The classification module is used to identify distortion features in the core extraction process based on the geometric reference of the core extraction process, integrate environmental data to analyze the residual stress distribution in the core extraction process, and generate distortion feature classification labels.

[0077] The adjustment module is used to identify the cause of distortion based on the distortion feature classification label, and to predict the core extraction processing path using dynamic simulation prediction technology. It integrates dynamic working condition data to dynamically adjust the core extraction processing path and determine the core extraction process path adjustment scheme.

[0078] The optimization module is used to adjust the scheme according to the core extraction process path, quantify the influence weight of each core extraction parameter on the core extraction accuracy, adjust the stress release control parameters, and determine whether the adjusted core extraction accuracy exceeds the preset range. If it does, an optimized core extraction execution scheme is generated.

[0079] The analysis module is used to track the dynamic evolution record of the processing according to the core extraction execution scheme, analyze the real-time decay trend of multi-source distortion causes, and obtain the geometric residual error distribution after processing.

[0080] The update module is used to adjust the assembly and clamping tilt compensation parameters, update the digital twin model, and determine the final core extraction control strategy if the geometric residual error distribution does not meet the preset accuracy standard.

[0081] The output module is used to acquire real-time evolution record data during the core extraction process based on the final core extraction control strategy, integrate stress release control effect and working condition dynamic data, and output the final core extraction processing control result.

[0082] In summary, the digital twin control method and system for core taking of special optical components disclosed in this application constructs a closed-loop control system for the entire process based on digital twin technology. It constructs a twin model and determines the processing benchmark by collecting data from multiple sources of sensors, accurately identifies distortion characteristics and correlates their causes, and dynamically optimizes the processing path to address the coupling effect of residual stress and clamping tilt. By combining algorithms such as weighted regression and PID control, it quantifies parameter weights and regulates stress release, tracks the distortion attenuation trend to obtain the residual error distribution, iteratively optimizes clamping parameters and the model to determine the final control strategy, and finally verifies the accuracy and outputs the results through Kalman filtering and data fusion. This solution effectively solves the pain points of traditional core taking processing, such as difficulty in controlling processing accuracy and large multi-source distortion interference, ensuring that processing accuracy and optical performance meet standards, providing reliable support for subsequent process optimization, and possessing significant engineering application value.

[0083] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0084] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0085] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A digital twin control method for core extraction of special optical components, characterized in that, The method includes: Step S1: Collect geometric topography data of the surface to be processed of the special optical component using multi-source sensors, construct a digital twin model of the special optical component, and determine the geometric reference for core processing; Step S2: Based on the core extraction processing geometric datum, identify the distortion features during the processing, integrate environmental data to analyze the residual stress distribution during the core extraction process, and generate distortion feature classification labels; Step S3: Identify the cause of distortion based on the distortion feature classification identifier, and use dynamic simulation prediction technology to predict the core extraction processing path. Combine dynamic working condition data to dynamically adjust the core extraction processing path and determine the core extraction process path adjustment scheme. Step S4: According to the core extraction process path adjustment scheme, quantify the influence weight of each core extraction parameter on the core extraction accuracy, adjust the stress release control parameters, and determine whether the adjusted core extraction accuracy exceeds the preset range. If it does, generate an optimized core extraction execution scheme. Step S5: Based on the core extraction execution scheme, track the dynamic evolution record of the processing process, analyze the real-time decay trend of multi-source distortion causes, and obtain the geometric residual error distribution after processing; Step S6: If the geometric residual error distribution does not meet the preset accuracy standard, adjust the assembly and clamping tilt compensation parameters, update the digital twin model, and determine the final core extraction control strategy. Step S7: Based on the final core extraction control strategy, acquire real-time evolution record data during the core extraction process, integrate stress release control effect and working condition dynamic data, and output the final core extraction processing control result.

2. The digital twin control method for core extraction of special optical components according to claim 1, characterized in that, Step S1 includes: Geometric topography data, core-taking tool condition data, and clamping and positioning data of the surface to be processed of special optical components are collected by multi-source sensors and integrated to form complete core-taking physical state data. Based on the core-taking physical state data, a digital twin model of the special optical components is constructed using virtual modeling mapping technology, and the collected data and model state are synchronized in real time. The synchronized digital twin model is compared with the preset design target model to determine the geometric reference for core-taking processing.

3. The digital twin control method for core extraction of special optical components according to claim 1, characterized in that, Step S2 includes: A preliminary phase distribution reference map is generated based on the core processing geometric reference. The phase distribution reference map is scanned using a phase gradient calculation method to calculate the phase change rate of each pixel. If the phase change rate exceeds a preset threshold, the corresponding pixel is defined as a distortion feature and a distortion key region is generated. Environmental data is collected during the core extraction process. The information of the key distortion area is fused with the environmental data. The stress tensor is calculated on the fused data to obtain a stress heat map of the residual stress distribution. Based on the stress heat map, a mapping relationship between the distortion features and the distortion causes is established. Based on the mapping relationship, the distortion features are divided into multiple categories. An identification code is assigned to each category of distortion features to generate a distortion feature classification identifier that represents the association between the distortion features and their causes.

4. The digital twin control method for core extraction of special optical components according to claim 1, characterized in that, Step S3 includes: Based on the distortion feature classification identifier, the distortion cause is identified. Based on the distortion cause, it is determined whether the dominant influencing factor of the core taking process is the coupling effect of residual stress and clamping tilt. If so, the dynamic simulation prediction process is triggered. According to the determination result of coupling dominance, the core taking process path is simulated using dynamic simulation prediction technology. Combined with fused data, the phase distribution evolution trend and stress release trend of the core taking process path under the coupling dominance state are iteratively calculated. Real-time acquisition of dynamic operating data during the core extraction process is performed and integrated into the simulation process. The weights of various parameters in the simulation process are adjusted through a data calibration algorithm to correct the simulation results of the core extraction path. The corrected simulation path is compared with the preset core extraction path benchmark, and the path deviation value is calculated. If the path deviation value exceeds the preset threshold, a core extraction process path adjustment scheme is generated.

5. The digital twin control method for core extraction of special optical components according to claim 1, characterized in that, Step S4 includes: Extract the core extraction parameters related to the core extraction processing accuracy from the core extraction process path adjustment scheme, combine them with the real-time working condition data during the core extraction process, and use a weighted regression algorithm to calculate the influence weight of each core extraction parameter on the core extraction accuracy, and generate a parameter influence weight mapping diagram. The parameter influence weight mapping diagram is input into the closed-loop feedback control mechanism. A proportional-integral-derivative control algorithm is adopted. The stress release control parameters are dynamically adjusted based on the real-time deviation of the core sampling accuracy. The adjusted stress distribution data and core sampling accuracy data are collected simultaneously and fed back to the control algorithm for iterative correction. The real-time core extraction accuracy data is compared with a preset threshold to determine whether the core extraction accuracy exceeds the allowable range. If the core extraction accuracy exceeds the preset range, the real-time operating data and accuracy deviation data are integrated to optimize and correct the core extraction process path adjustment scheme, and finally generate an optimized core extraction execution scheme that can be directly executed.

6. The digital twin control method for core extraction of special optical components according to claim 5, characterized in that, Based on the core extraction accuracy, real-time deviation dynamic adjustment of stress release control parameters includes: The real-time deviation of the core extraction accuracy includes dimensional deviation components and phase deviation components. The difference between the two types of deviation components and the preset threshold is calculated to determine the deviation priority. According to the deviation priority, the corresponding stress release control parameter adjustment strategy is matched, and the adjustment amount of the output stress release control parameter is calculated by the proportional-integral-derivative control algorithm.

7. The digital twin control method for core extraction of special optical components according to claim 1, characterized in that, Step S5 includes: Core extraction is performed based on the optimized core extraction execution scheme. Time-series data of the entire core extraction process is collected simultaneously. The time-series data includes phase distribution data, clamping posture data, cutting force data and environmental parameter data. Feature data corresponding to multi-source distortion causes are extracted from the time-series data, the attenuation rate of each distortion cause is calculated, and a multi-source distortion attenuation trend curve is generated. Based on the geometric reference of the core extraction process and combined with the phase distribution data in the time-series data, the difference between the actual phase value and the reference phase value at each moment is calculated to generate the geometric residual error distribution after processing.

8. The digital twin control method for core extraction of special optical components according to claim 1, characterized in that, Step S6 includes: Based on the geometric residual error distribution, the substandard areas and the corresponding dominant error sources are located. Combined with the multi-source distortion attenuation trend curve, the clamping posture parameters are adapted and optimized. The clamping posture parameters include clamping angle compensation value and clamping force distribution parameters. The optimized clamping posture parameters are input into the digital twin model to update the model. The core extraction process is then simulated and verified based on the updated digital twin model until the residual error of core extraction meets the standard. Finally, the optimized clamping parameters, core extraction execution scheme and model calibration parameters are integrated to determine the final core extraction control strategy.

9. A digital twin control method for core extraction of special optical components according to claim 1, characterized in that, Step S7 includes: Core removal processing is performed according to the final core removal control strategy, and real-time evolution record data of the core removal process is acquired simultaneously. The real-time evolution record data includes phase distribution timing data, clamping posture dynamic data, and real-time core removal dimension monitoring data. Collect stress release control effect data and working condition dynamic data, and use Kalman filter fusion algorithm to weight and integrate the real-time evolution record data, stress release control effect data and working condition dynamic data to calculate the joint influence factor of multi-source data on core extraction accuracy, verify whether the core extraction accuracy meets the preset benchmark, and output the final core extraction processing control result of special optical components.

10. A digital twin control system for core extraction of special optical components, used to implement the digital twin control method for core extraction of special optical components as described in any one of claims 1-9, characterized in that, The system includes: The module is used to collect geometric topography data of the surface to be processed of special optical components through multi-source sensors, construct a digital twin model of the special optical components, and determine the geometric reference for core processing. The classification module is used to identify distortion features during the core extraction process based on the core extraction geometric reference, analyze the residual stress distribution during the core extraction process by integrating environmental data, and generate distortion feature classification labels. The adjustment module is used to identify the cause of distortion based on the distortion feature classification identifier, and to predict the core extraction processing path using dynamic simulation prediction technology. It also integrates dynamic working condition data to dynamically adjust the core extraction processing path and determine the core extraction process path adjustment scheme. The optimization module is used to quantify the influence weight of each core extraction parameter on the core extraction accuracy according to the core extraction process path adjustment scheme, adjust the stress release control parameters, and determine whether the adjusted core extraction accuracy exceeds the preset range. If it does, an optimized core extraction execution scheme is generated. The analysis module is used to track the dynamic evolution record of the processing process according to the core extraction execution scheme, analyze the real-time decay trend of multi-source distortion causes, and obtain the geometric residual error distribution after processing. The update module is used to adjust the assembly and clamping tilt compensation parameters, update the digital twin model, and determine the final core extraction control strategy if the geometric residual error distribution does not meet the preset accuracy standard. The output module is used to acquire real-time evolution record data during the core extraction process based on the final core extraction control strategy, integrate stress release control effect and working condition dynamic data, and output the final core extraction processing control result.

Images