Method for predicting properties of an engineering plastic

Abstract

The application discloses an engineering plastic performance prediction method and relates to the field of data processing.The method comprises the following steps: reading basic elements of a pretest engineering plastic; combining an intelligent inference model to determine a first performance prediction result; interacting with multiple application scenarios of the pretest engineering plastic to determine a second performance prediction result; based on a controllable deviation range, mutually testing the first performance prediction result and the second performance prediction result to determine an effective test result; if the effective test result is yes, performing mean value processing to serve as a performance prediction result.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and more particularly to the field of performance monitoring data processing, specifically a method for predicting the performance of engineering plastics. Background Technology

[0002] Materials science provides the fundamental theories and methods for predicting the performance of engineering plastics. The performance of engineering plastics depends on various factors, including their chemical composition, molecular structure, crystallinity, and orientation. Materials scientists have developed various prediction models, such as molecular dynamics simulations and finite element analysis, to predict the performance of engineering plastics under different conditions by deeply studying the relationship between these factors and performance. Performance predictions are often conducted under specific conditions, but in practical applications, engineering plastics face a variety of complex environmental and stress conditions. If the prediction model does not fully consider these actual usage conditions, the predicted results will differ significantly from the actual performance. Current prediction techniques may have limitations, such as insufficient computational power or inappropriate algorithm selection, which may lead to inaccurate or unstable prediction results.

[0003] In summary, the existing technology has a technical problem: the performance prediction of engineering plastics is carried out under specific conditions, which leads to insufficient consideration of actual conditions and a large error in the prediction results. Summary of the Invention

[0004] Therefore, it is necessary to provide a method for predicting the performance of engineering plastics that can perform performance prediction through finite element reasoning and multi-scenario joint simulation testing, and verify the prediction results to obtain more accurate prediction results.

[0005] In view of the above problems, this application provides a method for predicting the properties of engineering plastics.

[0006] This application discloses a method for predicting the performance of engineering plastics. The method includes: reading the basic elements of the engineering plastic to be tested, the basic elements including raw material ratio elements and processing elements; combining an intelligent derivation model, performing single-element derivation prediction and hierarchical collaborative analysis based on the basic elements to determine a first performance prediction result, wherein the intelligent derivation model is trained based on the finite element analysis principle, and the performance dimensions include at least mechanical properties, thermal properties, electrical properties, and optical properties; interacting with multiple application scenarios of the engineering plastic to be tested, performing scenario switching tests and scenario performance joint analysis based on a visualization simulation module to determine a second performance prediction result; based on a controllable deviation range, cross-checking the first performance prediction result and the second performance prediction result to determine a valid check result; if the valid check result is positive, averaging the first performance prediction result and the second performance prediction result to obtain the performance prediction result of the engineering plastic to be tested.

[0007] The above-mentioned method for predicting the performance of engineering plastics solves the technical problem that the performance prediction of engineering plastics in the prior art is carried out under specific conditions, which leads to a large error in the prediction results due to insufficient consideration of actual conditions. The method uses finite element reasoning and multi-scenario joint simulation test to perform performance prediction and cross-verify the prediction results to obtain more accurate prediction results.

[0008] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0009] Figure 1 This is a flowchart illustrating a method for predicting the properties of engineering plastics in one embodiment; Figure 2 This is a flowchart illustrating the process of determining the effective verification effect of an engineering plastics performance prediction method in one embodiment. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0011] This application provides a method for predicting the performance of engineering plastics. This method solves the technical problem in the prior art where the performance prediction of engineering plastics is carried out under specific conditions, which leads to insufficient consideration of actual conditions and a large error in the prediction results.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0014] like Figure 1 As shown, this application provides a method for predicting the properties of engineering plastics, the method comprising: Read the basic elements of the pre-tested engineering plastics, including raw material proportions and processing elements; Engineering plastics performance prediction refers to the process of estimating and judging the physical, chemical, and mechanical properties of engineering plastics under specific conditions using a series of methods and technologies. Engineering plastics possess characteristics such as high strength, high mechanical modulus, wear resistance, fatigue resistance, chemical resistance, excellent electrical properties, flame retardancy, and weather resistance, making them suitable for use as engineering structural components and capable of long-term use in harsh chemical and physical environments. This application provides a method for predicting the performance of engineering plastics, estimating and judging the physical, chemical, and mechanical properties of engineering plastics under specific conditions. Performance prediction is based on finite element method inference and multi-scenario joint simulation testing, and the prediction results are cross-validated. If the difference between the two methods is within an acceptable range, the average value is taken as the final prediction result. By predicting the performance of engineering plastics, the cost of experimentation and trial and error is reduced, improving the efficiency and success rate of product development.

[0015] By reading the fundamental elements of engineering plastics for pre-testing, we can gain a comprehensive understanding of their composition and manufacturing process, thereby predicting their performance. These fundamental elements are key information closely related to the performance of engineering plastics. The raw material ratios include the types of raw materials, additive components, and proportions. The types of raw materials include different basic polymer types such as polyamide and polycarbonate. The additive components include stabilizers, pigments, and fillers. The raw material ratios include the proportion of polymers to additives. The manufacturing process refers to the various stages of the plastic manufacturing process, such as molding temperature, injection pressure and speed, mold temperature, and drying conditions. By reading these fundamental elements, we can gain a comprehensive understanding of the composition and manufacturing process of engineering plastics, and thus predict their performance. These fundamental elements are not only the basis for performance prediction but also an important reference for optimizing product design, improving manufacturing processes, and enhancing product quality.

[0016] Combining the intelligent derivation model, single-element derivation prediction and hierarchical collaborative analysis are performed based on the basic elements to determine the first performance prediction result. The intelligent derivation model is trained based on the finite element analysis principle, and the performance dimensions include at least mechanical performance, thermal performance, electrical performance and optical performance. A smart derivation model is constructed, establishing a complex mapping relationship between the fundamental elements and properties of engineering plastics. Based on the finite element analysis (FEM) principle, the model can simulate the stress, heat transfer, electrical conductivity, and optical behavior of plastics under different conditions, thereby predicting their performance. FEM is a numerical analysis method that approximates the behavior of a continuous system by dividing a continuous object into a finite number of small units. The core idea of ​​FEM is to discretize a continuous system into a finite number of interconnected units. This discretization approach is used to handle the complex performance prediction problem of engineering plastics when training the smart derivation model. By dividing the continuous raw material ratios and processing elements into a finite number of discrete regions, the model can better capture the nonlinear relationship between these elements and performance. During training, the model analyzes each discrete region separately to obtain the mapping relationship between raw material ratios, processing elements, and performance. During training, the smart derivation model comprehensively considers information from all discrete points or regions and continuously adjusts model parameters through optimization algorithms to improve the accuracy of performance prediction. Single-factor derivation and prediction are performed, considering each factor in the raw material ratio and processing parameters separately, and analyzing their impact on performance using the intelligent derivation model. For example, adjusting the dosage of a certain additive alone and observing its effect on mechanical or thermal properties; or changing the molding temperature and observing its effect on the crystallinity and properties of the plastic, etc. Through the above methods, the individual contribution of each factor to performance is obtained. Hierarchical synergistic analysis is then performed, considering the interaction and synergistic effects between multiple factors. The intelligent derivation model can analyze the mutual influence between different factors, as well as the synergistic effect of the basic elements in the hierarchical structure. For example, raw material ratio and processing conditions may jointly affect the crystallinity and mechanical properties of the plastic; or a certain additive and molding temperature may jointly affect the thermal properties of the plastic. Through hierarchical collaborative analysis, the complex relationships between various factors are obtained, leading to more accurate performance prediction results, denoted as the first performance prediction result. The first performance prediction result output by the intelligent derivation model includes mechanical properties, thermal properties, electrical properties, and optical properties. Mechanical properties include density, hardness, water absorption rate, and water vapor transmission rate; thermal properties include melting temperature, melt index, and heat distortion temperature; electrical properties include dielectric strength, breakdown voltage, and withstand voltage; and optical properties include light transmittance, gloss, and haze. By combining the principles of finite element analysis, the intelligent derivation model can better handle complex nonlinear relationships, improving the accuracy and reliability of performance prediction.

[0017] Determine the mapping relationship between the basic elements and the performance indicators of the pre-tested engineering plastics, and determine the element association network; Identify the element association network, predict and derive the mapping performance index of each basic element, and determine the performance of a single element; Based on the synergy of elements, the performance of the individual elements is weighted in one layer based on the importance of the elements and in a second layer based on the correlation of the indicators, and the characteristic value of the overall indicator is calculated. By integrating the characteristic values ​​of the assembly indicators, the first performance prediction result is determined.

[0018] By reviewing literature, experimental data, and expert experience, the mapping relationship between raw material proportioning factors, processing factors, and engineering plastic performance indicators is determined. Based on this mapping relationship, a factor association network is constructed. This network can intuitively display the connection relationship between each basic factor and performance indicator. In the factor association network, basic factors directly related to specific performance indicators are identified. Using the intelligent derivation model, the predicted performance indicators are derived for each identified basic factor; that is, the basic factor is used as input, and the corresponding predicted performance indicator value is calculated by the model. According to the specific needs of engineering plastic performance prediction and expert experience, the importance of each basic factor to the performance indicators is determined, and corresponding weights are assigned. This reflects the relative importance of different factors in affecting performance. Considering the correlation between performance indicators, the single-factor performance obtained through the first-level weighting is reweighted a second time to further adjust the weight of each performance indicator in the overall indicators, so as to reflect the interaction and influence between the basic factors. The single-factor performances after the first-level and second-level weighting are weighted and summed or other suitable combinations are used to calculate the characteristic values ​​of the overall indicators. These characteristic values ​​comprehensively reflect the overall performance of the pre-tested engineering plastics on multiple performance indicators. The calculated assembly index characteristic values ​​are integrated to form the first performance prediction result, which provides comprehensive and accurate information on the pre-tested engineering plastics in multiple key performance indicators, and provides an important reference for subsequent product design, process optimization and performance improvement.

[0019] Obtain homologous production records of the pre-tested engineering plastics, and identify performance influencing factors. These performance influencing factors are identified by influence coefficients and include external factors and internal processing elements. The first performance prediction result is compensated based on the aforementioned performance influencing factors.

[0020] Obtain the production records of the pre-tested engineering plastics from the same source. These records refer to the production records of other batches or models of engineering plastics produced under the same or similar production conditions as the pre-tested engineering plastics. This includes information such as raw material sources, production processes, production equipment, and environmental conditions. By collecting these records, we can understand the various variables and conditions during the production process of the engineering plastics. From these same-source production records, we identify factors affecting the performance of the engineering plastics, categorized into external factors and internal processing factors. External factors include batch differences in raw materials and changes in environmental temperature and humidity; internal processing factors involve minor adjustments to the production process and changes in equipment status. We determine the degree of influence on performance and assign influence coefficients to them. For each performance-influencing factor identified, quantification is performed, transforming it into calculable and analyzable values. For example, batch variations in raw materials can be quantified by the fluctuation range of chemical composition; changes in environmental temperature and humidity can be represented by specific temperature and humidity values. The influence coefficient, a crucial indicator reflecting the degree of influence of performance-influencing factors on the prediction results, is determined based on historical data, expert experience, or experimental results. Based on these coefficients, a compensation mechanism is constructed to correct the initial performance prediction results. This mechanism can adjust the prediction results according to different influencing factors and their coefficients. Applying this mechanism to the initial performance prediction results, compensation is made based on the quantified values ​​and influence coefficients of each performance-influencing factor. This method yields prediction results closer to actual performance, improving prediction accuracy. By acquiring source production records, identifying performance-influencing factors, and compensating for prediction results based on these factors, the engineering plastics performance prediction model can be further improved, making it more closely aligned with actual production conditions and providing more accurate and reliable guidance for performance optimization and process improvement of engineering plastics.

[0021] The interaction of the pre-tested engineering plastics with diverse application scenarios allows for scenario switching tests and joint scenario performance analysis to be performed in the visualization simulation module, thereby determining the second performance prediction result. The interactive pre-testing of engineering plastics involves diverse application scenarios, including high-temperature, low-temperature, high-humidity, and mechanical stress environments, depending on the application field. For each scenario, the environmental conditions and performance challenges faced by the engineering plastics are understood. These application scenarios are then imported into a visualization simulation module. This module is a software system capable of simulating real-world environmental conditions, simulating changes in environmental parameters under different application scenarios, and displaying the performance response of the engineering plastics in real time. Scenario switching tests are performed, simulating the switching process of the engineering plastics between different scenarios to observe continuous changes in performance and obtain the adaptability and stability of the engineering plastics under different environmental conditions. Joint scenario performance analysis involves comprehensively evaluating the performance indicators of the engineering plastics under different scenarios and analyzing their mutual influence and correlation. Through joint analysis, the overall performance of the engineering plastics in diverse application scenarios is obtained. Based on the results of the scenario switching test and the joint scenario performance analysis, a second performance prediction result is determined, reflecting the comprehensive performance of the engineering plastics in diverse application scenarios and providing a more comprehensive and accurate performance prediction for practical applications. By interactively pre-testing various application scenarios of engineering plastics and performing scenario switching tests and joint scenario performance analysis in the visualization simulation module, we can understand the performance of engineering plastics and provide more accurate and reliable prediction results for practical applications.

[0022] Perform scene switching tests and determine M performance test results, where M is the same as the number of the diverse application scenarios; Conduct industrial big data retrieval to determine the practicality of each application scenario, assign scenario weights, and determine the scenario weight distribution; Based on the scenario weight distribution, the M performance test results are weighted and summed according to the same performance index to determine the comprehensive performance index. The integrated performance indicators are used as the second performance prediction result.

[0023] For each of the multiple application scenarios of the pre-tested engineering plastics, scenario switching tests were conducted sequentially. This means that in a simulation environment, the process of switching the engineering plastic from one scenario to another was simulated, and its performance changes were recorded in real time, resulting in M ​​performance test results, where M is consistent with the number of multiple application scenarios. Each performance test result represents the performance of the engineering plastic in the corresponding scenario. Through the analysis of industrial big data, the actual usage of the engineering plastic in each application scenario was obtained, including usage frequency, importance, and stability requirements. Based on the practical situation of each application scenario, scenario weights were assigned, that is, a weight was assigned to each scenario, which reflects the importance and influence of the scenario in actual application. According to the scenario weight distribution, the M performance test results were weighted and summed for the same performance index. That is, for each performance index, the performance index values ​​under each scenario were weighted and summed according to the weight of its corresponding scenario. Through the above method, the degree of influence of different scenarios on the same performance index can be obtained, thereby obtaining a more accurate comprehensive evaluation result, namely the comprehensive performance index. The comprehensive performance index can comprehensively reflect the overall performance of the pre-tested engineering plastic in multiple application scenarios. By integrating all comprehensive performance indicators together, a second performance prediction result is obtained, which can more comprehensively and accurately reflect the performance of the pre-tested engineering plastics in various application scenarios.

[0024] Based on the controllable deviation range, the first performance prediction result and the second performance prediction result are cross-checked to determine the valid check result; A controllable deviation range is determined based on historical data of engineering plastic properties, expert experience, and practical application requirements. This controllable deviation range reflects the acceptable error range between the predicted results and actual performance. The performance indicators in the first performance prediction result are compared with the corresponding indicators in the second performance prediction result, including comparing the magnitude and trend of the same performance indicator in the two prediction results. The deviation value between the two prediction results for the same performance indicator is calculated. The deviation value can be calculated using absolute deviation, relative deviation, or other suitable deviation measurement methods. The calculated deviation value is compared with the controllable deviation range. If the deviation value is within the controllable range, the two prediction results are considered consistent for that performance indicator, and the prediction result is valid. If the deviation value exceeds the controllable range, further analysis of the cause is required. The performance indicators of the valid verification results are integrated as the final prediction result. The final prediction result ensures the accuracy and reliability of the prediction results through mutual verification.

[0025] like Figure 2 As shown, N distribution points of the pre-tested engineering plastic are set, and a spatial coordinate system is established, wherein the points are distributed at any position from the surface to the center; Establish a first performance distribution map and a second performance distribution map based on the first performance prediction result and the second performance prediction result. The performance distribution map uses the performance index as the coordinate axis, and each distribution map point corresponds to an index plural. Map the spatial coordinate system, the first performance distribution map, and the second performance distribution map, perform corresponding verification, and determine the valid verification result.

[0026] N distribution points are set on the pre-tested engineering plastic. These N distribution points are located at arbitrary positions from the surface to the center, covering different areas of the engineering plastic to more comprehensively reflect its performance distribution. Based on the set distribution points, a three-dimensional spatial coordinate system is established. Each distribution point has a clear coordinate position in the coordinate system, which facilitates subsequent data mapping and verification. The performance index data from the first and second performance prediction results are extracted. Each distribution point corresponds to a set of performance index data. Using the performance index as the coordinate axis, a first performance distribution map and a second performance distribution map are established. Each distribution point in the performance distribution map corresponds to a tuple of indicators, that is, the numerical combination of various performance indicators for that point. Each distribution point in the spatial coordinate system is mapped to a corresponding tuple of indicators in the performance distribution map. This allows for a visual representation of the performance indicator distribution at each point in the spatial coordinate system. The mapped results are then verified by comparing the performance indicator data for the same distribution point in the first and second performance distribution maps, analyzing the differences and consistency between them. If the performance indicator data for a given distribution point in the two performance distribution maps are essentially consistent and within a controllable deviation range, then the prediction result for that point is considered valid, and a valid verification result is determined. This valid verification result provides information about the performance distribution of the pre-tested engineering plastics in different locations and regions, helping to comprehensively understand its performance characteristics and providing guidance for practical applications.

[0027] Identify the first mapping relationship, perform performance difference calibration for points with the same distribution, and determine the first test result; Identify the second mapping relationship, verify the performance distribution trend, and determine the second verification result. Based on the first test result and the second test result, the valid test result is determined.

[0028] Identifying the mapping relationship between the spatial coordinate system and the first performance distribution map includes determining the specific location of each distribution point in three-dimensional space and the correspondence between the corresponding indicator tuples in the first performance distribution map. For each distribution point, the performance indicators in the first performance prediction result are compared with the corresponding indicators in the second performance prediction result. By calculating the degree of difference between performance indicators, the consistency of the two prediction results at the same distribution point can be evaluated, and performance difference correction at the same distribution point is performed. Based on the performance difference correction result, a first verification result is determined, identifying the mapping relationship between the spatial coordinate system and the overall performance distribution map. The first performance distribution map and the second performance distribution map are compared to observe whether the changing trends of performance indicators in different regions are consistent, that is, checking whether the performance indicators show similar increasing, decreasing, or fluctuating patterns, and whether the distribution of these patterns in space is consistent. Through the verification of performance distribution trends, a second verification result is determined. If the two performance distribution maps show a high degree of consistency in the overall trend, it indicates that the prediction result is reliable in spatial distribution. Based on the first verification result and the second verification result, the final valid verification result is determined. When both verification results indicate that the prediction result is valid, the prediction result is considered accurate and reliable. The above verification process ensures that the performance prediction results of the pre-tested engineering plastics are highly accurate and reliable, providing strong support for practical applications.

[0029] The first mapping relationship is the correspondence between the spatial coordinate system, the first performance distribution map, and the second performance distribution map; the second mapping relationship is the correspondence between the spatial coordinate system and the first performance distribution map, and the spatial coordinate system and the second performance distribution map; the performance distribution trend includes surface uniformity and depth distribution trend.

[0030] The first mapping relationship refers to the correspondence between the spatial coordinate system, the first performance distribution map, and the second performance distribution map. Each distribution point set in the spatial coordinate system should have corresponding performance index data represented in both the first and second performance distribution maps, ensuring that the performance index of a certain distribution point can be directly located in the spatial coordinate system, and the differences between the two performance distribution maps at that point can be compared. The second mapping relationship is the correspondence between the spatial coordinate system and the performance distribution map. The spatial coordinate system and the first performance distribution map ensure that each distribution point in the spatial coordinate system accurately matches the performance index data in the first performance distribution map, obtaining the first performance prediction of engineering plastics at different locations; the correspondence between the spatial coordinate system and the second performance distribution map ensures that the distribution points in the spatial coordinate system accurately correspond to the performance index data in the second performance distribution map, obtaining the performance of the same distribution point in the second performance prediction. Performance distribution trends include surface uniformity and depth distribution trends. Surface uniformity refers to whether the performance indicators at various points on the surface of the engineering plastic exhibit a consistent distribution pattern. If the performance indicators at different points on the surface differ too much, it means that the prediction model's predictions in the surface area are not accurate enough. Depth distribution trends refer to how the performance indicators change with depth from the surface to the center. If the two performance distribution maps show inconsistent trends in the depth direction—for example, one showing a gradual increase in performance while the other shows a gradual decrease—it indicates a deviation in the prediction results. After identifying the first mapping relationship, the second mapping relationship, and verifying the performance distribution trends, valid verification results are determined. By verifying the first mapping relationship, the performance indicators at the same distribution point are compared in the two performance distribution maps. If the difference is within a controllable range, the prediction result for that point is initially considered valid. By verifying the second mapping relationship, the correspondence between the spatial coordinate system and the performance distribution map is ensured. By verifying the performance distribution trends, the consistency and reliability of the prediction results in spatial distribution are confirmed. If both surface uniformity and depth distribution trends show high consistency in the two performance distribution maps, the prediction results are considered valid. By combining the verification results of the first mapping relationship, the second mapping relationship, and the performance distribution trend, valid test results are determined, thereby ensuring the accuracy and reliability of the performance prediction results of the pre-tested engineering plastics.

[0031] Identify the scene weight distribution and determine the optimal collision avoidance principle; Determine whether the performance prediction result meets the expected performance standard. If not, generate optimization instructions for the poor performance. Upon receiving the optimization instruction, and constrained by the optimization avoidance principle, multi-objective optimization is performed on the basic elements associated with the poor performance to determine the optimization scheme, wherein the optimization iteration and the poor rollback are used as optimization criteria. Based on the optimization scheme, feedback production management is carried out for the pre-tested engineering plastics.

[0032] Based on the actual application scenarios of the pre-tested engineering plastics, the weight distribution of different scenarios is identified, and the optimization avoidance principle for scenario collisions is determined. This optimization avoidance principle clarifies which performance indicators should be prioritized during the optimization process, which can be appropriately compromised, and avoids potential performance conflicts or contradictions. For example, when optimizing a certain indicator that does not meet the standard, there may be a scenario collision in one adjustment direction, such as a superior performance indicator versus a inferior one in another scenario; a decision is made on whether to execute this step. The performance prediction results are evaluated to determine whether they meet the expected performance standards. The expected performance indicators are set and obtained by the staff. The prediction results are compared with the set performance thresholds. If the prediction results do not meet the expected performance standards, optimization instructions are generated for the inferior performance. The inferior performance refers to the performance characteristics of the engineering plastics that do not meet the expected or design standards during the prediction process. The optimization instructions should include specific optimization objectives, optimization methods, and optimization steps to ensure the smooth progress of subsequent optimization work. Upon receiving optimization instructions, and constrained by the principle of optimization avoidance, multi-objective optimization is performed on the fundamental elements related to suboptimal performance. This requires a strategy combining optima-oriented iteration and suboptimal backoff. Optima-oriented iteration means continuously trying to improve performance indicators and find better solutions; suboptimal backoff, when encountering situations where further performance improvement is not possible, appropriately lowers the requirements of certain indicators in exchange for improvements in other indicators. While ensuring overall performance meets requirements, the optimal balance point is found. Based on the optimization solution, feedback is used in production management, translating the optimization solution into specific production process parameters, operating procedures, or equipment adjustments, and applying them to the actual production process. Through these measures, the performance of pre-tested engineering plastics is improved, thereby increasing product quality and production efficiency.

[0033] If the valid test result is yes, the average of the first performance prediction result and the second performance prediction result is used as the performance prediction result of the pre-tested engineering plastic.

[0034] If the valid verification result is yes, it means that the first performance prediction result and the second performance prediction result are consistent in terms of key performance indicators and meet the preset verification standards. The first and second performance prediction results can then be averaged. Averaging is a commonly used data fusion method that can effectively integrate data from different sources, reduce the impact of random and systematic errors, and thus obtain more stable and reliable results. The performance indicators corresponding to the first and second performance prediction results are compared one by one, and the average value is calculated. The average values ​​are then integrated to obtain the final performance prediction result for the pre-tested engineering plastic. Through averaging, using the information from the two prediction results, a more comprehensive and accurate performance prediction result is obtained, which helps to understand the performance of the pre-tested engineering plastic and provides strong support for subsequent production and application. When averaging, it should be ensured that the first and second performance prediction results are compared under the same coordinate system and the same performance indicator system to ensure the accuracy and comparability of the calculation. Performance prediction is performed based on finite element reasoning and multi-scenario joint simulation testing, and the prediction results are mutually verified. If the difference between the two is within the allowable range, the average value is taken as the final prediction result for plastic performance prediction.

[0035] In summary, the beneficial effects of this method include: The technical effect of using finite element reasoning and multi-scenario joint simulation testing to perform performance prediction and cross-validate the prediction results is to obtain more accurate prediction results.

[0036] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific and sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0037] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0038] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for predicting the properties of engineering plastics, characterized in that, The method includes: Read the basic elements of the pre-tested engineering plastics, including raw material proportions and processing elements; Combining the intelligent derivation model, single-element derivation prediction and hierarchical collaborative analysis are performed based on the basic elements to determine the first performance prediction result. The intelligent derivation model is trained based on the finite element analysis principle, and the performance dimensions include at least mechanical performance, thermal performance, electrical performance and optical performance. The interaction of the pre-tested engineering plastics with diverse application scenarios allows for scenario switching tests and joint scenario performance analysis to be performed in the visualization simulation module, thereby determining the second performance prediction result. Based on the controllable deviation range, the first performance prediction result and the second performance prediction result are cross-checked to determine the valid check result, wherein the controllable deviation range reflects the acceptable error range between the prediction result and the actual performance. If the valid test result is yes, the average of the first performance prediction result and the second performance prediction result is used as the performance prediction result of the pre-tested engineering plastic. The method, which performs single-element derivation prediction and hierarchical collaborative analysis based on the aforementioned basic elements, further includes: Determine the mapping relationship between the basic elements and the performance indicators of the pre-tested engineering plastics, and determine the element association network; Identify the element association network, predict and derive the mapping performance index of each basic element, and determine the performance of a single element; Based on the synergy of elements, the performance of the individual elements is weighted in one layer based on the importance of the elements and in a second layer based on the correlation of the indicators, and the characteristic value of the overall indicator is calculated. By integrating the characteristic values ​​of the assembly indicators, the first performance prediction result is determined.

2. The method as described in claim 1, characterized in that, After determining the first performance prediction result, the method further includes: Obtain homologous production records of the pre-tested engineering plastics, and identify performance influencing factors. These performance influencing factors are identified by influence coefficients and include external factors and internal processing elements. The first performance prediction result is compensated based on the aforementioned performance influencing factors.

3. The method as described in claim 1, characterized in that, The method further includes: interactively performing scenario switching tests and joint scenario performance analysis on the pre-tested engineering plastics across various application scenarios, and determining the second performance prediction result by executing scenario switching tests and joint scenario performance analysis in the visualization simulation module. Perform scene switching tests and determine M performance test results, where M is the same as the number of the diverse application scenarios; Conduct industrial big data retrieval to determine the practicality of each application scenario, assign scenario weights, and determine the scenario weight distribution; Based on the scenario weight distribution, the M performance test results are weighted and summed according to the same performance index to determine the comprehensive performance index. The integrated performance indicators are used as the second performance prediction result.

4. The method as described in claim 1, characterized in that, The method for cross-validating the first performance prediction result and the second performance prediction result further includes: N distribution points are set for the pre-tested engineering plastic, and a spatial coordinate system is established, wherein the points are distributed at any position from the surface to the center; Establish a first performance distribution map and a second performance distribution map based on the first performance prediction result and the second performance prediction result. The performance distribution map uses the performance index as the coordinate axis, and each distribution map point corresponds to an index plural. Map the spatial coordinate system, the first performance distribution map, and the second performance distribution map, perform corresponding verification, and determine the valid verification result.

5. The method as described in claim 4, characterized in that, The method for performing corresponding verification further includes: Identify the first mapping relationship, perform performance difference calibration for points with the same distribution, and determine the first test result; Identify the second mapping relationship, verify the performance distribution trend, and determine the second verification result. Based on the first test result and the second test result, the valid test result is determined.

6. The method as described in claim 5, characterized in that, The first mapping relationship is the correspondence between the spatial coordinate system, the first performance distribution map, and the second performance distribution map; the second mapping relationship is the correspondence between the spatial coordinate system and the first performance distribution map, and between the spatial coordinate system and the second performance distribution map. The performance distribution trends include surface uniformity and depth distribution trends.

7. The method as described in claim 3, characterized in that, After obtaining the performance prediction results, the method further includes: Identify the scene weight distribution and determine the optimization avoidance principle for scene collisions. The optimization avoidance principle refers to clarifying the performance indicators that need to be prioritized and compromised during the optimization process to avoid possible performance conflicts or contradictions. Determine whether the performance prediction result meets the expected performance standard. If not, generate optimization instructions for the poor performance. Upon receiving the optimization instruction, and constrained by the optimization avoidance principle, multi-objective optimization is performed on the basic elements associated with the poor performance to determine the optimization scheme, wherein the optimization iteration and the poor rollback are used as optimization criteria. Based on the optimization scheme, feedback production management is carried out for the pre-tested engineering plastics.

Images