A smart supply chain optimization scheduling method and system for carbon fiber materials
Patent Information
- Application Number
- CN202610443963.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-07
- Publication Date
- 2026-07-03
Smart Images
Figure CN122334824A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of intelligent supply chain scheduling technology, and more specifically, to an intelligent supply chain optimization scheduling method and system for carbon fiber materials. Background Technology
[0002] The production process of carbon fiber materials is complex and demands extremely high quality, presenting numerous challenges to its supply chain management. While traditional intelligent supply chain scheduling systems can optimize production and logistics, they often struggle to detect and assess the potential impact of minor process deviations on product quality in real time. This can lead to the system continuing to schedule downstream high-value resources according to a predetermined plan even when product quality risks exist, resulting in resource waste and delivery risks.
[0003] For example, the core of existing intelligent supply chain optimization scheduling for carbon fiber materials is to coordinate and plan the entire production and logistics process based on downstream customer orders. Under normal operating conditions, the system generates the optimal scheduling plan based on historical production data, material inventory information at each stage, standard processing efficiency of equipment, and average logistics transportation time.
[0004] However, in actual production, the smooth operation of the supply chain can easily be disrupted by special circumstances. For example, when the system receives an urgent order with high-quality requirements, adjustments to the production process are needed to meet the technical specifications of the urgent order. During the production of this order, the system responsible for monitoring the operation of the carbonization furnace detected a minor deviation in parameters. However, this deviation is within the acceptable process window for producing ordinary carbon fiber products, and therefore it is not considered a serious fault requiring immediate shutdown. However, for carbon fiber products with current high-quality requirements, even a minor parameter deviation is enough to have an irreversible impact on the product.
[0005] There is currently no effective technical solution to the above problems. Summary of the Invention
[0006] This application aims to solve at least one of the technical problems mentioned in the background art, and provides an intelligent supply chain optimization scheduling method and system for carbon fiber materials. Its purpose is to improve the ability to proactively respond to potential quality risks and delivery pressures in the production process when handling high-priority orders for carbon fiber production.
[0007] To solve the above-mentioned technical problems, the solution proposed in this application is as follows: As one aspect of this application, a smart supply chain optimization scheduling method for carbon fiber materials is provided. This method is used to optimize the scheduling of carbon fiber production orders executed on a carbon fiber production line. The method includes: Obtain process data of carbon fiber materials during the production process, the process data including at least one of carbonization furnace temperature zone temperature, protective gas flow rate and carbon fiber tension; The performance of carbon fiber products is evaluated based on the obtained process data, and the performance evaluation results are obtained. Obtain the order delivery information for the current carbon fiber production order, wherein the order delivery information includes the promised delivery date, production schedule, and information on available alternative production lines; By combining performance evaluation results with order delivery information, a delivery confidence index is generated. Based on the delivery confidence index, determine whether there is any risk to order delivery.
[0008] Respond to order delivery risks and develop corresponding strategies.
[0009] Furthermore, obtaining process data of carbon fiber materials during the production process includes: Detection sensors deployed on the carbon fiber production line are used to collect continuous process data at a set frequency. Interpolation is used to process continuous process data to obtain processed process data.
[0010] Furthermore, obtaining process data of carbon fiber materials during the production process includes: Using sensor arrays deployed on the carbon fiber material production line, the output data of the sensing elements in each sensor array is continuously monitored, and the output data is compared in real time to obtain the comparison results. When the comparison result indicates that the output data of a sensor element deviates from the output data of the remaining sensor elements belonging to the same sensor array, the sensor element is identified as a failed element, and the identification result is output. Based on the identification results, the output data of the sensing element is corrected or excluded, and the output data of the remaining sensing elements under the same sensor array is used as the process data of carbon fiber material in the production process.
[0011] Furthermore, the evaluation of the performance of the carbon fiber product based on the obtained process data, to obtain the performance evaluation results, includes: The process data is input into a preset carbon fiber production evolution module, wherein the preset carbon fiber production evolution module is used to simulate the carbon fiber production structure evolution process and predict the structural mechanical properties of the produced carbon fiber based on the input process data. The carbon fiber production quality qualification index is quantitatively output based on the structural mechanical properties of the produced carbon fiber, and is used as the performance evaluation result.
[0012] Furthermore, the step of quantitatively outputting a carbon fiber production quality qualification index based on the structural mechanical properties of the produced carbon fiber, and using it as a performance evaluation result, includes: Energy excitation is applied to the precursor material entering the carbon fiber production line, and response data about the precursor material is collected. The response data includes the instantaneous temperature change value, surface potential shift, infrared absorption spectrum intensity change, and trace gas release concentration of the precursor material. The response data is converted into quantitative characteristics of the precursor material's response under energy excitation. These quantitative characteristics include instantaneous temperature change rate, peak and duration of surface potential shift, shift of characteristic peaks in infrared absorption spectrum, peak and decay rate of trace gas release concentration. The quantitative characteristics are compared with preset precursor material behavior patterns to identify whether there are any unidentified intrinsic differences in the current precursor material. When an unidentified intrinsic difference is identified in the current precursor material, a correction coefficient is determined based on the degree of deviation between the response quantification characteristics corresponding to the intrinsic difference and the preset precursor material behavior pattern, combined with the production practice data of different types of precursor materials. The carbon fiber production quality qualification index is then corrected based on this correction coefficient.
[0013] Furthermore, after quantifying the carbon fiber production quality qualification index based on the structural mechanical properties of the produced carbon fiber and using it as a performance evaluation result, the process also includes: Energy excitation is applied to the precursor material entering the carbon fiber production line, and response data of the precursor material is collected. The response data specifically includes the instantaneous temperature change value, surface potential shift, infrared absorption spectrum intensity change, and trace gas release concentration of the precursor material. A correlation model between Raman features and ultrasonic features is pre-established. This correlation model is used to realize the mutual conversion between Raman feature vectors and ultrasonic feature vectors. The Raman feature vector is obtained by inputting the response data of the precursor material into a Raman spectrometer. Based on the obtained Raman feature vector, the predicted ultrasonic feature vector is obtained by using the correlation model between Raman features and ultrasonic features. The actual ultrasonic feature vector is obtained by inputting the response data of the precursor material into an ultra-high frequency acoustic attenuation sensor array. Determine the consistency deviation between the actual ultrasonic feature vector and the predicted ultrasonic feature vector. When the consistency deviation exceeds a preset threshold, it is confirmed that the acquired precursor material response data is subject to external interference. Based on the consistency deviation, the corresponding correction coefficient is found from the preset correction factor list, and the carbon fiber production quality qualification index is corrected according to the correction coefficient.
[0014] Furthermore, the generation of delivery confidence metrics by combining performance evaluation results and order delivery information includes: The order time margin percentage is calculated based on the delivery date and production schedule, and the available alternative path index is calculated based on the information on alternative production lines. The delivery confidence index is calculated by combining the carbon fiber production quality qualification index, order time margin percentage, and available alternative path index using a weighted summation algorithm. The delivery confidence index is obtained according to the following calculation formula:
[0015] , Preset weighting coefficients and satisfying + + =1; A indicates the carbon fiber production quality qualification index, which is a value between 0 and 1. B indicates the order time margin percentage, which is the ratio of the buffer time between the current remaining production cycle and the promised delivery time, and is a value between 0 and 1. C indicates the available alternative path index, which is a value between 0 and 1.
[0016] Furthermore, the determination of whether there is order delivery risk based on the delivery confidence index includes: Preset delivery confidence risk threshold; When the delivery confidence index falls below the preset delivery confidence risk threshold, an order delivery risk is confirmed.
[0017] Furthermore, the response strategies include: The initiation of downstream resources for carbon fiber production orders may be suspended or delayed. These downstream resources include at least one of special resin formulation and special temperature-controlled logistics vehicle reservation. Provide suggestions for adjusting process parameters during the production of carbon fiber materials; And generate customer communication reports.
[0018] As a second aspect of this application, an intelligent supply chain optimization scheduling system for carbon fiber materials is provided. This system is used to optimize the scheduling of carbon fiber production orders executed on the carbon fiber production line. The system includes: A process data acquisition module is used to acquire process data of carbon fiber materials during the production process. The process data includes at least one of the following: carbonization furnace temperature zone temperature, protective gas flow rate, and carbon fiber tension. A performance evaluation module is used to evaluate the performance of carbon fiber products based on the obtained process data and obtain performance evaluation results. The order delivery information acquisition module is used to acquire the order delivery information of the current carbon fiber production order, wherein the order delivery information includes the promised delivery date, production progress and available alternative production line information; The delivery confidence metric generation module is used to generate delivery confidence metrics by combining performance evaluation results and order delivery information. A risk identification module is used to determine whether there is any risk to order delivery based on a delivery confidence index. The strategy output module is used to respond to order delivery risks and output corresponding strategies.
[0019] This application discloses an intelligent supply chain optimization scheduling method and system for carbon fiber materials. It acquires process data during the carbon fiber production process and evaluates the performance of carbon fiber products based on this data to obtain performance evaluation results. Simultaneously, it acquires order delivery information for current carbon fiber production orders and, combined with the performance evaluation results and order delivery information, generates a delivery confidence index. This index is used to determine whether there are any order delivery risks. Finally, it outputs response strategies to address these risks. This method improves the ability to proactively respond to potential quality risks and delivery pressures during the production process when handling high-priority carbon fiber production orders. Through real-time process data monitoring and performance evaluation, potential quality risks can be identified early and combined with order delivery information to form a quantified delivery confidence index. This avoids resource waste and delivery risks caused by scheduling downstream high-value resources according to the established plan even when product quality risks exist. Attached Figure Description
[0020] Figure 1 A flowchart illustrating an intelligent supply chain optimization scheduling method for carbon fiber materials provided in this application embodiment; Figure 2 A system structure diagram of an intelligent supply chain optimization scheduling system for carbon fiber materials provided in this application embodiment; Figure reference numerals: 100, Intelligent supply chain optimization and scheduling system for carbon fiber materials; 101, Process data acquisition module; 102, Performance evaluation module; 103, Order delivery information acquisition module; 104, Delivery confidence index generation module; 105, Risk identification module; 106, Strategy output module. Detailed Implementation
[0021] To better illustrate the present invention, the invention will now be described in further detail with reference to the accompanying drawings.
[0022] It should be understood that, in order to make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. The components of the embodiments of this disclosure 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 disclosure provided in the accompanying drawings is not intended to limit the scope of the claimed disclosure, but merely represents selected embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without inventive effort are within the scope of protection of this disclosure.
[0023] The following description uses at least one specific embodiment as an example. In this embodiment: Firstly, such as Figure 1 As shown, an intelligent supply chain optimization scheduling method for carbon fiber materials is provided. This method is used to optimize the scheduling of carbon fiber production orders executed on the carbon fiber production line. The method includes the following steps: S1. Obtain process data of carbon fiber material during production, wherein the process data includes at least one of carbonization furnace temperature zone temperature, protective gas flow rate and carbon fiber tension. S2. Evaluate the performance of carbon fiber products based on the obtained process data and obtain the performance evaluation results; S3. Obtain the order delivery information of the current carbon fiber production order, wherein the order delivery information includes the promised delivery date, production progress and available alternative production line information; S4. Combine performance evaluation results with order delivery information to generate delivery confidence metrics; S5. Based on the delivery confidence index, determine whether there is any risk to order delivery.
[0024] S6. Respond to order delivery risks and provide corresponding strategies.
[0025] To implement this method, the first step is to acquire process data on carbon fiber materials during production. This process data are key parameters reflecting the production status and product quality of carbon fiber. For example, the temperature of the carbonization furnace can be acquired, which directly affects the graphitization degree and microstructure of the carbon fiber; the flow rate of the protective gas can also be acquired, which is crucial for atmosphere control and impurity removal during the carbonization process; and the carbon fiber tension can be acquired, which affects fiber orientation and mechanical properties. All of this process data can serve as the basis for evaluating the performance of carbon fiber products. For instance, this can be done manually, with operators periodically inspecting and recording temperature readings of each zone of the carbonization furnace, the readings of the protective gas flow meter, and the readings of the tension sensor. Another approach is to deploy various detection sensors on the carbon fiber production line, such as thermocouples for temperature measurement, flow meters for gas flow measurement, and tension sensors for fiber tension measurement. These sensors can collect continuous process data in real time or at a preset frequency.
[0026] Next, the performance of carbon fiber products is evaluated based on the obtained process data, and performance evaluation results are obtained. The aim is to transform abstract process data into a quantitative assessment of product quality. For example, process data such as carbonization furnace temperature, protective gas flow rate, and carbon fiber tension can be mapped to performance indicators such as tensile strength and modulus of carbon fiber using empirical formulas or pre-set lookup tables.
[0027] Simultaneously, the system retrieves order delivery information for current carbon fiber production orders. This information can include the promised delivery date (the date the customer expects to receive the products), production progress (the current order's completion status on the production line), and information on available alternative production lines (whether other production lines can take over if the main production line experiences problems). For example, this can be done manually by order managers inputting the promised delivery date from the customer's order, the current production progress reported by the production workshop, and a list of available alternative production lines within the factory. Alternatively, this order delivery information can be automatically retrieved by interfaceing with an Enterprise Resource Planning (ERP) system.
[0028] Then, combining the performance evaluation results and order delivery information, a delivery confidence index is generated. The delivery confidence index is a comprehensive quantitative indicator used to reflect the likelihood of on-time and high-quality order delivery. For example, a weighted average algorithm can be used to comprehensively calculate a value between 0 and 100, based on performance evaluation results (e.g., a quantified quality compliance index), order time margin percentage (e.g., the ratio of the current remaining production cycle to the promised delivery time), and available alternative path index (e.g., the number or capacity of alternative production lines), resulting in a delivery confidence index. A delivery confidence risk threshold, such as 60 points, can then be preset. When the calculated delivery confidence index falls below this threshold, the system determines that there is a risk to order delivery.
[0029] The method proposed in this application combines detailed process data from carbon fiber production with order delivery information to generate a quantitative delivery confidence index, thereby determining whether there is any risk in order delivery and achieving intelligent optimization and scheduling of the supply chain.
[0030] Furthermore, the acquisition of process data for carbon fiber materials during production can be further refined into the following steps: Detection sensors deployed on the carbon fiber production line are used to collect continuous process data at a set frequency. Interpolation is used to process continuous process data to obtain processed process data.
[0031] The aforementioned detection sensor can be understood as a physical device used to monitor various key process parameters in the carbon fiber production process in real time. As a preferred embodiment, the detection sensor can be a thermocouple, flow meter, tension sensor, etc., and its selection and deployment should be based on specific production process requirements and data acquisition accuracy needs.
[0032] Acquiring continuous process data at a set frequency means that sensors capture dynamic changes in process parameters during production at preset time intervals, such as per second, per minute, or shorter intervals, to provide detailed time-series data for subsequent performance evaluation.
[0033] Interpolation is used to process continuous process data to obtain processed process data. The purpose of this method is to address potential missing or anomaly issues that may occur during data acquisition. By estimating and filling in missing or anomaly data through interpolation, complete and smooth processed process data is obtained, ensuring the reliability of the data used in subsequent performance evaluations.
[0034] Based on the above-described embodiments, process data of carbon fiber materials during production is acquired and used for subsequent performance evaluation and risk assessment. However, in actual production environments, detection sensors deployed on production lines may malfunction or experience data deviations for various reasons, leading to inaccurate process data and potentially causing deviations in delivery confidence indicators, resulting in misjudgments of order delivery risks.
[0035] To ensure the accuracy of the acquired data, the acquisition of process data for carbon fiber materials during production includes: Using sensor arrays deployed on the carbon fiber material production line, the output data of the sensing elements in each sensor array is continuously monitored, and the output data is compared in real time to obtain the comparison results. When the comparison result indicates that the output data of a sensor element deviates from the output data of the remaining sensor elements belonging to the same sensor array, the sensor element is identified as a failed element, and the identification result is output. Based on the identification results, the output data of the sensing element is corrected or excluded, and the output data of the remaining sensing elements under the same sensor array is used as the process data of carbon fiber material in the production process.
[0036] Continuous monitoring refers to each sensing element in the sensor array continuously collecting and outputting data at a preset frequency.
[0037] Real-time consistency comparison refers to comparing and analyzing data collected by different sensing elements in the same sensor array at the same point in time or within a similar time period to determine whether there are differences between these data.
[0038] When the comparison results indicate that the output data of a certain sensing element deviates from the output data of other sensing elements in the same array beyond a preset range, that sensing element is identified as a failed element. The identification results may include information such as the ID of the failed element, the time of failure, and the degree of deviation.
[0039] Based on the identification results, the output data of the failed component can be corrected, for example, through interpolation, regression analysis, or estimation using data from other normal sensing components in the same array; or the output data of the failed component can be directly excluded and no longer used for subsequent process data processing. Finally, the output data of the remaining sensing components considered to be normal in the same sensor array, after appropriate fusion or processing, serves as reliable process data for the production of carbon fiber materials.
[0040] The solution proposed in this application solves the problem that data deviations from a single sensor may affect the subsequent performance evaluation of carbon fiber products by introducing a sensor array and a real-time consistency comparison mechanism.
[0041] In some preferred embodiments, it is assumed that a sensor array consisting of three temperature sensors is deployed in the carbonization furnace temperature zone of a carbon fiber production line. These three temperature sensors continuously monitor the real-time temperature of this zone. The system collects the output data of the three sensors once per second and performs real-time consistency comparison. For example, at a certain moment, sensor 1 outputs 200°C, sensor 2 outputs 201°C, and sensor 3 outputs 250°C. The system detects a significant deviation between the output data of sensor 3 and the output data of sensors 1 and 2 (e.g., exceeding a preset 5°C threshold). At this time, the system identifies sensor 3 as a failed component. To ensure the accuracy of process data, the system can choose to exclude the output data of sensor 3 and use only the average value of sensors 1 and 2 (e.g., 200.5°C) as the real-time process temperature data for this zone. Alternatively, the system can also attempt to correct the data of sensor 3 based on the data of sensors 1 and 2, for example, through weighted averaging or predictive correction using historical data models. In this way, even if some sensors fail, the system can still obtain reliable process data, avoiding problems in the monitoring and scheduling of the entire production process due to the failure of a single sensor.
[0042] Based on the above embodiments, the step of evaluating the performance of carbon fiber products according to the obtained process data and obtaining performance evaluation results specifically includes: The process data is input into a preset carbon fiber production evolution module, wherein the preset carbon fiber production evolution module is used to simulate the carbon fiber production structure evolution process and predict the structural mechanical properties of the produced carbon fiber based on the input process data. The carbon fiber production quality qualification index is quantitatively output based on the structural mechanical properties of the produced carbon fiber, and is used as the performance evaluation result.
[0043] Specifically, the purpose of the preset carbon fiber production evolution module is to transform the input process data into predictions of the performance of carbon fiber products. This module can be built using data-driven models such as machine learning and deep learning. Through this module, the structural evolution process of carbon fiber under different process conditions can be simulated, such as changes in microstructural characteristics like fiber orientation, and its macroscopic mechanical properties, such as tensile strength, elastic modulus, and elongation at break, can be predicted.
[0044] Furthermore, based on the predicted structural mechanical properties of carbon fibers, a carbon fiber production quality qualification index can be quantitatively output to measure the quality level of the current production batch of carbon fibers. This index can be calculated based on a weighted average of multiple mechanical performance indicators, or by comparing it with preset quality standards. This carbon fiber production quality qualification index directly serves as a performance evaluation result, providing a quantitative basis for generating confidence indicators for subsequent order delivery.
[0045] Furthermore, the quantitative output of the carbon fiber production quality qualification index based on the structural mechanical properties of the produced carbon fiber, and its use as a performance evaluation result, includes: Energy excitation is applied to the precursor material entering the carbon fiber production line, and response data about the precursor material is collected. The response data includes the instantaneous temperature change value, surface potential shift, infrared absorption spectrum intensity change, and trace gas release concentration of the precursor material. The response data is converted into quantitative characteristics of the precursor material's response under energy excitation. These quantitative characteristics include instantaneous temperature change rate, peak and duration of surface potential shift, shift of characteristic peaks in infrared absorption spectrum, peak and decay rate of trace gas release concentration. The quantitative characteristics are compared with preset precursor material behavior patterns to identify whether there are any unidentified intrinsic differences in the current precursor material. When an unidentified intrinsic difference is identified in the current precursor material, a correction coefficient is determined based on the degree of deviation between the response quantification characteristics corresponding to the intrinsic difference and the preset precursor material behavior pattern, combined with the production practice data of different types of precursor materials. The carbon fiber production quality qualification index is then corrected based on this correction coefficient.
[0046] Specifically, precursor materials are those used in the preparation of carbon fibers. Applying energy excitation to the precursor materials entering the carbon fiber production line refers to exciting the precursor materials through external energy input. The purpose is to induce microstructural or chemical bond responses within the precursor materials, thereby revealing their potential physicochemical properties. The acquisition of precursor material response data involves using sensors and detection equipment to capture in real-time physicochemical parameters such as instantaneous temperature changes, surface potential shifts, infrared absorption spectrum intensity changes, and trace gas release concentrations under energy excitation. These data directly reflect the intrinsic properties of the precursor materials.
[0047] Furthermore, converting the response data into quantitative characteristics of the physicochemical response of the precursor material under energy excitation refers to processing and analyzing the collected raw instantaneous data to extract quantitative indicators that characterize the properties of the precursor material. For example, instantaneous temperature change values can be converted into instantaneous temperature change rates, surface potential shifts can be extracted for their peak values and durations, changes in infrared absorption spectrum intensity can be analyzed for the shift of characteristic peaks, and trace gas release concentrations can be extracted for their peak values and decay rates. This can describe the behavior patterns of the precursor material under energy excitation.
[0048] In practical applications, the quantitative characteristics of the physicochemical response are compared with a preset precursor material behavior pattern to identify any unidentified intrinsic differences in the current precursor material. This preset precursor material behavior pattern is a reference model established based on historical production data, experimental analysis, and the performance of different batches of precursor materials. By comparing the quantitative characteristics of the current batch of precursor materials with this behavior pattern, significant deviations can be identified, thereby determining whether the current precursor material has any unidentified intrinsic defects that may affect the performance of the final product.
[0049] When unidentified intrinsic differences are found in the precursor materials used to prepare carbon fibers, a correction coefficient is determined based on the degree of deviation between the quantified response characteristics corresponding to these intrinsic differences and the preset behavior patterns of the precursor materials, combined with production practice data of different types of precursor materials. Specifically, the greater the deviation, the more significant the intrinsic difference in the precursor materials. By combining historical production practice data, a mapping relationship between the degree of deviation and its impact on the performance of the final product can be established, thereby determining a correction coefficient. This correction coefficient is then used to correct the carbon fiber production quality qualification index to ensure the accuracy of the performance evaluation results.
[0050] As a specific implementation method, it is assumed that in carbon fiber production, the precursor material is excited by laser energy of a specific wavelength before entering the carbon fiber production line. During this process, the instantaneous temperature change, surface potential shift, infrared absorption spectrum intensity change, and trace gas release concentration of the precursor material are collected in real time using a high-precision infrared spectrometer, surface potential sensor, and trace gas analyzer. These instantaneous data are then processed to extract physicochemical response quantitative characteristics such as the instantaneous temperature change rate, surface potential shift peak and duration, infrared absorption spectrum characteristic peak shift, and trace gas release concentration peak and decay rate. These quantitative characteristics are compared with a pre-established "behavioral pattern" based on a standard batch of precursor material. For example, if a significant deviation is found in the infrared absorption spectrum characteristic peak shift of the batch of precursor material from the standard pattern, and this deviation matches a known intrinsic defect pattern in historical data that leads to a decrease in carbon fiber strength, the system will identify this "unidentified intrinsic difference." Based on the degree of deviation and historical production practice data (e.g., this degree of deviation typically results in a 5% decrease in the strength of the final product), the system will determine a correction factor (e.g., 0.95). Finally, the carbon fiber production quality compliance index (e.g., 0.85) initially calculated by the carbon fiber production evolution module will be multiplied by this correction factor to obtain the corrected quality compliance index (0.8075).
[0051] Furthermore, traditional methods for assessing the production quality of carbon fiber may face the problem of external interference affecting the response data of precursor materials when predicting the structural mechanical properties of carbon fiber based on process data and quantifying the production quality qualification index. To address this, this application proposes introducing a multimodal data cross-validation mechanism to identify and correct the impact of external interference on precursor material data.
[0052] Specifically, after quantifying the carbon fiber production quality qualification index based on the structural mechanical properties of the produced carbon fiber and using it as a performance evaluation result, the process also includes: Energy excitation is applied to the precursor material entering the carbon fiber production line, and response data of the precursor material is collected. The response data specifically includes the instantaneous temperature change value, surface potential shift, infrared absorption spectrum intensity change, and trace gas release concentration of the precursor material. A correlation model between Raman features and ultrasonic features is pre-established. This correlation model is used to realize the mutual conversion between Raman feature vectors and ultrasonic feature vectors. The Raman feature vector is obtained by inputting the response data of the precursor material into a Raman spectrometer. Based on the obtained Raman feature vector, the predicted ultrasonic feature vector is obtained by using the correlation model between Raman features and ultrasonic features. The actual ultrasonic feature vector is obtained by inputting the response data of the precursor material into an ultra-high frequency acoustic attenuation sensor array. Determine the consistency deviation between the actual ultrasonic feature vector and the predicted ultrasonic feature vector. When the consistency deviation exceeds a preset threshold, it is confirmed that the acquired precursor material response data is subject to external interference. Based on the consistency deviation, the corresponding correction coefficient is found from the preset correction factor list, and the carbon fiber production quality qualification index is corrected according to the correction coefficient.
[0053] Applying energy excitation to precursor materials aims to induce physicochemical responses that reflect the intrinsic properties of the materials.
[0054] The correlation model between Raman and ultrasonic features represents a multimodal data fusion and transformation mechanism. This model uses machine learning algorithms to learn the intrinsic mapping relationship between Raman spectral data and ultrasonic attenuation data. In this embodiment, its purpose is to accurately predict data of another modality (such as ultrasonic feature vectors) when data of one modality (e.g., Raman feature vectors) is acquired, thereby providing a benchmark for data consistency verification.
[0055] In practical applications, Raman eigenvectors are acquired using a Raman spectrometer, while the actual ultrasonic eigenvectors are acquired using an ultra-high frequency (UHF) acoustic attenuation sensor array. The Raman spectrometer provides molecular structure information of the material by analyzing scattered light, while the UHF acoustic attenuation sensor array reflects the material's physical structure and defect information by measuring the attenuation of sound waves within the material. Furthermore, by comparing the consistency deviation between the predicted and actual ultrasonic eigenvectors, external interference can be effectively detected. If the deviation exceeds a preset threshold, it indicates that the original acquired response data may be abnormal. Once external interference is confirmed, the system will search for the corresponding correction coefficient from a preset correction factor list based on the degree of deviation. This correction coefficient is used to quantify the impact of interference on the quality pass index and adjust the index to eliminate or mitigate the negative impact of the interference.
[0056] In some preferred embodiments, it is assumed that during the carbon fiber production process, the infrared absorption spectroscopy sensor used to collect the response data of the precursor material is affected by ambient temperature fluctuations, causing abnormal changes in the intensity of its output infrared absorption spectrum. According to the method described above, energy excitation is first applied to the precursor material, and response data, including the abnormal infrared data, is collected. This data is input into a Raman spectrometer to obtain Raman feature vectors. Using a pre-established correlation model between Raman and ultrasonic features, the Raman feature vector is converted into a predicted ultrasonic feature vector. Simultaneously, an ultra-high frequency acoustic attenuation sensor array independently collects the actual ultrasonic feature vector of the precursor material. When the system detects a significant consistency deviation between the predicted and actual ultrasonic feature vectors, and this deviation exceeds a preset threshold, it confirms that the originally collected response data (especially infrared data) has been subjected to external interference. At this point, the system searches for a corresponding correction coefficient from a preset correction factor list based on the degree of deviation; for example, a correction coefficient for temperature-induced interference with infrared spectral data. Ultimately, the carbon fiber production quality qualification index was corrected based on this correction factor, thereby eliminating the evaluation error caused by abnormal infrared sensor data and ensuring the accuracy of quality assessment.
[0057] In this embodiment, the step of generating a delivery confidence index by combining performance evaluation results and order delivery information includes: The order time margin percentage is calculated based on the delivery date and production schedule, and the available alternative path index is calculated based on the information on alternative production lines. The delivery confidence index is calculated by combining the carbon fiber production quality qualification index, order time margin percentage, and available alternative path index using a weighted summation algorithm. The delivery confidence index is obtained according to the following calculation formula:
[0058] , Preset weighting coefficients and satisfying + + =1; A indicates the carbon fiber production quality qualification index, which is a value between 0 and 1. B indicates the order time margin percentage, which is the ratio of the buffer time between the current remaining production cycle and the promised delivery time, and is a value between 0 and 1. C indicates the available alternative path index, which is a value between 0 and 1.
[0059] Specifically, the order time margin percentage refers to the ratio of the buffer time between the current remaining production cycle and the promised delivery time. This metric quantifies the flexibility of an order in the time dimension, reflecting the degree of leeway in production schedule relative to the delivery date.
[0060] The available alternative path index (ALOS) measures the number or capacity of other production lines available to complete an order if the current production line experiences a problem. It reflects the likelihood that an order can be transferred to another production line and delivered on time in the event of an unforeseen situation. For example, if there are N alternative production lines, each with different capacities and suitability, the ALOS index C can be quantified by comprehensively evaluating factors such as the availability and capacity matching of these alternative production lines. Its value ranges from 0 to 1; a higher value indicates stronger substitutability and lower delivery risk.
[0061] In practical applications, the delivery confidence index L is calculated using a weighted summation algorithm. This algorithm comprehensively considers three key factors: the carbon fiber production quality compliance index A, the order time margin percentage B, and the available alternative path index C.
[0062] In some preferred embodiments, it is assumed that the carbon fiber production quality qualification index A of a certain carbon fiber production order is 0.95 (indicating good product quality), the order time margin percentage B is 0.70 (indicating sufficient time buffer), and the available alternative path index C is 0.60 (indicating the existence of certain alternative production capacity). The preset weighting coefficients are W1=0.5 (quality factor is the most important), W2=0.3 (time factor is the second most important), and W3=0.2 (flexibility factor).
[0063] Based on the above calculation formula, the delivery confidence index L will be calculated as follows: L=(0.5×0.95+0.3×0.70+0.2×0.60)×100=80.5; Furthermore, based on delivery confidence indicators, we determine whether there are any order delivery risks, including: Preset delivery confidence risk threshold; When the delivery confidence index falls below the preset delivery confidence risk threshold, an order delivery risk is confirmed.
[0064] Based on the preferred example above, the delivery confidence index for this order is 80.5. If the preset delivery confidence risk threshold is 75, then the delivery confidence index for this order is higher than the threshold, indicating that the current order delivery risk is low. Conversely, if the calculated L value is lower than the threshold, then an order delivery risk will be confirmed, and corresponding response strategies will be triggered.
[0065] Furthermore, in the step of responding to order delivery risks and outputting response strategies, the response strategies include: The initiation of downstream resources for carbon fiber production orders may be suspended or delayed. These downstream resources include at least one of special resin formulation and special temperature-controlled logistics vehicle reservation. Provide suggestions for adjusting process parameters during the production of carbon fiber materials; And generate customer communication reports.
[0066] Specifically, "suspending or delaying the initiation of downstream resources for carbon fiber production orders" means that once an order delivery risk is identified, the system will automatically or semi-automatically block or postpone the allocation of resources related to subsequent stages of the order. Here, "downstream resources for carbon fiber production orders" can be understood as all resources required to achieve final product delivery or further processing after carbon fiber production is completed. Its scope can be adjusted according to the actual production process, but it must at least include one of the following: special resin formulation and reservation of special temperature-controlled logistics vehicles.
[0067] Furthermore, the "output of adjustment suggestions for process parameters in the production process of carbon fiber materials" refers to generating adjustment schemes for process parameters on the carbon fiber production line based on the specific causes of delivery risks, combined with historical data and preset optimization models. Its purpose is to reduce delivery risks by optimizing the production process to improve product quality or accelerate production progress.
[0068] In practical applications, "generating a customer communication report" refers to automatically summarizing the current order's risk status, expected impact, and countermeasures already taken or recommended, forming a structured report. This report may include information such as risk level, expected new delivery window, and recommendations for customers, enabling timely and transparent communication with customers, managing customer expectations, and maintaining customer relationships.
[0069] This solution utilizes specific strategies such as suspending downstream resources, adjusting process parameters, and generating customer communication reports to enable enterprises to respond quickly and effectively to potential delivery crises, avoiding losses caused by information delays or untimely decision-making.
[0070] Secondly, such as Figure 2 As shown, an intelligent supply chain optimization scheduling system 100 for carbon fiber materials is provided. This system is used to optimize the scheduling of carbon fiber production orders executed on the carbon fiber production line. The system includes: The process data acquisition module 101 is used to acquire process data of carbon fiber materials during the production process. The process data includes at least one of the following: carbonization furnace temperature zone temperature, protective gas flow rate, and carbon fiber tension. Performance evaluation module 102 is used to evaluate the performance of carbon fiber products based on the obtained process data and obtain performance evaluation results; The order delivery information acquisition module 103 is used to acquire the order delivery information of the current carbon fiber production order, wherein the order delivery information includes the promised delivery date, production progress and available alternative production line information. Delivery confidence index generation module 104 is used to generate delivery confidence index by combining performance evaluation results and order delivery information. Risk identification module 105, which is used to determine whether there is any risk to order delivery based on delivery confidence indicators; The strategy output module 106 is used to respond to order delivery risks and output response strategies.
[0071] This application discloses an intelligent supply chain optimization scheduling method and system for carbon fiber materials. It acquires process data during the carbon fiber production process and evaluates the performance of carbon fiber products based on this data to obtain performance evaluation results. Simultaneously, it acquires order delivery information for current carbon fiber production orders and, combined with the performance evaluation results and order delivery information, generates a delivery confidence index. This index is used to determine whether there are any order delivery risks. Finally, it outputs response strategies to address these risks. This method improves the ability to proactively respond to potential quality risks and delivery pressures during the production process when handling high-priority carbon fiber production orders. Through real-time process data monitoring and performance evaluation, potential quality risks can be identified early and combined with order delivery information to form a quantified delivery confidence index. This avoids resource waste and delivery risks caused by scheduling downstream high-value resources according to the established plan even when product quality risks exist.
[0072] Finally, it should be noted that the above-described embodiments are merely specific implementations of this disclosure, used to illustrate the technical solutions of this disclosure, and not to limit them. The protection scope of this disclosure is not limited thereto. Although this disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features within the scope of the technology disclosed in this disclosure. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this disclosure, and should all be covered within the protection scope of this disclosure.
Claims
1. A smart supply chain optimization scheduling method for carbon fiber materials, wherein the method is used to optimize the scheduling of carbon fiber production orders executed on the carbon fiber production line, characterized in that, The method includes: Obtain process data of carbon fiber materials during the production process, the process data including at least one of carbonization furnace temperature zone temperature, protective gas flow rate and carbon fiber tension; The performance of carbon fiber products is evaluated based on the obtained process data, and the performance evaluation results are obtained. Obtain the order delivery information for the current carbon fiber production order, wherein the order delivery information includes the promised delivery date, production schedule, and information on available alternative production lines; By combining performance evaluation results with order delivery information, a delivery confidence index is generated. Based on the delivery confidence index, determine whether there is any risk to order delivery; Respond to order delivery risks and develop corresponding strategies.
2. The intelligent supply chain optimization scheduling method for carbon fiber materials according to claim 1, characterized in that, The acquisition of process data for carbon fiber materials during production includes: Detection sensors deployed on the carbon fiber production line are used to collect continuous process data at a set frequency. Interpolation is used to process continuous process data to obtain processed process data.
3. The intelligent supply chain optimization scheduling method for carbon fiber materials according to claim 1, characterized in that, The acquisition of process data for carbon fiber materials during production includes: Using sensor arrays deployed on the carbon fiber material production line, the output data of the sensing elements in each sensor array is continuously monitored, and the output data is compared in real time to obtain the comparison results. When the comparison result indicates that the output data of a sensor element deviates from the output data of the remaining sensor elements belonging to the same sensor array, the sensor element is identified as a failed element, and the identification result is output. Based on the identification results, the output data of the sensing element is corrected or excluded, and the output data of the remaining sensing elements under the same sensor array is used as the process data of carbon fiber material in the production process.
4. The intelligent supply chain optimization scheduling method for carbon fiber materials according to claim 1, characterized in that, The process of evaluating the performance of carbon fiber products based on the obtained process data to obtain performance evaluation results includes: The process data is input into a preset carbon fiber production evolution module, wherein the preset carbon fiber production evolution module is used to simulate the carbon fiber production structure evolution process and predict the structural mechanical properties of the produced carbon fiber based on the input process data. Based on the structural mechanical properties of the produced carbon fibers, a quantitative carbon fiber production quality qualification index is output and used as the performance evaluation result.
5. The intelligent supply chain optimization scheduling method for carbon fiber materials according to claim 4, characterized in that, The process of quantifying and outputting a carbon fiber production quality qualification index based on the structural mechanical properties of the produced carbon fiber, and using this index as a performance evaluation result, includes: Energy excitation is applied to the precursor material entering the carbon fiber production line, and response data about the precursor material is collected. The response data includes the instantaneous temperature change value, surface potential shift, infrared absorption spectrum intensity change, and trace gas release concentration of the precursor material. The response data is converted into quantitative characteristics of the precursor material's response under energy excitation. These quantitative characteristics include instantaneous temperature change rate, peak and duration of surface potential shift, shift of characteristic peaks in infrared absorption spectrum, peak and decay rate of trace gas release concentration. The quantitative characteristics are compared with preset precursor material behavior patterns to identify whether there are any unidentified intrinsic differences in the current precursor material. When an unidentified intrinsic difference is identified in the current precursor material, a correction coefficient is determined based on the degree of deviation between the response quantification characteristics corresponding to the intrinsic difference and the preset precursor material behavior pattern, combined with the production practice data of different types of precursor materials. The carbon fiber production quality qualification index is then corrected based on this correction coefficient.
6. The intelligent supply chain optimization scheduling method for carbon fiber materials according to claim 4, characterized in that, After quantifying and outputting the carbon fiber production quality qualification index based on the structural mechanical properties of the produced carbon fiber, and using it as the performance evaluation result, the following is also included: Energy excitation is applied to the precursor material entering the carbon fiber production line, and response data of the precursor material is collected. The response data specifically includes the instantaneous temperature change value, surface potential shift, infrared absorption spectrum intensity change, and trace gas release concentration of the precursor material. A correlation model between Raman features and ultrasonic features is pre-established. This correlation model is used to realize the mutual conversion between Raman feature vectors and ultrasonic feature vectors. The Raman feature vector is obtained by inputting the response data of the precursor material into a Raman spectrometer. Based on the obtained Raman feature vector, the predicted ultrasonic feature vector is obtained by using the correlation model between Raman features and ultrasonic features. The actual ultrasonic feature vector is obtained by inputting the response data of the precursor material into an ultra-high frequency acoustic attenuation sensor array. Determine the consistency deviation between the actual ultrasonic feature vector and the predicted ultrasonic feature vector. When the consistency deviation exceeds a preset threshold, it is confirmed that the acquired precursor material response data is subject to external interference. Based on the consistency deviation, the corresponding correction coefficient is found from the preset correction factor list, and the carbon fiber production quality qualification index is corrected according to the correction coefficient.
7. The intelligent supply chain optimization scheduling method for carbon fiber materials according to claim 4, characterized in that, The process of generating delivery confidence metrics by combining performance evaluation results and order delivery information includes: The order time margin percentage is calculated based on the delivery date and production schedule, and the available alternative path index is calculated based on the information on alternative production lines. The delivery confidence index is calculated by combining the carbon fiber production quality qualification index, order time margin percentage, and available alternative path index using a weighted summation algorithm. The delivery confidence index is obtained according to the following calculation formula: , Preset weighting coefficients and satisfying + + =1; A indicates the carbon fiber production quality qualification index, which is a value between 0 and 1. B indicates the order time margin percentage, which is the ratio of the buffer time between the current remaining production cycle and the promised delivery time, and is a value between 0 and 1. C indicates the available alternative path index, which is a value between 0 and 1.
8. The intelligent supply chain optimization scheduling method for carbon fiber materials according to claim 1, characterized in that, The determination of whether there is order delivery risk based on delivery confidence indicators includes: Preset delivery confidence risk threshold; When the delivery confidence index falls below the preset delivery confidence risk threshold, an order delivery risk is confirmed.
9. The intelligent supply chain optimization scheduling method for carbon fiber materials according to claim 1, characterized in that, The response strategies include: The initiation of downstream resources for carbon fiber production orders may be suspended or delayed. These downstream resources include at least one of special resin formulation and special temperature-controlled logistics vehicle reservation. Provide suggestions for adjusting process parameters during the production of carbon fiber materials; And generate customer communication reports.
10. A smart supply chain optimization scheduling system for carbon fiber materials, the system being used to optimize the scheduling of carbon fiber production orders executed on a carbon fiber production line, characterized in that, The system includes: A process data acquisition module is used to acquire process data of carbon fiber materials during the production process. The process data includes at least one of the following: carbonization furnace temperature zone temperature, protective gas flow rate, and carbon fiber tension. A performance evaluation module is used to evaluate the performance of carbon fiber products based on the obtained process data and obtain performance evaluation results. The order delivery information acquisition module is used to acquire the order delivery information of the current carbon fiber production order, wherein the order delivery information includes the promised delivery date, production progress and available alternative production line information; The delivery confidence metric generation module is used to generate delivery confidence metrics by combining performance evaluation results and order delivery information. A risk identification module is used to determine whether there is any risk to order delivery based on a delivery confidence index. The strategy output module is used to respond to order delivery risks and output corresponding strategies.