Light diffusion plate production process data analysis method and system

Abstract

The invention provides a light diffusion plate production process data analysis method and system, and relates to the field of data analysis, and the method comprises the steps: obtaining various types of information in the whole production process of a light diffusion plate, comprising batch characteristic information of raw materials, processing operation information in a production process, operation state information of production equipment, offline quality information of the light diffusion plate and environmental condition information of the light diffusion plate in a storage link, and the multi-source heterogeneous information is effectively associated to construct a complete life cycle record of the light diffusion plate. The defects that in the prior art, data information is broken, deep reasons of product quality problems are difficult to trace, and long-term quality degradation of products cannot be predicted are overcome, and accurate recognition and quantitative evaluation of product quality influence factors can be achieved by comprehensively integrating and deeply associating data of different links. Therefore, the problem of information isolated island in the prior art is effectively solved.

Description

Technical Field

[0001] This application relates to the field of data analysis technology, and more specifically, to a data analysis method and system for the production process of light diffusion plates. Background Technology

[0002] In the continuous production of light diffusion plates, product quality stability and long-term reliability are crucial. However, existing production data analysis methods fail to fully consider the numerous factors affecting product quality. Information from different stages varies in data type, structure, and time scale, making effective correlation and integration difficult. This information gap makes it difficult for factories to accurately trace the root causes of product quality problems, such as yellowing, thus impacting overall quality control capabilities and supply chain efficiency.

[0003] For example, in a factory specializing in the production of high-performance light diffusion plates, numerous sensors are deployed on the production line to collect real-time operational information from critical equipment. Regarding raw materials, even within the same batch of certified raw materials, there may be subtle, undeclared differences in particle size distribution or additive concentration.

[0004] Secondly, the condition of the production equipment itself is an influencing factor. Although the extruder's temperature sensor indicates that the set value has been reached, the entire mechanical and thermal system may not have reached a truly stable equilibrium state after being idle for two days. The first few hours after startup are a "warm-up" process, during which the thermal expansion of metal components and the flow characteristics of the polymer melt are still undergoing subtle changes. This "thermal history" or operating status of the equipment is an unmonitored variable that genuinely affects the processing. On the other hand, the storage environment for high-performance light diffusers, especially temperature and humidity fluctuations and potential indirect sunlight exposure, is not uniform throughout the warehouse. This factor has never been considered in the scope of factors affecting the long-term stability of the product's optical performance. The data from the production process and the data from the storage environment are completely isolated, making it impossible to foresee the occurrence of such quality degradation problems. Summary of the Invention

[0005] This application discloses a data analysis method and system for the production process of light diffusion plates, aiming to solve the technical problem that existing data analysis methods for light diffusion plates fail to fully consider various factors affecting product quality, making it difficult to accurately trace the root causes of quality problems and predict product quality deterioration, thereby affecting overall quality control capabilities and supply chain efficiency.

[0006] The technical solution of this application is as follows: In a first aspect, this application discloses a data analysis method for the production process of a light diffusion plate, including: Acquire various types of information throughout the entire production process of light diffusion plates, including batch characteristics of raw materials, processing operation information during production, operating status information of production equipment, quality information of light diffusion plates after production, and environmental conditions of light diffusion plates during warehousing. The batch characteristics of raw materials are associated with processing operation information and operating status information. The processing operation information and operating status information are associated with the quality information of the finished product and the environmental conditions information to construct a complete life cycle record of the light diffusion plate. Based on the complete lifecycle record, the impact of raw material batch characteristics, processing operations, operating status, and environmental conditions on the quality of the light diffusion plate is determined, and the probability data of quality degradation of the light diffusion plate during subsequent storage and use is output.

[0007] Secondly, this application also discloses a data analysis system for the production process of light diffusion plates, used to analyze data from the production process of light diffusion plates. The system includes: The information acquisition module is used to acquire various types of information throughout the entire production process of the light diffusion plate. These various types of information include batch characteristic information of raw materials, processing operation information during the production process, operating status information of production equipment, quality information of the light diffusion plate after production, and environmental condition information of the light diffusion plate in the warehousing process. The information association module is used to associate the batch characteristics information of raw materials with processing operation information and operating status information, associate the processing operation information and operating status information with the quality information after production line, and associate the quality information after production line with environmental condition information, so as to construct a complete life cycle record of the light diffusion plate. The quality judgment and prediction module is used to determine the impact of raw material batch characteristics, processing operations, operating status and environmental conditions on the quality of the light diffusion plate based on the complete life cycle record, and output the probability data of quality degradation of the light diffusion plate during subsequent storage and use.

[0008] This technical solution enables the comprehensive acquisition, efficient correlation, and intelligent analysis of data from the entire production process of light diffusion plates through modular design. This allows for the construction of a complete quality traceability and prediction system, effectively improving the quality control capabilities and management efficiency of light diffusion plate production. Beneficial effects

[0009] This application discloses a data analysis method for the production process of light diffusion plates. By acquiring various types of information throughout the entire production process, including batch characteristics of raw materials, processing operation information, operating status information of production equipment, quality information of the finished light diffusion plates, and environmental conditions during warehousing, this method effectively correlates these multi-source heterogeneous information to construct a complete lifecycle record for the light diffusion plates. Based on this complete lifecycle record, this application can determine the impact of raw material batch characteristics, processing operations, operating status, and environmental conditions on the quality of the light diffusion plates and output probability data of quality degradation during subsequent storage and use. This method overcomes the shortcomings of existing technologies, such as fragmented data, difficulty in tracing the root causes of product quality problems, and inability to predict long-term product quality deterioration. By comprehensively integrating and deeply correlating data from different stages, this application can achieve accurate identification and quantitative assessment of factors affecting product quality, thereby effectively solving the problem of information silos in existing technologies. Furthermore, by predicting the probability of future quality deterioration, this application can provide factories with forward-looking quality warnings, enabling enterprises to take intervention measures in advance, optimize production plans and warehouse management, significantly improve overall quality control capabilities and supply chain efficiency, and avoid economic losses caused by quality problems such as product yellowing. It has significant practical value and economic benefits. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of a data analysis method for the production process of a light diffusion plate provided in this application.

[0011] Figure 2 A schematic diagram of a data analysis system for the production process of a light diffusion plate provided in this application. Detailed Implementation

[0012] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0013] Reference Figure 1 The diagram illustrates an embodiment of a data analysis method for the production process of a light diffusion plate according to an embodiment of the present invention, which may specifically include the following steps: S101, acquire various types of information throughout the entire production process of the light diffusion plate, including batch characteristic information of raw materials, processing operation information during production, operating status information of production equipment, quality information of the light diffusion plate after production, and environmental condition information of the light diffusion plate in the warehousing process. S102, the batch characteristic information of the raw materials is associated with the processing operation information and the operating status information, the processing operation information and the operating status information are associated with the off-line quality information, and the off-line quality information is associated with the environmental condition information, so as to construct a complete life cycle record of the light diffusion plate; S103, based on the complete lifecycle record, determine the impact of raw material batch characteristics, processing operations, operating status and environmental conditions on the quality of the light diffusion plate, and output the probability data of the light diffusion plate undergoing quality degradation during subsequent storage and use.

[0014] This application integrates various types of information throughout the entire production process of light diffusion plates and constructs a complete lifecycle record, enabling a comprehensive and accurate assessment and prediction of the quality of light diffusion plates. This effectively solves the problems of information fragmentation, difficulty in traceability, and inability to predict product quality degradation in existing technologies.

[0015] To better understand the data analysis method for the production process of light diffusion plates proposed in this application, some key terms involved will be explained below.

[0016] A "light diffusion plate" typically refers to a type of sheet material that can evenly scatter light, and it is widely used in lighting, display, and other fields. Its quality directly affects the optical performance and lifespan of the final product.

[0017] "Batch characteristics of raw materials" refers to the differences in the physical and chemical properties of raw materials such as polymers and additives used in the production of light diffusion plates between different batches, such as molecular weight distribution, impurity content, and thermal stability. These characteristics have a significant impact on the performance of the final product.

[0018] "Processing operation information" refers to the specific operational parameters of the extrusion, injection molding, calendering, and other processes during the production of light diffusion plates, such as temperature, pressure, speed, and cooling rate. These parameters directly determine the microstructure and macroscopic properties of the material.

[0019] "Production equipment operating status information" refers to the real-time operating status data of various equipment on the production line, such as extruder screw speed, heating zone temperature, die pressure, cooling water flow rate, motor current, vibration frequency, etc. Abnormal operation of equipment may lead to product defects.

[0020] "Off-line quality information of light diffusion plates" refers to the various quality indicators detected at the end of the production line, such as light transmittance, haze, yellowness index, surface flatness, dimensional accuracy, and internal defects. These are the final quality inspection results before the product leaves the factory.

[0021] "Environmental conditions during warehousing" refers to the environmental factors that the light diffusion panels experience during storage in the warehouse, such as temperature, humidity, light intensity, ultraviolet radiation, and air pollutant concentration. These environmental factors may cause the product to age or deteriorate in performance during storage.

[0022] "Complete lifecycle record" refers to a comprehensive database formed by linking and integrating information from all the above-mentioned links. It can trace the entire process of each batch or each light diffusion plate from raw materials to production and warehousing, thereby achieving comprehensive traceability and analysis of product quality.

[0023] "Probability data of quality degradation" refers to the assessment, through data analysis and model prediction, of the possibility that the performance indicators (such as yellowness, light transmittance, etc.) of the light diffuser will fall below the preset standard threshold due to various factors in a specific period of time in the future.

[0024] The core of the data analysis method for the production process of light diffusion plates proposed in this application lies in acquiring, correlating, judging, and predicting various types of information in the entire production process of light diffusion plates.

[0025] Multiple methods can be employed to acquire various types of information throughout the entire light diffusion plate production process. For example, batch characteristic information of raw materials can be obtained by manually entering or scanning batch reports provided by suppliers when raw materials are received. This information can include the chemical composition and physical performance parameters of the raw materials. During production, sensors can be installed on production equipment to collect processing operation information and equipment operating status information in real time. For example, temperature sensors can collect extrusion temperature, pressure sensors can collect mold pressure, and current sensors can collect motor current. This sensor data can be automatically recorded in the Manufacturing Execution System (MES). For the quality information of the light diffusion plates after production, automated optical inspection (AOI) equipment can be installed at the end of the production line to scan and analyze the finished products, obtaining data such as their optical performance and surface defects. Alternatively, manual sampling can be used to conduct laboratory tests on the products to obtain more detailed quality data. In the warehousing stage, an environmental sensor network can be deployed in the warehouse to monitor environmental conditions such as temperature, humidity, and light intensity in real time, and this data can be associated with the batch number or unique identifier of the products entering the warehouse.

[0026] To construct a complete lifecycle record for the light diffusion plate by associating the batch characteristics of the raw materials with the processing operation information and the operating status information, associating the processing operation information and the operating status information with the off-line quality information, and associating the off-line quality information with the environmental condition information, database technology and data mining technology can be employed. For example, a central database can be established to integrate data from different sources and in different formats. By assigning a unique identifier to each batch or each product, the batch characteristics of the raw materials, processing operation information, operating status information, off-line quality information, and storage environmental condition information can be associated. Specifically, when a batch of raw materials is put into production, its batch characteristics can be bound to the processing operation information and the operating status information of the production equipment experienced by that batch of products during production. When the product comes off the production line and undergoes quality inspection, its off-line quality information can be associated with the corresponding processing operation information and operating status information. Finally, when the product enters the storage stage, its off-line quality information can be associated with the environmental condition information experienced by the product during storage. This association can be achieved through various methods such as timestamps, batch numbers, and product serial numbers, thereby forming a full lifecycle data chain from raw materials to final products and then to warehousing.

[0027] To determine the impact of raw material batch characteristics, processing operations, operating status, and environmental conditions on the quality of light diffusion plates based on the complete lifecycle record, and to output the probability data of quality degradation during subsequent storage and use, statistical analysis, machine learning, and other methods can be employed. For example, historical data can be used to establish mathematical models between various factors and product quality degradation. By analyzing data in the complete lifecycle record, it is possible to identify which raw material batch characteristics, processing operation parameters, equipment operating status, or storage environment conditions are significantly correlated with product quality degradation (e.g., yellowing, decreased light transmittance). For instance, if it is found that products produced from a specific batch of raw materials are more prone to yellowing, or products stored at a specific temperature are more prone to performance degradation, then these factors can be identified as key factors affecting product quality. Based on these analytical results, predictive models, such as regression or classification models, can be constructed. Inputting the complete lifecycle record data of the current product, the output is the probability data of quality degradation during subsequent storage and use. This probability data can be presented in the form of percentages, risk levels, etc., providing decision support for enterprises.

[0028] The data analysis method for the production process of light diffusion plates proposed in this application can comprehensively and accurately evaluate and predict the quality of light diffusion plates by integrating various types of information in the entire production process and constructing a complete life cycle record.

[0029] Specifically, this method first acquires various types of information throughout the entire production process of light diffusion plates, including batch characteristics of raw materials, processing operation information during production, operating status information of production equipment, quality information of the light diffusion plates after production, and environmental conditions information of the light diffusion plates during warehousing. This information covers all the key aspects affecting the quality of light diffusion plates.

[0030] Subsequently, information from these different stages is correlated to construct a complete lifecycle record for the light diffusion plate. This correlation breaks down the information silos inherent in traditional data analysis, enabling the traceability and analysis of data throughout the entire process, from raw materials to the final product and then to warehousing. For example, by correlating raw material batch characteristic information with processing operation information and operational status information, the performance of specific raw materials under specific process conditions can be analyzed; by correlating processing operation information and operational status information with off-line quality information, the impact of the production process on product quality can be assessed; and by correlating off-line quality information with environmental condition information, the impact of the warehousing environment on the long-term performance of the product can be predicted.

[0031] Finally, based on the constructed complete lifecycle record, the impact of raw material batch characteristics, processing operations, operating status, and environmental conditions on the quality of the light diffusion plate is determined, and the probability data of quality degradation of the light diffusion plate during subsequent storage and use is output. This step is the core value of this application, enabling enterprises to shift from passive post-event handling to proactive pre-event prevention. Through in-depth mining and analysis of historical data, key factors leading to quality problems can be identified, and predictive models can be established. For example, if a batch of raw materials has high photosensitivity and has experienced high extrusion temperatures during production, while being exposed to strong ultraviolet light during storage, then the probability of yellowing in subsequent use of the product will be predicted to be high. This probability data provides enterprises with a quantitative risk assessment, which helps to optimize production processes, improve raw material selection, and adjust storage strategies, thereby significantly improving the stability and reliability of product quality.

[0032] Compared to existing technologies, the advantages of this application lie in its comprehensiveness and predictive capabilities. Traditional methods often focus only on real-time process parameters during production and the quality performance of products upon completion, failing to effectively integrate data from different stages and types. This information gap makes it difficult for companies to accurately trace the root causes of product quality problems, let alone predict potential quality degradation during subsequent storage and use. This application constructs a complete lifecycle record for the light diffusion plate, deeply linking and integrating information from all key stages, including raw materials, production, quality upon completion, and warehousing, thereby achieving a comprehensive understanding of factors influencing product quality. Furthermore, this application can output probability data on the quality degradation of the light diffusion plate during subsequent storage and use. This allows companies to provide early warnings of potential quality risks and take preventative measures, significantly improving quality control capabilities and supply chain efficiency, and reducing economic losses caused by product quality degradation.

[0033] This application proposes a more refined method for acquiring and associating batch characteristic information of raw materials. By quantifying the inherent sensitivity of raw materials to light, a complete life cycle record of the light diffusion plate can be constructed more accurately.

[0034] The steps described above for constructing a complete lifecycle record for light diffusion plates include: associating batch characteristic information of raw materials with processing operation information and operating status information; associating processing operation information and operating status information with off-line quality information; and associating off-line quality information with environmental condition information. When batches of polycarbonate raw materials arrive at the factory, standardized accelerated aging stimulation is applied to the collected raw material samples in a microreactor, and the rate of change of the yellowness index of the raw material samples during the accelerated aging stimulation is monitored to quantify the inherent sensitivity of the batch of raw materials to light. The rate of change of the yellowness index is used as a micro-activity imprint of the raw material batch, and the micro-activity imprint is bound to the raw material batch number as batch characteristic information of the raw material, which is then integrated into the complete life cycle record of the light diffusion plate.

[0035] Specifically, the arrival of polycarbonate raw material batches refers to the stage where the supplier delivers polycarbonate granules or sheets to the production plant and completes preliminary acceptance. During this stage, representative raw material samples are randomly selected from each batch. A microreactor can be understood as a miniaturized, controllable experimental device capable of simulating and accelerating the aging environment that a light diffuser might encounter in actual use, such as high-intensity ultraviolet radiation and high temperatures. Standardized accelerated aging stimulation refers to applying simulated aging effects to raw material samples according to preset, uniform experimental conditions and parameters, ensuring the comparability of test results between different batches. For example, continuous irradiation with a specific wavelength and intensity of ultraviolet light can be used, while maintaining constant temperature and humidity. The rate of change of the yellowness index refers to the rate at which the degree of yellowing of the raw material sample changes over time during accelerated aging stimulation; it can serve as a key indicator for measuring the material's resistance to photoaging. By monitoring the rate of change of the yellowness index, the inherent sensitivity of a batch of raw materials to light can be quantified, i.e., the tendency and speed at which that batch of raw materials yellows under light exposure. The micro-active imprint can be understood as a unique, quantified identifier of the inherent characteristics of raw materials, reflecting their microscopic sensitivity to specific environmental stimuli (such as light). Linking the micro-active imprint to the raw material batch number means associating this quantified data with the raw material's unique batch identification code, ensuring data traceability. As batch characteristic information of the raw material, the micro-active imprint will serve as important, in-depth raw material attribute data, integrated into the complete lifecycle record of the light diffusion plate, providing more insightful foundational data for subsequent product quality analysis and prediction.

[0036] This application's solution involves standardized accelerated aging stimulation of polycarbonate raw material samples during the raw material arrival stage, and monitoring the rate of change in their yellowness index. This allows for the early and accurate quantification of the inherent sensitivity of raw material batches to light exposure. This inherent sensitivity is abstracted as a "micro-active imprint" and linked to the raw material batch number, serving as a novel and more predictive characteristic information for raw material batches. Because this micro-active imprint can reveal the potential response of raw materials to light aging at the molecular level, it allows for the incorporation of deeper raw material quality attributes when constructing a complete lifecycle record for the light diffusion plate. By integrating this inherent sensitivity information into the lifecycle record, subsequent quality assessment and prediction modules can more accurately evaluate the impact of raw material batch characteristics, processing operations, operating status, and environmental conditions on the long-term quality of the light diffusion plate, especially quality degradation issues such as yellowing and fogging that may occur under light exposure.

[0037] In some preferred embodiments, the process is as follows: When a new batch of polycarbonate raw materials (e.g., batch number PC20230815-001) arrives at the factory and undergoes preliminary inspection, approximately 50 grams of raw material particles are randomly selected from the batch as samples. These samples are then fed into a specially designed microreactor. This microreactor is configured to simulate the operating conditions of a light diffuser in outdoor or high-light environments. For example, it is equipped with a 100W xenon lamp to simulate the solar spectrum, with an ultraviolet intensity set to 0.8W / m²@340nm. The internal temperature of the reactor is maintained at 80°C, and the humidity is controlled at 50%RH. Accelerated aging stimulation is applied to the raw material samples for 24 consecutive hours. During this period, the built-in spectrometer in the microreactor automatically collects the reflectance spectrum data of the samples every hour and calculates their yellowness index. By performing linear regression analysis on the changes in the yellowness index over 24 hours, the rate of change of the yellowness index of this batch of raw materials is obtained, for example, 0.05 units / hour. This value of 0.05 units / hour was identified as the micro-activity imprint of batch PC20230815-001. Subsequently, this micro-activity imprint data was linked to batch number PC20230815-001 and entered into the database of the light diffusion plate production management system as batch characteristic information of the raw material, becoming part of the complete life cycle record of the light diffusion plates produced from this batch of raw materials.

[0038] In this regard, this application further proposes that the steps of associating processing operation information, operating status information, and offline quality information, and associating offline quality information with environmental condition information, to construct a complete lifecycle record for the light diffusion plate, include: Sensing units are deployed on the highly polished structural components of the new electric forklift; The intensity of the beam reflected by the forklift is monitored by the sensing unit; When the intensity of the reflected beam exceeds a preset threshold and the proportion of ultraviolet components increases abnormally, the control system of the new electric forklift calculates the projection area of ​​the reflected beam based on data from its built-in positioning module and attitude sensor. The novel electric forklift is controlled to send instantaneous exposure event commands with timestamps, peak intensity, and ultraviolet dose estimates to environmental response tags of all light diffusers within the projection area of ​​the reflected beam via a directional wireless communication module. The environmental response tag is controlled to record detailed information about the instantaneous exposure event command; The system receives and integrates instantaneous high-intensity environmental impact data from the novel electric forklift and the environmental response tag, and uses the instantaneous high-intensity environmental impact data as the environmental condition information to update the complete life cycle record of the light diffuser. The instantaneous high-intensity environmental impact data includes the timestamp, peak intensity, and UV dose estimate after integration from the novel electric forklift and the environmental response tag.

[0039] Specifically, the highly polished structural components of the new electric forklift can be understood as parts whose surfaces have undergone special treatment to achieve high reflectivity. Their purpose is to provide a stable source of reflected light for the sensing unit, or to serve as a reference surface for ambient light detection. The sensing unit is deployed on this structural component to monitor the intensity and spectral composition of ambient light in real time, particularly the ultraviolet (UV) component. When the sensing unit detects that the intensity of the reflected light suddenly exceeds a preset safety threshold, and simultaneously detects an abnormally high proportion of UV radiation, this indicates a potential, instantaneous, high-intensity environmental impact event that could negatively affect the quality of the light diffuser plate.

[0040] The new electric forklift features a built-in positioning module and attitude sensors that work together to accurately determine the forklift's position and orientation in space. Based on this data, the projection area of ​​the reflected light beam can be calculated, i.e., the specific spatial range affected by the instantaneous environmental impact. A directional wireless communication module is configured to send commands to specific devices within this projection area, ensuring accurate information transmission. An environmental response tag, a smart tag attached to the light diffusion plate, has built-in storage and communication functions. It can receive and record instantaneous exposure event commands from the forklift, including the timestamp of the event, peak light intensity, and estimated ultraviolet dose.

[0041] In practical applications, instantaneous high-intensity environmental impact data refers to detailed information about a specific environmental event captured and integrated by new electric forklifts and environmental response tags. This data not only includes the time, intensity, and UV dose of the event, but also, by being linked to specific light diffuser products, enables each product's lifecycle record to accurately reflect the localized, instantaneous environmental impact it has experienced.

[0042] This application's solution utilizes sensing units deployed on a novel electric forklift to achieve real-time, mobile monitoring of instantaneous high-intensity light and UV exposure in warehouse environments. When an abnormal lighting event that may damage the light diffuser is detected, the forklift's positioning module and attitude sensors accurately calculate the affected area. Subsequently, via a directional wireless communication module, a command for the instantaneous exposure event with detailed parameters is sent to the environmental response tags of all light diffusers within that area. These tags receive and record the event information, thereby accurately linking previously difficult-to-track localized, instantaneous environmental impact events to specific individual products. This allows for the updating of the complete lifecycle record of the light diffuser, containing more refined and accurate environmental condition information, overcoming the shortcomings of traditional methods in capturing such instantaneous high-intensity environmental impacts.

[0043] In some preferred embodiments, suppose a new type of electric forklift is performing routine material handling operations within a warehouse area covered by light diffusers. Sensing units deployed on the forklift's highly polished structural components continuously monitor the ambient light intensity and UV composition. Suddenly, a brief but extremely intense UV flash occurs due to a malfunction in one of the warehouse's lighting fixtures. The sensing units immediately detect that the intensity of the reflected beam momentarily exceeds a preset 5000 lux threshold, and the UV component abnormally increases to over 15% of the total light intensity. At this point, the forklift's built-in positioning module and attitude sensors quickly calculate the projection area of ​​the UV flash, for example, identifying it as the third to fifth shelves of shelf A, covering approximately 20 light diffusers. The forklift then sends a momentary exposure event command via its directional wireless communication module to environmental response tags attached to these 20 light diffusers. This command includes a precise timestamp (e.g., October 26, 2023, 14:35:12), peak intensity (e.g., 8000 lux), and an estimated UV dose (e.g., 0.5 joules / cm²). Upon receiving an instruction, the environmental response tag of each light diffuser plate records its detailed information in its own memory. Subsequently, this recorded instantaneous high-intensity environmental impact data is received and integrated into the complete lifecycle record of the corresponding light diffuser plate as part of its environmental condition information. In this way, even brief, localized environmental anomalies can be accurately tracked and associated with the affected products, providing crucial, high-precision environmental exposure data for subsequent quality analysis and prediction.

[0044] In some embodiments described above in this application, a method is proposed to determine the impact of raw material batch characteristics, processing operations, operating status, and environmental conditions on the quality of the light diffusion plate based on its complete lifecycle record, and to output probability data of quality degradation during subsequent storage and use. Specifically, the steps described above for determining the impact of raw material batch characteristics, processing operations, operating status, and environmental conditions on the quality of the light diffusion plate based on its complete lifecycle record, and outputting probability data of quality degradation during subsequent storage and use, can be further refined as follows.

[0045] Obtain a complete lifecycle record for each batch of light diffusion plate products. This complete lifecycle record includes the micro-activity imprint of raw materials, the melt micro-stress imprint during the production process, instantaneous environmental exposure information during storage, and the molecular aging evolution trajectory. The micro-activity imprint refers to the inherent sensitivity of raw materials to light. The melt micro-stress imprint refers to the shear and tensile mechanical stresses experienced by the polymer melt during processing such as extrusion and injection molding. The instantaneous environmental exposure information refers to the captured localized, instantaneous high-intensity environmental impacts. Based on the information in the complete life cycle record, the molecular aging state evolution within the product is calculated and updated, including the cumulative concentration of oxidation products and the average molecular weight. Using the molecular aging state evolution as the predicted center value, and combining the micro-activity imprint of the raw material, the micro-stress imprint of the melt, and the fluctuation range and uncertainty of the instantaneous environmental exposure information, multiple simulated trajectories of the molecular aging state evolution are generated through Monte Carlo simulation. Based on the multiple simulated trajectories, the probability distribution of quality degradation of the light diffusion plate during subsequent storage and use is calculated. Based on the probability distribution, the quality degradation potential index, confidence interval, and corresponding risk classification of the light diffusion plate are generated.

[0046] The complete lifecycle record is a key data set for comprehensively tracing the entire process of the light diffusion plate from raw materials to the final product. Specifically, the micro-activity imprint of the raw materials characterizes the inherent sensitivity of the raw materials to external environmental stimuli (especially light), which can be quantified through standardized accelerated aging experiments. The melt micro-stress imprint during the production process reflects the mechanical stress history experienced by the polymer during processing and molding, such as the magnitude and duration of shear or tensile stress in extrusion or injection molding. These stresses may cause the polymer molecular chains to break or rearrange, thereby affecting the performance of the final product. The instantaneous environmental exposure information in warehousing refers to the localized, instantaneous high-intensity environmental shocks that the product may encounter during storage, such as unexpected strong light exposure, high temperature or high humidity events, which may accelerate the aging process of the product. The molecular aging state evolution trajectory is a dynamic indicator used to describe the changes in the internal chemical structure and physical properties of the product over time, such as the accumulation of oxidation products and changes in molecular weight. These are core indicators for measuring the degree of material aging.

[0047] Furthermore, the calculation and updating of the molecular aging state evolution aims to quantify the cumulative damage within the product. The cumulative concentration of oxidation products can serve as an indicator of the degree of oxidative degradation of the material, while the average molecular weight reflects the degree of polymer chain breakage or cross-linking. Accurate calculation of these indicators is crucial for assessing the current aging state of the product.

[0048] As a preferred implementation, the Monte Carlo simulation method is used to handle the inherent fluctuations and uncertainties of various input information, such as micro-activity imprints, melt micro-stress imprints, and transient environmental exposure information. By generating multiple simulated trajectories of the molecular aging state evolution, this method can comprehensively explore the aging path of a product under different scenarios, thereby more accurately assessing its future quality risks.

[0049] Therefore, the calculation of the probability distribution provides a probabilistic view of the quality degradation of the light diffusion plate during subsequent storage and use, rather than a simple binary judgment. Based on this probability distribution, a quality degradation potential index for the light diffusion plate can be generated. This index is a comprehensive quantitative indicator used to assess the potential risk of future quality degradation of the product; it also provides a confidence interval to reflect the reliability range of the prediction results and gives a corresponding risk classification to facilitate managers in quickly identifying and handling high-risk product batches.

[0050] Through the above technical solution, this application provides a more refined, quantitative, and forward-looking method for assessing the quality risk of light diffusion plates. Compared to traditional methods that only assess the impact, this application can output specific data on the probability of quality degradation and provide a quality deterioration potential index, confidence interval, and risk classification. This enables quality management personnel to more accurately understand and predict the long-term performance of products. By integrating key data throughout the entire lifecycle from raw materials to warehousing and utilizing Monte Carlo simulation to handle uncertainty, this solution significantly improves the reliability and robustness of quality prediction. Consequently, enterprises can more effectively manage inventory, trace product batches, and provide after-sales service, thereby reducing potential quality risks, enhancing product competitiveness, and providing customers with more reliable product quality assurance.

[0051] This application further proposes a procedure for constructing a complete lifecycle record of a light diffuser plate, which includes: Obtain a unique identifier for each batch of light diffusion plates; Based on the unique identifier, obtain the batch number of the raw materials used in the production process of this batch of products and the corresponding production time period; By analyzing the micro-activity imprints of the raw material batch number and combining them with the melt micro-stress imprints during the production period, the cumulative impact of raw material batch characteristics and production process on the initial molecular aging state of the product is estimated. Track the physical location changes of the light diffusion plate batch during the warehousing process and correlate them with the instantaneous environmental exposure information experienced by the light diffusion plate batch. Based on the instantaneous environmental exposure information accumulated by the light diffusion plate product during the warehousing process and the cumulative impact of the initial molecular aging state of the product, the secondary impact of the warehousing environment on the molecular aging state of the product is estimated. Based on the secondary impact of raw material batch characteristics, production process, and storage environment on the molecular aging state of the product, a nonlinear correlation is established between information from different stages, and the complete life cycle record of the light diffusion plate is updated.

[0052] Specifically, obtaining the unique identifier for each batch of light diffusion panels involves scanning barcodes on the product packaging, RFID tags, or querying a database to acquire a unique identification code for each batch. This unique identifier is crucial for tracing the entire product lifecycle. Based on this unique identifier, obtaining the batch number of the raw materials used in the production process and the corresponding production time period involves using a traceability system to link a specific batch of light diffusion panels to the specific batch of raw materials used in their manufacture (e.g., the batch number of polycarbonate granules) and the precise time period that batch of products spent on the production line. This ensures that subsequent analysis can accurately trace back to the source.

[0053] By analyzing the micro-activity imprints of the raw material batches and combining them with the melt micro-stress imprints during the production period, the cumulative impact of raw material batch characteristics and the production process on the initial molecular aging state of the product can be calculated. The micro-activity imprint can be understood as the inherent sensitivity of the raw material to environmental factors such as light and heat, for example, the rate of change in yellowness index measured through accelerated aging experiments. The melt micro-stress imprint refers to the mechanical stresses such as shear and tension experienced by the polymer melt during extrusion, injection molding, and other processing. These stresses may lead to polymer chain breakage or structural changes, thereby affecting the initial molecular aging state of the product. By comprehensively analyzing these two types of information, the initial cumulative impact of inherent defects in the raw materials and stresses during the production process on the molecular structural stability of the product at the time of shipment can be quantified.

[0054] The physical location changes of the light diffuser batches during warehousing are tracked, and the transient environmental exposure information experienced by these batches is correlated. This can be achieved by deploying positioning systems (such as UWB, RFID) and environmental sensor networks in the warehousing area. Transient environmental exposure information refers to localized, transient, high-intensity environmental shocks that the product may encounter during warehousing, such as unexpected strong light exposure, sudden temperature changes, or abnormal humidity. Through precise tracking and correlation, potential degradation factors that the product may be subjected to during warehousing can be identified. Based on the accumulated transient environmental exposure information of the light diffuser products during warehousing and the cumulative impact of the product's initial molecular aging state, the secondary impact of the warehousing environment on the product's molecular aging state is calculated. This means treating environmental shocks during warehousing as a further effect on the product's initial molecular aging state, and calculating how these shocks accelerate or alter the product's molecular aging process.

[0055] Based on the secondary impacts of raw material batch characteristics, production processes, and storage environments on the molecular aging state of products, nonlinear correlations are established between information from different stages, and the complete lifecycle record of the light diffusion plate is updated. Specifically, machine learning models (such as neural networks and support vector machines) can be used to capture these complex, nonlinear interactions. For example, the micro-active imprints of a certain raw material may exhibit a stronger tendency to deteriorate under specific processing stresses, or accelerate aging under specific storage environment shocks. By establishing these nonlinear correlations, the molecular aging trajectory of the product throughout its entire lifecycle can be more accurately reflected, thereby updating and improving the complete lifecycle record of the light diffusion plate.

[0056] In some preferred embodiments, this application is implemented as follows: Suppose a batch of light diffusion plates has a unique identifier of "PC-DP-20230815-001". First, the system uses this unique identifier to retrieve from the database that this batch of products used polycarbonate raw material with batch number "RAW-PC-A001" and was extruded between 8:00 AM and 10:00 AM on August 15, 2023.

[0057] Next, the system analyzed the micro-activity imprint data of the "RAW-PC-A001" batch of raw materials. For example, its yellowness index change rate under standard ultraviolet irradiation was 0.05 / hour, indicating that it has moderate sensitivity to light. Simultaneously, combined with the melt micro-stress imprint data recorded by the extruder melt pressure sensor during this production period, several instantaneous shear stress peaks were found, which may cause some damage to the polymer chains. Using a pre-set model, the system calculated the cumulative impact of the raw material batch characteristics and the production process on the initial molecular aging state of this batch of products. For example, the initial oxidation product concentration was slightly higher than the average level, and the average molecular weight decreased slightly.

[0058] Subsequently, the system tracked the physical location changes of batch "PC-DP-20230815-001" within the warehouse area. For example, this batch of products was initially placed near a window during the initial storage phase and was detected by deployed sensors to have experienced a momentary high-intensity direct sunlight event between 2:00 PM and 2:30 PM on a certain afternoon, with an estimated UV dose of 5000 J / m². The system correlated this momentary environmental exposure information with the product.

[0059] Based on the cumulative impact of the initial molecular aging state of the product and the information on instantaneous environmental exposure accumulated during storage, the system further calculates the secondary impact of the storage environment on the molecular aging state of the product. For example, by combining the ultraviolet dose and the photosensitivity characteristics of the raw materials, it is calculated that the instantaneous exposure event caused an additional 0.02% increase in the concentration of oxidation products.

[0060] Ultimately, the system utilizes a pre-trained nonlinear model (e.g., a deep learning-based model) to establish a nonlinear correlation between the micro-activity imprints of raw materials, melt micro-stress imprints, and instantaneous exposure information from the storage environment, as inputs, and the molecular aging state evolution trajectory. For example, the model might reveal that raw materials with high micro-activity imprints exhibit significantly increased sensitivity to UV exposure after experiencing specific melt stresses. Through this nonlinear correlation, the system updated the complete lifecycle record of the "PC-DP-20230815-001" batch of light diffusion plates, enabling it to contain a more accurate molecular aging state evolution trajectory, thus providing a more reliable data foundation for subsequent prediction of quality degradation probability.

[0061] This application further proposes to refine the above calculation steps in order to accurately distinguish the cumulative effects of inherent defects in raw materials and instantaneous fluctuations in the production process on the initial molecular aging state of the product.

[0062] Specifically, the steps for analyzing the micro-activity imprints of raw material batch numbers and combining them with the melt micro-stress imprints during the production period to estimate the cumulative impact of raw material batch characteristics and the production process on the initial molecular aging state of the product include: Receive the micro-active imprints and conventional physicochemical properties of raw material batch numbers, as well as the melt micro-stress imprints during the production process; The micro-active imprint is compared with the conventional physicochemical indicators to identify whether there are significant differences between the micro-active imprint and the conventional physicochemical indicators; When there is a significant difference between the micro-active imprint and the conventional physicochemical indicators, the raw material batch will be marked as having a potential intrinsic defect; The micro-stress imprint of the melt is monitored in real time to identify whether there are local abnormally high values ​​in the micro-stress imprint of the melt. When a local abnormally high value is identified in the micro-stress imprint of the melt, the equipment operation log during the production period is used to analyze whether the local abnormally high value coincides with the instantaneous fluctuation event of the equipment in time. When the local abnormally high value coincides with the instantaneous fluctuation event of the equipment, the abnormality of the melt micro-stress imprint is attributed to the instantaneous fluctuation of the production process. When the local abnormally high value does not coincide with the instantaneous fluctuation event of the equipment, the abnormality of the melt micro-stress imprint is attributed to the inherent defects of the raw materials that manifest during the processing, in conjunction with the potential inherent defect markers of the raw material batch. The cumulative effects of inherent defects in raw materials and instantaneous fluctuations in the production process on the initial molecular aging state of the product are calculated and differentiated.

[0063] Specifically, when receiving the micro-active imprint and conventional physicochemical indicators of raw material batches, the micro-active imprint refers to the inherent sensitivity of the raw material to light, quantified by applying standardized accelerated aging stimulation to the raw material sample and monitoring the rate of change of its yellowness index. The conventional physicochemical indicators typically include parameters traditionally used to assess raw material quality, such as melt index, density, ash content, and moisture content. The melt micro-stress imprint refers to the mechanical stress, such as shear and tension, experienced by the polymer melt during extrusion, injection molding, and other processing, which can be collected in real time using an online rheometer or dedicated sensors. The comparison between the micro-active imprint and the conventional physicochemical indicators aims to discover potential defects in the raw material that may be overlooked under traditional testing methods. For example, some batches of raw materials may meet the standards for conventional physicochemical indicators, but their micro-active imprints may show abnormally high photosensitivity, indicating a potential photosensitive defect in that batch. When the comparison results show a significant difference, the raw material batch will be marked as having a potential inherent defect, providing an important basis for subsequent defect attribution.

[0064] Furthermore, real-time monitoring of the melt micro-stress imprint allows for the timely detection of abnormal stress states experienced by the polymer melt during production. When localized abnormally high values ​​are identified in the melt micro-stress imprint, analysis is required in conjunction with equipment operation logs for the specific production period. These logs record various operating parameters of the production equipment over a specific time period, such as temperature, pressure, screw speed, and current fluctuations, as well as potential instantaneous fluctuation events, such as power outages, mechanical vibrations, and cooling system anomalies. By comparing the temporal overlap between abnormally high values ​​in the melt micro-stress imprint and instantaneous equipment fluctuation events, it is possible to accurately determine whether the abnormal stress originates from a momentary equipment malfunction or an inherent defect in the raw material during processing.

[0065] This application's solution, by introducing a comparison between the micro-active imprints of raw materials and conventional physicochemical indicators, enables earlier and more comprehensive identification of potential inherent defects in raw materials. Simultaneously, through real-time monitoring of the melt's micro-stress imprints and cross-validation using equipment operation logs, it achieves precise attribution of the sources of abnormal stress occurring during production. This allows for a clear distinction between whether the cumulative impact of the initial molecular aging state of the product is primarily caused by inherent defects in the raw materials or by transient fluctuations during production. This refined attribution mechanism provides data support for subsequent quality control and process optimization.

[0066] This application further proposes that the steps for marking raw material batches as having potential inherent defects include: The differences in micro-active imprints and conventional physicochemical indicators between batches of raw materials were analyzed, and the differential characteristics were extracted. Based on the extracted differential features, a search is performed in a preset defect type mapping table to obtain the matching defect type. The mapping table sets the correspondence between different differential features and defect types. The defect types include photosensitive defects, thermosensitive defects, and oxidation-sensitive defects. Based on the matched defect type and the degree of impact of the defect type on the long-term quality deterioration of the product in historical data, the risk level of the raw material batch is calculated. The defect type and risk level are bound to the raw material batch number to generate a refined defect marker, which is then integrated into the complete lifecycle record of the light diffusion plate. During production planning and scheduling, based on the refined defect marking, high-risk batches of raw materials are allocated to products with preset quality requirement levels for production, or a special pre-processing procedure for high-risk batches of raw materials is triggered.

[0067] Specifically, analyzing the discrepancy patterns between micro-active imprints and conventional physicochemical indicators of raw material batches involves using data analysis techniques, such as pattern recognition algorithms or machine learning models, to identify specific deviation patterns or abnormal correlations between micro-active imprints and conventional physicochemical indicators, and extracting key features that characterize the nature of defects. These discrepancies can be numerical deviations, inconsistencies in trends, or abnormal performance of specific indicator combinations. The pre-defined defect type mapping table can be understood as a knowledge base or rule set, aiming to associate the extracted discrepancies with known, specific defect types in light diffusion plate raw materials. This mapping table predefines the correspondence between different combinations of discrepancies and specific defect types such as photosensitive defects, thermosensitive defects, and oxidation-sensitive defects. For example, if the micro-active imprint shows an abnormally high rate of change in the yellowness index under a specific wavelength of light, and the antioxidant content in the conventional physicochemical indicators is low, it may be mapped as a composite type of photosensitive and oxidation-sensitive defects. In practical applications, calculating the risk level of a raw material batch refers to a quantitative assessment of the actual impact of the matched defect type on the long-term quality deterioration of the light diffusion plate product (e.g., yellowing, increased haze, decreased mechanical properties) based on historical production and usage data. This can be accomplished through statistical analysis, regression models, or expert systems, thereby assigning a specific risk value or risk level (e.g., low, medium, high risk) to each raw material batch. Furthermore, generating a refined defect tag involves binding the identified defect type (e.g., photosensitive defects) and the calculated risk level (e.g., high risk) to the raw material batch number, forming a tag containing detailed defect information. This refined defect tag is then integrated into the complete lifecycle record of the light diffusion plate, serving as an important component of the batch's raw material characteristics for traceability and utilization in subsequent stages. As a preferred implementation, during production planning and scheduling, based on the refined defect marking, two strategies can be adopted: First, batches of raw materials marked as high-risk can be specifically allocated to products with relatively lower quality requirements for production, thereby reducing overall quality risk; second, for high-risk batches of raw materials, special pretreatment processes can be triggered, such as additional purification, modification, or mixing treatments, to mitigate their impact. In some preferred embodiments, it is assumed that the micro-active imprint of a batch of polycarbonate raw materials shows that its yellowness index changes at a rate significantly higher than the historical average under standard ultraviolet light, while its conventional physicochemical indicators (such as molecular weight and melt index) are within the normal range. By analyzing this difference pattern, the system extracts the difference feature of "high light sensitivity". According to a preset defect type mapping table, this feature is matched to "photosensitive defect". Further, combined with historical data analysis, it is found that light diffusion plates produced from raw materials with "photosensitive defects" are more prone to yellowing during long-term use, and the degree of yellowing is positively correlated with light sensitivity. Based on this, the system calculates the risk level of this batch of raw materials as "high risk". Subsequently, the system binds "photosensitive defect" and "high risk" with the batch of raw materials number to generate a refined defect label. During production planning and scheduling, if there are product orders with low requirements for yellowing (such as diffusion plates for indoor indirect light sources), the system will prioritize the allocation of this high-risk batch of raw materials to such products for production. Conversely, if all products to be manufactured have strict requirements for yellowing, the system will trigger a special pretreatment process for that batch of raw materials, such as adding additional UV absorbers or performing mixed modification treatment to reduce their photosensitivity, thereby ensuring that the quality of the final product meets the requirements.

[0068] This application further proposes a more refined method for raw material allocation, which aims to achieve synergistic optimization of quality risk and production efficiency by constructing a multi-objective optimization model.

[0069] In response, this application further proposes the following steps for allocating high-risk batches of raw materials to product grades with preset quality requirements for production scheduling based on refined defect tags, or triggering a special pre-processing procedure for high-risk batches of raw materials: obtaining a list of high-risk batches of raw materials currently to be scheduled, wherein the list of raw materials includes refined defect tags and available quantities for each batch; obtaining the quality requirement grades of all products in the current production order pool; for each high-risk batch of raw materials, calculating the overall quality risk introduced when allocating it to products with different quality requirement grades based on the corresponding refined defect tags; and for each production order, calculating the overall quality risk introduced when allocating it to products with different quality requirement grades based on the corresponding quality requirements. The system calculates the production efficiency when raw materials are allocated to different risk levels. A multi-objective optimization function is constructed under preset constraints, aiming to minimize overall quality risk and maximize production efficiency. These constraints include available raw material quantity, production order demand, and equipment capacity. Through iterative search and evaluation, a solution set of the multi-objective optimization function is found, representing a raw material allocation scheme that balances quality risk and production efficiency. The optimal raw material allocation scheme is selected from the solution set, generating allocation instructions for the high-risk batches of raw materials. These instructions include each high-risk batch of raw materials and the assigned specific product production task.

[0070] Specifically, obtaining the list of high-risk batches of raw materials currently awaiting scheduling refers to the system automatically or manually inputting information on all high-risk batches of raw materials that currently require production planning and scheduling. This raw material list not only includes batch numbers but also detailed records of refined defect markers for each batch, such as photosensitive defects, heat-sensitive defects, or oxidation-sensitive defects, their risk levels, and the current available quantity of raw materials for that batch. This information forms the basis for subsequent calculations and optimizations. Obtaining the quality requirement levels of all products in the current production order pool refers to extracting order information for all products awaiting production from the production management system and identifying the corresponding quality requirement level for each product. For example, some products may require extremely high optical performance, while others may have higher requirements for weather resistance; these different quality requirement levels will affect the raw material allocation strategy.

[0071] For each high-risk batch of raw materials, the overall quality risk introduced when allocating them to products with different quality requirement levels is calculated based on the corresponding refined defect markings. This means that for each high-risk raw material, the potential quality risk of using it to produce products with different quality requirements is assessed, taking into account its specific defect type and risk level. For example, raw materials with photosensitive defects will introduce significantly higher quality risks when used to produce products with high photostability requirements than when used to produce products with low photostability requirements. For each production order, the production efficiency allocated to raw materials of different risk levels is calculated based on the corresponding quality requirement level. This step aims to assess the impact of using raw materials of different risk levels on production efficiency while meeting specific product quality requirements. For example, using high-risk raw materials may require stricter process control or longer production cycles, thus affecting production efficiency.

[0072] A multi-objective optimization function is constructed under pre-defined constraints, aiming to minimize overall quality risk and maximize production efficiency. These constraints include available raw material quantity, production order demand, and equipment capacity. The multi-objective optimization function is a mathematical model that simultaneously optimizes two or more conflicting objectives (in this case, minimizing quality risk and maximizing production efficiency) while satisfying a series of practical production limitations, such as the total amount of raw materials in a specific batch not exceeding its available quantity, each production order meeting its demand, and production tasks not exceeding the actual capacity of the equipment. Through iterative search and evaluation, a solution set of the multi-objective optimization function is found, representing a raw material allocation scheme that balances quality risk and production efficiency. This is typically achieved using optimization algorithms, such as genetic algorithms, particle swarm optimization, or simulated annealing. These algorithms generate a series of possible allocation schemes (solution sets), each balancing quality risk and production efficiency to some extent. The optimal raw material allocation scheme is selected from the solution set, generating allocation instructions for the high-risk batches of raw materials. These instructions contain the specific product production task allocated to each high-risk batch of raw materials. The selection of the optimal solution can be based on preset weights or the decision-maker's preferences. For example, in some cases, the focus may be on minimizing quality risk, while in others, the focus may be on maximizing production efficiency. The resulting allocation instructions will guide the production line on how to use high-risk batches of raw materials.

[0073] In some embodiments described above, a method is proposed to compare the micro-active imprints of raw material batches with conventional physicochemical indicators to identify significant differences and thereby mark the raw material batches for potential inherent defects. However, simple comparisons may not accurately quantify and classify these differences, especially when raw material characteristics are complex or defect manifestations are diverse. Such coarse difference identification may lead to inaccurate judgments of potential defects, thereby affecting the effectiveness of subsequent risk assessments and production decisions.

[0074] In this regard, this application further proposes a step of comparing the aforementioned micro-active imprint with conventional physicochemical indicators to identify whether there are significant differences between the micro-active imprint and the conventional physicochemical indicators, including: Obtain micro-active imprint data and conventional physicochemical index data for each batch of raw materials; Feature extraction is performed on the micro-active imprint data to obtain a first feature vector; Feature extraction is performed on the conventional physicochemical index data to obtain a second feature vector; The first feature vector and the second feature vector are fused to form a comprehensive feature vector for the raw material batch; Calculate the similarity between the comprehensive feature vector and the preset feature vectors of multiple known difference patterns; Based on the similarity, the difference pattern to which the raw material batch belongs is identified, and a type identifier for the difference pattern is generated; the difference pattern refers to specific and actionable defect type information.

[0075] Specifically, obtaining micro-active imprint data and conventional physicochemical index data for a batch of raw materials refers to collecting all micro-active imprint data and conventional physicochemical index data related to that batch of raw materials from sources such as incoming inspection reports and accelerated aging test results from microreactors. Micro-active imprint data may include, but is not limited to, dynamic response data such as the rate of change of yellowness index, spectral absorption or emission intensity at a specific wavelength, and oxidation induction time. Conventional physicochemical index data may include static physicochemical parameters such as melt flow index (MFI), density, ash content, moisture content, molecular weight distribution, impact strength, and tensile strength.

[0076] Feature extraction is performed on the micro-active imprint data to obtain a first feature vector, and feature extraction is performed on the conventional physicochemical index data to obtain a second feature vector. This refers to extracting a set of numerical values, i.e., feature vectors, from the original, high-dimensional micro-active imprint data and conventional physicochemical index data through data dimensionality reduction, statistical analysis, or machine learning algorithms. For example, principal component analysis (PCA), independent component analysis (ICA), or deep learning models can be used to extract key features.

[0077] Fusing the first feature vector and the second feature vector to form the comprehensive feature vector of the raw material batch refers to combining feature vectors extracted from two different types of data to form a more comprehensive and representative comprehensive feature representation. The fusion method may include simple concatenation, weighted averaging, or deep fusion through neural networks, etc.

[0078] Calculating the similarity between the comprehensive feature vector and the preset feature vectors of multiple known difference patterns involves comparing the comprehensive feature vector of the current raw material batch with a pre-established feature vector library containing multiple known defect types (i.e., difference patterns). Similarity calculation can employ various metrics such as cosine similarity, Euclidean distance, and Mahalanobis distance. The preset feature vectors of known difference patterns are obtained by analyzing and modeling raw material batches known to have specific defects from a large amount of historical data.

[0079] Based on the similarity score, identifying the difference pattern to which the raw material batch belongs and generating a type identifier for that difference pattern means determining, based on the calculated similarity value, which known difference pattern the current raw material batch has the highest similarity to, and thus classifying it into the corresponding difference pattern. The type identifier for the difference pattern is a concise description of the defect pattern, such as "photosensitive defect," "thermal sensitive defect," or "oxidation-sensitive defect." The difference pattern refers to specific and actionable defect type information, meaning that the identified defect type is not merely an abstract classification, but also concrete information that can guide subsequent processing or production adjustments.

[0080] Secondly, referring to Figure 2 This application further proposes a data analysis system for the production process of light diffusion plates, the system comprising: The information acquisition module 201 is used to acquire various types of information throughout the entire production process of the light diffusion plate. These various types of information include batch characteristic information of raw materials, processing operation information during the production process, operating status information of production equipment, quality information of the light diffusion plate after production, and environmental condition information of the light diffusion plate in the warehousing process. The information association module 202 is used to associate the batch characteristic information of the raw materials with the processing operation information and the operating status information, associate the processing operation information and the operating status information with the off-line quality information, and associate the off-line quality information with the environmental condition information, so as to construct a complete life cycle record of the light diffusion plate. The quality judgment and prediction module 203 is used to judge the impact of raw material batch characteristics, processing operations, operating status and environmental conditions on the quality of the light diffusion plate based on the complete life cycle record, and output the probability data of the light diffusion plate undergoing quality degradation during subsequent storage and use.

[0081] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for analyzing data during the production process of a light diffusion plate, characterized in that, include: Acquire various types of information throughout the entire production process of light diffusion plates, including batch characteristic information of raw materials, processing operation information during production, operating status information of production equipment, quality information of light diffusion plates after production, and environmental conditions information of light diffusion plates in the warehousing process. The batch characteristic information of the raw materials is associated with the processing operation information and the operating status information. The processing operation information and the operating status information are associated with the off-line quality information. The off-line quality information is associated with the environmental condition information to construct a complete life cycle record of the light diffusion plate. Based on the complete lifecycle record, the impact of raw material batch characteristics, processing operations, operating status, and environmental conditions on the quality of the light diffusion plate is determined, and the probability data of quality degradation of the light diffusion plate during subsequent storage and use is output.

2. The data analysis method for the production process of a light diffusion plate according to claim 1, characterized in that, The steps of associating the batch characteristic information of the raw materials with the processing operation information and the operating status information, associating the processing operation information and the operating status information with the off-line quality information, and associating the off-line quality information with the environmental condition information to construct a complete lifecycle record for the light diffusion plate include: When batches of polycarbonate raw materials arrive at the factory, standardized accelerated aging stimulation is applied to the collected raw material samples in a microreactor, and the rate of change of the yellowness index of the raw material samples during the accelerated aging stimulation is monitored to quantify the inherent sensitivity of the batch of raw materials to light. The rate of change of the yellowness index is used as a micro-activity imprint of the raw material batch, and the micro-activity imprint is bound to the raw material batch number as batch characteristic information of the raw material, which is then integrated into the complete life cycle record of the light diffusion plate.

3. The data analysis method for the production process of a light diffusion plate according to claim 1, characterized in that, The step of associating the processing operation information, the operating status information, and the offline quality information, and associating the offline quality information with the environmental condition information to construct a complete lifecycle record for the light diffusion plate includes: Sensing units are deployed on the highly polished structural components of the new electric forklift; The intensity of the beam reflected by the forklift is monitored by the sensing unit; When the intensity of the reflected beam exceeds a preset threshold and the proportion of ultraviolet components increases abnormally, the control system of the new electric forklift calculates the projection area of ​​the reflected beam based on data from its built-in positioning module and attitude sensor. The novel electric forklift is controlled to send instantaneous exposure event commands with timestamps, peak intensity, and ultraviolet dose estimates to environmental response tags of all light diffusers within the projection area of ​​the reflected beam via a directional wireless communication module. The environmental response tag is controlled to record detailed information about the instantaneous exposure event command; The system receives and integrates instantaneous high-intensity environmental impact data from the novel electric forklift and the environmental response tag, and uses the instantaneous high-intensity environmental impact data as the environmental condition information to update the complete life cycle record of the light diffuser. The instantaneous high-intensity environmental impact data includes the timestamp, peak intensity, and UV dose estimate after integration from the novel electric forklift and the environmental response tag.

4. The data analysis method for the production process of a light diffusion plate according to claim 1, characterized in that, The step of determining the impact of raw material batch characteristics, processing operations, operating status, and environmental conditions on the quality of the light diffusion plate based on the complete lifecycle record, and outputting the probability data of quality degradation of the light diffusion plate during subsequent storage and use, includes: Obtain a complete lifecycle record for each batch of light diffusion plate products. This complete lifecycle record includes the micro-activity imprint of raw materials, the melt micro-stress imprint during the production process, instantaneous environmental exposure information during storage, and the molecular aging evolution trajectory. The micro-activity imprint refers to the inherent sensitivity of raw materials to light. The melt micro-stress imprint refers to the shear and tensile mechanical stresses experienced by the polymer melt during processing such as extrusion and injection molding. The instantaneous environmental exposure information refers to the captured localized, instantaneous high-intensity environmental impacts. Based on the information in the complete life cycle record, the molecular aging state evolution within the product is calculated and updated, including the cumulative concentration of oxidation products and the average molecular weight. Using the molecular aging state evolution as the predicted center value, and combining the micro-activity imprint of the raw material, the micro-stress imprint of the melt, and the fluctuation range and uncertainty of the instantaneous environmental exposure information, multiple simulated trajectories of the molecular aging state evolution are generated through Monte Carlo simulation. Based on the multiple simulated trajectories, the probability distribution of quality degradation of the light diffusion plate during subsequent storage and use is calculated. Based on the probability distribution, the quality degradation potential index, confidence interval, and corresponding risk classification of the light diffusion plate are generated.

5. The data analysis method for the production process of a light diffusion plate according to claim 1, characterized in that, The steps of associating the batch characteristic information of the raw materials with the processing operation information and the operating status information, associating the processing operation information and the operating status information with the off-line quality information, and associating the off-line quality information with the environmental condition information to construct a complete lifecycle record for the light diffusion plate include: Obtain a unique identifier for each batch of light diffusion plates; Based on the unique identifier, obtain the batch number of the raw materials used in the production process of this batch of products and the corresponding production time period; By analyzing the micro-activity imprints of the raw material batch number and combining them with the melt micro-stress imprints during the production period, the cumulative impact of raw material batch characteristics and production process on the initial molecular aging state of the product is estimated. Track the physical location changes of the light diffusion plate batch during the warehousing process and correlate them with the instantaneous environmental exposure information experienced by the light diffusion plate batch. Based on the instantaneous environmental exposure information accumulated by the light diffusion plate product during the warehousing process and the cumulative impact of the initial molecular aging state of the product, the secondary impact of the warehousing environment on the molecular aging state of the product is estimated. Based on the secondary impact of raw material batch characteristics, production process, and storage environment on the molecular aging state of the product, a nonlinear correlation is established between information from different stages, and the complete life cycle record of the light diffusion plate is updated.

6. The data analysis method for the production process of a light diffusion plate according to claim 5, characterized in that, The steps of analyzing the micro-activity imprint of the raw material batch number and combining it with the melt micro-stress imprint during the production period to estimate the cumulative impact of raw material batch characteristics and the production process on the initial molecular aging state of the product include: Receive the micro-active imprints and conventional physicochemical properties of raw material batch numbers, as well as the melt micro-stress imprints during the production process; The micro-active imprint is compared with the conventional physicochemical indicators to identify whether there are significant differences between the micro-active imprint and the conventional physicochemical indicators; When there is a significant difference between the micro-active imprint and the conventional physicochemical indicators, the raw material batch will be marked as having a potential intrinsic defect; The micro-stress imprint of the melt is monitored in real time to identify whether there are local abnormally high values ​​in the micro-stress imprint of the melt. When a local abnormally high value is identified in the micro-stress imprint of the melt, the equipment operation log during the production period is used to analyze whether the local abnormally high value coincides with the instantaneous fluctuation event of the equipment in time. When the local abnormally high value coincides with the instantaneous fluctuation event of the equipment, the abnormality of the melt micro-stress imprint is attributed to the instantaneous fluctuation of the production process. When the local abnormally high value does not coincide with the instantaneous fluctuation event of the equipment, the abnormality of the melt micro-stress imprint is attributed to the inherent defects of the raw materials that manifest during the processing, in conjunction with the potential inherent defect markers of the raw material batch. The cumulative effects of inherent defects in raw materials and instantaneous fluctuations in the production process on the initial molecular aging state of the product are calculated and differentiated.

7. The data analysis method for the production process of a light diffusion plate according to claim 6, characterized in that, The step of marking the raw material batch as having a potential intrinsic defect when there is a significant difference between the micro-active imprint and the conventional physicochemical indicators includes: The differences in micro-active imprints and conventional physicochemical indicators between batches of raw materials were analyzed, and the differential characteristics were extracted. Based on the extracted differential features, a search is performed in a preset defect type mapping table to obtain the matching defect type. The defect type mapping table sets the correspondence between different differential features and defect types. The defect types include photosensitive defects, thermosensitive defects, and oxidation-sensitive defects. Based on the matched defect type and the degree of impact of the defect type on the long-term quality deterioration of the product in historical data, the risk level of the raw material batch is calculated. The defect type and risk level are bound to the raw material batch number to generate a refined defect marker, which is then integrated into the complete lifecycle record of the light diffusion plate. During production planning and scheduling, based on the refined defect marking, high-risk batches of raw materials are allocated to products with preset quality requirement levels for production, or a special pre-processing procedure for high-risk batches of raw materials is triggered.

8. The data analysis method for the production process of a light diffusion plate according to claim 7, characterized in that, The steps of allocating high-risk batches of raw materials to products with preset quality requirements for production production based on the refined defect markings, or triggering a special pretreatment process for high-risk batches of raw materials during production planning and scheduling, include: Obtain a list of high-risk batches of raw materials currently awaiting scheduling, the list of high-risk batches of raw materials including detailed defect markers and available quantities for each batch; Retrieve the quality requirement level for all products in the current production order pool; For each high-risk batch of raw materials, the overall quality risk introduced during the production of products with different quality requirements is calculated based on the corresponding refined defect markers. For each production order, calculate the production efficiency when allocating raw materials of different risk levels according to the corresponding quality requirement level; A multi-objective optimization function is constructed under preset constraints, the multi-objective optimization function aiming to minimize overall quality risk and maximize production efficiency; the preset constraints include the availability of raw materials, the demand of production orders, and the capacity of equipment. Through iterative search and evaluation, a solution set of the multi-objective optimization function is found, which represents a raw material allocation scheme under the premise of balancing quality risk and production efficiency; The optimal raw material allocation scheme is selected from the solution set, and the allocation instruction for the high-risk batch of raw materials is generated. The allocation instruction includes each high-risk batch of raw materials and the specific product production task assigned to it.

9. The data analysis method for the production process of a light diffusion plate according to claim 6, characterized in that, The step of comparing the micro-active imprint with the conventional physicochemical indicators to identify whether there are significant differences between the micro-active imprint and the conventional physicochemical indicators includes: Obtain micro-active imprint data and conventional physicochemical index data for each batch of raw materials; Feature extraction is performed on the micro-active imprint data to obtain a first feature vector; Feature extraction is performed on the conventional physicochemical index data to obtain a second feature vector; The first feature vector and the second feature vector are fused to form a comprehensive feature vector for the raw material batch; Calculate the similarity between the comprehensive feature vector and the preset feature vectors of multiple known difference patterns; Based on the similarity, the difference pattern to which the raw material batch belongs is identified, and a type identifier for the difference pattern is generated; the difference pattern refers to specific and actionable defect type information.

10. A data analysis system for the production process of light diffusion plates, used to analyze data from the production process of light diffusion plates, characterized in that, The system includes: The information acquisition module is used to acquire various types of information throughout the entire production process of the light diffusion plate. These various types of information include batch characteristic information of raw materials, processing operation information during the production process, operating status information of production equipment, quality information of the light diffusion plate after production, and environmental condition information of the light diffusion plate in the warehousing process. The information association module is used to associate the batch characteristic information of the raw materials with the processing operation information and the operating status information, associate the processing operation information and the operating status information with the off-line quality information, and associate the off-line quality information with the environmental condition information, so as to construct a complete life cycle record of the light diffusion plate. The quality judgment and prediction module is used to determine the impact of raw material batch characteristics, processing operations, operating status and environmental conditions on the quality of the light diffusion plate based on the complete life cycle record, and output the probability data of the light diffusion plate undergoing quality degradation during subsequent storage and use.

Images