Effect analysis method and system for communication mold components based on deep learning

Abstract

The present invention relates to the field of intelligent analysis, and proposes a method and system for analyzing the effect of communication mold components based on deep learning. The method includes: obtaining the production and manufacturing data of the communication mold components, extracting key characteristic parameters, analyzing the component effects, identifying core factors, detecting their mutual connection and calculating the correlation degree value, based on which the leading factors of the effect are screened, simulating the actual working scene to obtain performance data, analyzing the data change trend to determine the mold performance area, performing defect location to construct a performance defect curve, analyzing the influence of the curve to obtain defect influencing factors, formulating mold improvement measures based on this, simulating and implementing the simulation effect parameters, and finally generating an effect analysis report based on the parameters. The present invention can improve the accuracy of the effect analysis of the communication mold components.

Description

Technical Field

[0001] The present invention relates to the field of intelligent analysis, and in particular to a method and system for analyzing the effects of communication mold components based on deep learning. Background Art

[0002] In today's era of rapid technological development, the continuous innovation of communication technology has put forward higher requirements on the performance and quality of mold components. Moreover, with the popularization of 5G networks and the development of 6G technology, the accuracy, stability and reliability of communication mold components have become key factors in ensuring the efficient operation of communication equipment. However, in current production and application, the analysis of the effects of communication mold components still faces many challenges.

[0003] At present, the traditional communication mold component effect analysis method mainly relies on manual inspection and simple instrument measurement methods, such as dimension measurement, appearance inspection and other methods. This method can often only obtain surface and local information, and it is difficult to comprehensively and deeply evaluate the performance of the mold component. It is easy to miss detection and misjudgment, resulting in the inability to guarantee the accuracy of the communication mold component effect. Therefore, a communication mold component effect analysis method based on deep learning is needed to improve the accuracy of communication mold component effect analysis. Summary of the invention

[0004] The present invention provides a method and system for analyzing the effects of communication mold components based on deep learning, the main purpose of which is to improve the accuracy of the analysis of the effects of communication mold components.

[0005] Acquire manufacturing data of the communication mold component, the manufacturing data including: material properties, processing parameters and component size specifications, and extract key characteristic parameters from the manufacturing data;

[0006] Based on the key characteristic parameters, analyzing the component effects corresponding to the communication mold components, identifying the core factors corresponding to the component effects, detecting the mutual connections between the core factors, and calculating the correlation degree values ​​corresponding to the mutual connections;

[0007] Based on the correlation degree value, the effect dominant factor corresponding to the communication mold component is screened; based on the effect dominant factor, the actual working scene corresponding to the communication mold component is simulated, the performance data in the actual working scene is obtained, and the data change trend corresponding to the performance data is analyzed;

[0008] Based on the data change trend, determine the mold performance area corresponding to the communication mold component, locate the defects in the mold performance area, obtain the performance defect points, and construct the performance defect curve corresponding to the communication mold component based on the performance defect points;

[0009] An impact analysis is performed on the performance defect curve to obtain defect influencing factors, and based on the defect influencing factors, mold improvement measures corresponding to the communication mold assembly are formulated, and the mold improvement measures are simulated and implemented to obtain simulation effect parameters, and based on the simulation effect parameters, an effect analysis report corresponding to the communication mold assembly is generated.

[0010] Optionally, extracting key characteristic parameters from the manufacturing data includes:

[0011] Performing data cleaning on the production and manufacturing data to obtain cleaned manufacturing data;

[0012] Extracting detailed features from the cleaning manufacturing data; performing preliminary feature screening on the detailed features to obtain screening features;

[0013] Calculating feature similarity corresponding to the screening feature;

[0014] Based on the feature similarity, determining key features corresponding to the production and manufacturing data;

[0015] Extract key feature parameters corresponding to the key features.

[0016] Optionally, calculating the feature similarity corresponding to the screening feature includes:

[0017] The feature similarity corresponding to the screening feature is calculated using the following formula:

[0018]

[0019] in, represents the feature similarity corresponding to the screening feature, represents the feature index corresponding to the screening feature, Indicates the number of features corresponding to the screening feature, Indicates The weight coefficient of the feature, Represents the feature vector in the screening feature No. elements, Represents the feature vector The mean of Represents the feature vector in the screening feature No. elements, Represents the feature vector The mean of .

[0020] Optionally, analyzing the component effect corresponding to the communication mold component based on the key characteristic parameter includes:

[0021] Performing data integration on the key characteristic parameters to obtain integrated data;

[0022] Performing data visualization on the integrated data to obtain a visual scatter plot;

[0023] Performing correlation analysis on the visual scatter plot to obtain a correlation coefficient matrix;

[0024] Based on the correlation coefficient matrix, determining the component factors corresponding to the communication mold;

[0025] Based on the component factors, component effects corresponding to the communication mold components are analyzed.

[0026] Optionally, the calculating the association degree value corresponding to the mutual connection degree includes:

[0027] The correlation degree value corresponding to the mutual connection degree is calculated using the following formula:

[0028]

[0029] in, represents the correlation degree value corresponding to the mutual connection degree, represents the factor index of the key core factor pair in the mutual connection degree, represents the number of key core factor pairs in the mutual connection degree, Indicates The interrelationship values ​​between core factors are Indicates The importance weight of the core factors, represents the factor index of the key influencing factor in the mutual connection degree, represents the number of key influencing factors in the mutual connection degree, Indicates The actual weight of the influencing factors.

[0030] Optionally, the screening of the effect dominant factors corresponding to the communication mold assembly based on the correlation degree value includes:

[0031] Standardizing the correlation degree value to obtain a standard correlation value;

[0032] Based on the standard association value, setting an association threshold value corresponding to the communication mold component;

[0033] Based on the association threshold, factor verification is performed on the dominant factor in the communication mold component to obtain a verified dominant factor;

[0034] Analyze the effect factors corresponding to the verification leading factors;

[0035] Based on the effect factors, the effect dominant factors corresponding to the communication mold components are screened.

[0036] Optionally, simulating an actual working scenario corresponding to the communication mold assembly based on the effect-dominant factor includes:

[0037] Identify the factor range corresponding to the dominant factor of the effect;

[0038] Based on the range of factors, analyzing the working scenario corresponding to the communication mold assembly;

[0039] Querying a scene parameter group in the working scene, and monitoring scene working indicators in the scene parameter group;

[0040] Based on the scenario work indicators, the actual work scenario corresponding to the communication mold assembly is simulated.

[0041] Optionally, determining the mold performance area corresponding to the communication mold component based on the data change trend includes:

[0042] Identify the performance change law corresponding to the data change trend;

[0043] Querying a performance change sequence corresponding to the performance change rule;

[0044] Based on the performance change sequence, determining a performance boundary corresponding to the communication mold component;

[0045] Based on the performance boundaries, mold performance data in the communication mold assembly is divided into mold performance areas.

[0046] Optionally, constructing a performance defect curve corresponding to the communication mold assembly based on the performance defect point includes:

[0047] Identify defect point data corresponding to the performance defect point, and analyze data frequency values ​​corresponding to the defect point data;

[0048] Querying the defect mode corresponding to the data frequency value;

[0049] Based on the defect mode, determining a defect distribution area corresponding to the communication mold component;

[0050] Performing defect density analysis on the defect distribution area to obtain a defect density distribution map;

[0051] Based on the defect density distribution diagram, a performance defect curve corresponding to the communication mold assembly is constructed.

[0052] In order to solve the above problems, the present invention also provides an effect analysis system for implementing a communication mold assembly based on deep learning, the system comprising:

[0053] A parameter identification module is used to obtain the manufacturing data of the communication mold component, the manufacturing data including: material properties, processing parameters and component size specifications, and extract key characteristic parameters from the manufacturing data;

[0054] A correlation degree value calculation module is used to analyze the component effect corresponding to the communication mold component based on the key characteristic parameters, identify the core factors corresponding to the component effect, detect the mutual connection between the core factors, and calculate the correlation degree value corresponding to the mutual connection;

[0055] A trend analysis module, for screening the effect dominant factors corresponding to the communication mold assembly based on the correlation degree value, simulating the actual working scene corresponding to the communication mold assembly based on the effect dominant factors, obtaining the performance data in the actual working scene, and analyzing the data change trend corresponding to the performance data;

[0056] A curve construction module, used to determine the mold performance area corresponding to the communication mold component based on the data change trend, locate the defects in the mold performance area, obtain performance defect points, and construct a performance defect curve corresponding to the communication mold component based on the performance defect points;

[0057] A report generation module is used to perform an impact analysis on the performance defect curve to obtain defect influencing factors, formulate mold improvement measures corresponding to the communication mold assembly based on the defect influencing factors, simulate the implementation of the mold improvement measures to obtain simulation effect parameters, and generate an effect analysis report corresponding to the communication mold assembly based on the simulation effect parameters.

[0058] First of all, by acquiring the production and manufacturing data of the communication mold assembly, the present invention can fully understand the details of the mold assembly in the production process, which helps to accurately grasp the key factors such as material properties, processing parameters and component size specifications, and provide a solid foundation for subsequent effect analysis and performance evaluation. At the same time, by analyzing the component effect corresponding to the communication mold assembly based on the key characteristic parameters, the present invention can help us deeply understand the direct relationship between the performance of the component and the key characteristic parameters, and can accurately evaluate the degree of functional realization of the component under specific working conditions, thereby improving production efficiency. The present invention screens the effect dominant factors corresponding to the communication mold assembly based on the correlation degree value, which can help accurately identify the effects on the performance of the communication mold assembly. The factors that have the most critical impact on the effect can be adjusted and optimized by taking corresponding measures in the early stage of production, thereby reducing the defective rate and the subsequent maintenance costs. The present invention determines the mold performance area corresponding to the communication mold assembly based on the data change trend, and can accurately locate the advantages and disadvantages of the mold in terms of performance, which helps to optimize and improve in a targeted manner, concentrate resources on key areas, and improve the efficiency and effect of subsequent mold improvements. Furthermore, the present invention analyzes the performance defect curve to obtain the defect influencing factors, and can fundamentally understand the causes of performance defects of communication mold assemblies. It can improve material selection, processing accuracy, assembly methods and other links in a targeted manner, thereby improving the overall quality and reliability of the mold. Therefore, the method and system for implementing the effect analysis of communication mold assemblies based on deep learning proposed in the present invention can improve the accuracy of the effect analysis of communication mold assemblies. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A schematic flow chart of a method for analyzing the effect of a communication mold assembly based on deep learning provided by an embodiment of the present invention;

[0060] Figure 2 A schematic diagram of a module for implementing an effect analysis system for a communication mold assembly based on deep learning provided in one embodiment of the present invention.

[0061] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0062] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0063] The embodiment of the present application provides a method for analyzing the effects of a communication mold component based on deep learning. The execution subject of the method for analyzing the effects of a communication mold component based on deep learning includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the method for analyzing the effects of a communication mold component based on deep learning can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0064] Reference Figure 1 FIG. 1 is a flow chart of a method for analyzing the effect of a communication mold component based on deep learning according to an embodiment of the present invention. In this embodiment, the method for analyzing the effect of a communication mold component based on deep learning includes:

[0065] S1. Acquire manufacturing data of a communication mold component, wherein the manufacturing data includes material properties, processing parameters, and component size specifications, and extract key characteristic parameters from the manufacturing data.

[0066] By acquiring the production and manufacturing data of the communication mold assembly, the present invention can fully understand the details of the mold assembly during the production process, which helps to accurately grasp key factors such as material properties, processing parameters, and component size specifications, providing a solid foundation for subsequent effect analysis and performance evaluation.

[0067] Among them, the communication mold assembly refers to a mold used to manufacture key components in communication equipment, which may include antennas, connectors, circuit boards, etc. For example, it may refer to a mold assembly of a specific structure used in 5G base station communication equipment, or a mold component that performs a specific function in a satellite communication system; the production and manufacturing data refers to all relevant information collected during the production of these communication mold assemblies, and the production and manufacturing data includes: material properties, processing parameters and component size specifications. In detail, the material properties refer to the properties of the materials that constitute the communication mold assembly, such as the physical and chemical properties of the material such as hardness, strength, toughness, conductivity, thermal conductivity, etc.; the processing parameters refer to the specific set values ​​and operating conditions of various processes used in the production and manufacturing process, such as cutting speed, feed rate, processing temperature, pressure, processing time, etc. The component size specifications clearly define the specific geometric dimensions of the communication mold assembly such as length, width, height, thickness, aperture, and tolerance range. For example, the length of a communication mold assembly is required to be accurate to ±0.1 Within the tolerance range of millimeters, optionally, the acquisition of the production and manufacturing data of the communication mold assembly can be achieved by directly reading the device controller, such as the controller of a CNC machine tool, an injection molding machine, and other equipment.

[0068] Furthermore, by extracting key characteristic parameters from the production and manufacturing data, the present invention can quickly understand which material properties play a decisive role in the performance of the communication mold components, such as hardness, toughness or conductivity, thereby clarifying the degree of influence of key factors such as specific processing temperature, pressure and time on the quality of the final product.

[0069] The key feature parameters refer to specific numerical values ​​or parameter values ​​that can accurately quantify and describe the key features.

[0070] As an embodiment of the present invention, the extraction of key feature parameters in the production and manufacturing data includes: performing data cleaning on the production and manufacturing data to obtain cleaned manufacturing data; extracting detailed features in the cleaned manufacturing data; performing initial feature screening on the detailed features to obtain screening features; calculating feature similarities corresponding to the screening features; determining key features corresponding to the production and manufacturing data based on the feature similarities; and extracting key feature parameters corresponding to the key features.

[0071] Among them, the cleaning manufacturing data refers to the relatively accurate and complete production and manufacturing data obtained after data cleaning operations such as removing duplications, errors, and missing values; the detailed features refer to the specific and detailed descriptive features extracted from the cleaning manufacturing data regarding material properties, processing parameters, and component size specifications; the screening features refer to the feature set that is obtained after preliminary screening of the detailed features and that may have a potential impact on the production and manufacturing results; the feature similarity refers to the measure of the degree of similarity between different features in terms of numerical values, trends, distributions, etc.; the key features refer to the features that have a significant impact and a decisive role on the key indicators such as the performance and quality of communication mold components among the numerous screening features.

[0072] Furthermore, the data cleaning of the production and manufacturing data can be achieved through data cleaning methods, such as: missing value processing, outlier detection, data standardization and other methods; the extraction of detailed features in the cleaned manufacturing data can be achieved through Python's Scikit-learn library, such as: using the PCA class in the Scikit-learn library to perform principal component analysis on the cleaned data to extract the main features; the initial feature screening of the detailed features can be achieved through feature screening tools, such as: SPSS, R language and other tools; the calculation of the feature similarity corresponding to the screened features can be achieved through the following calculation formula; the determination of the key features corresponding to the production and manufacturing data can be achieved through the WEKA tool, such as: using the CfsSubsetEval evaluator in WEKA combined with the BestFirst search method to select key features; the extraction of key feature parameters corresponding to the key features can be achieved through Python's NumPy library, such as: using NumPy's where function combined with a set threshold to extract key feature parameters.

[0073] As an embodiment of the present invention, the calculating the feature similarity corresponding to the screening feature includes:

[0074] The feature similarity corresponding to the screening feature is calculated using the following formula:

[0075]

[0076] in, represents the feature similarity corresponding to the screening feature, represents the feature index corresponding to the screening feature, Indicates the number of features corresponding to the screening feature, Indicates The weight coefficient of the feature, Represents the feature vector in the screening feature No. elements, Represents the feature vector The mean of Represents the feature vector in the screening feature No. elements, Represents the feature vector The mean of .

[0077] S2. Based on the key characteristic parameters, analyze the component effects corresponding to the communication mold components, identify the core factors corresponding to the component effects, detect the mutual connections between the core factors, and calculate the correlation degree value corresponding to the mutual connections.

[0078] By analyzing the component effects corresponding to the communication mold components based on the key characteristic parameters, the present invention can help us deeply understand the direct relationship between the performance of the components and the key characteristic parameters, and can accurately evaluate the degree of functional realization of the components under specific working conditions, thereby improving production efficiency.

[0079] The component effect refers to the comprehensive results of the performance, function, quality, etc. of the communication mold component in actual use.

[0080] As an embodiment of the present invention, the component effect corresponding to the communication mold component is analyzed based on the key characteristic parameters, including: data integration of the key characteristic parameters to obtain integrated data; data visualization of the integrated data to obtain a visual scatter plot; correlation analysis of the visual scatter plot to obtain a correlation coefficient matrix; based on the correlation coefficient matrix, determining the component factor corresponding to the communication mold; based on the component factor, analyzing the component effect corresponding to the communication mold component.

[0081] Among them, the integrated data refers to the data set obtained after sorting, summarizing and unifying the various types of scattered data collected about key characteristic parameters; the visual scatter plot refers to a data visualization chart that uses the values ​​of two variables as horizontal and vertical coordinates, and uses the distribution of points to show the relationship between the variables; the correlation coefficient matrix refers to a matrix used to reflect the pairwise correlation between multiple variables, and the elements in it represent the correlation coefficients between the corresponding two variables; the component factor refers to the key characteristic factor that has a significant impact on the effect of the communication mold component determined after analyzing the correlation coefficient matrix.

[0082] Furthermore, the data integration of the key characteristic parameters can be achieved through a data integration algorithm, such as a clustering algorithm, a classification algorithm, and the like; the data visualization of the integrated data can be achieved through a visualization tool, such as Tableau, PowerBI, matplotlib, and the like; the correlation analysis of the visual scatter plot can be achieved through an analysis algorithm, such as a Pearson correlation coefficient algorithm, a Spearman correlation coefficient algorithm, and the like; the determination of the component factors corresponding to the communication mold can be achieved through a component model, such as a factor analysis model, a principal component analysis model, and the like; the analysis of the component effects corresponding to the communication mold components can be achieved through an effect analysis tool, such as ANSYS, ABAQUS, and the like.

[0083] The present invention can help to quickly and accurately find the key factors affecting the component effect by identifying the core factors corresponding to the component effect and detecting the mutual connection between the core factors, which helps to more comprehensively grasp the formation mechanism of the component effect, thereby quickly locating the root cause of the problem and improving the efficiency of maintenance and problem solving.

[0084] Among them, the core factors refer to those important factors that play a key and dominant role among the many factors that affect the component effect and have a decisive impact on the performance, quality, function, etc. of the component; the mutual connection degree refers to the dynamic relationship value of mutual influence, mutual correlation, mutual restriction or mutual promotion between the core factors. Optionally, the identification of the core factors corresponding to the component effect can be achieved through Minitab tools, such as: using Minitab to perform variance analysis to find out the factors that have a significant impact on the component effect; the detection of the mutual connection degree between the core factors can be achieved through a deep learning framework, such as: TensorFlow, Keras, etc.

[0085] Furthermore, the present invention provides a quantitative index by calculating the correlation degree value corresponding to the mutual connection degree, which can clearly and intuitively measure the closeness of the interaction between the core factors.

[0086] The correlation degree value refers to a numerical value used to quantitatively describe the closeness or correlation strength between different core factors, which reflects the degree of interaction and influence between the core factors.

[0087] As an embodiment of the present invention, the calculating of the association degree value corresponding to the mutual connection degree includes:

[0088] The correlation degree value corresponding to the mutual connection degree is calculated using the following formula:

[0089]

[0090] in, represents the correlation degree value corresponding to the mutual connection degree, represents the factor index of the key core factor pair in the mutual connection degree, represents the number of key core factor pairs in the mutual connection degree, Indicates The interrelationship values ​​between core factors are Indicates The importance weight of the core factors, represents the factor index of the key influencing factor in the mutual connection degree, represents the number of key influencing factors in the mutual connection degree, Indicates The actual weight of the influencing factors.

[0091] In detail, the key core factor pair refers to a pair of core factors selected from many core factors for key analysis of their mutual relationship. For example, in the communication mold assembly, the hardness of the material and the temperature of the processing technology may constitute a key core factor pair; the mutual connection value refers to the specific measurement value of the correlation between a pair of core factors; the importance weight refers to the value assigned to the relative importance of different core factors in the overall analysis. The larger the weight value, the greater the impact of the core factor on the final result; the key influencing factors refer to those factors that have significant effects and important influences among the factors that can affect the effect of the communication mold assembly; the actual weight value refers to the specific value determined based on actual conditions, experience, data analysis, etc., which is used to reflect the real importance of its influencing factors in the entire system.

[0092] S3. Based on the correlation degree value, screen the effect dominant factors corresponding to the communication mold assembly; based on the effect dominant factors, simulate the actual working scene corresponding to the communication mold assembly, obtain the performance data in the actual working scene, and analyze the data change trend corresponding to the performance data.

[0093] The present invention screens the effect dominant factors corresponding to the communication mold assembly based on the correlation degree value, which can help accurately identify the factors that have the most critical impact on the performance and effect of the communication mold assembly, and can take corresponding measures to adjust and optimize in the early stage of production, thereby reducing the defective rate and subsequent maintenance costs.

[0094] The effect-dominant factors refer to the factors that play a key and main role in the effect of the communication mold components, which are finally determined after in-depth analysis and comprehensive evaluation of the effect factors.

[0095] As an embodiment of the present invention, the method of screening the effect dominant factors corresponding to the communication mold component based on the correlation degree value includes: standardizing the correlation degree value to obtain a standard correlation value; setting a correlation threshold corresponding to the communication mold component based on the standard correlation value; verifying the dominant factors in the communication mold component based on the correlation threshold to obtain a verified dominant factor; analyzing the effect factors corresponding to the verified dominant factors; and screening the effect dominant factors corresponding to the communication mold component based on the effect factors.

[0096] Among them, the standard correlation value refers to the value obtained after processing the original correlation degree value through a certain standardization method (such as mean normalization, Z-score standardization, etc.); the correlation threshold refers to a pre-set limit of the standard correlation value; the verification dominant factor refers to the factor that is considered to play a dominant role in the effect of the communication mold component after preliminary screening and verification based on the correlation threshold; the effect factor refers to a specific indicator or parameter that can reflect the influence of the verification dominant factor on the final effect of the communication mold component.

[0097] Furthermore, the standardization of the correlation degree value can be achieved through a standardization algorithm, such as: Z-score standardization, Min-Max standardization, etc.; the setting of the correlation threshold corresponding to the communication mold component can be achieved through the percentile method; the factor verification of the dominant factor in the communication mold component can be achieved through simulation tools, such as: ANSYS, ABAQUS and other tools; the analysis of the effect factor corresponding to the verified dominant factor can be achieved through an effect analysis method, such as: sensitivity analysis, variance analysis and other methods; the screening of the effect dominant factor corresponding to the communication mold component can be achieved through the numpy and pandas libraries in Python.

[0098] The present invention simulates the actual working scenario corresponding to the communication mold assembly based on the effect-dominant factors, and can clearly predict the performance under various complex and real working conditions, which helps to discover potential problems and deficiencies in advance and avoid failures or substandard performance in actual applications, thereby saving a lot of time and cost.

[0099] The actual working scenario refers to the actual environment and working conditions faced by the communication mold assembly in actual use.

[0100] As an embodiment of the present invention, simulating the actual working scenario corresponding to the communication mold component based on the effect-dominant factor includes: identifying the factor range corresponding to the effect-dominant factor; analyzing the working scenario corresponding to the communication mold component based on the factor range; querying the scene parameter group in the working scenario; monitoring the scene working indicators in the scene parameter group; and simulating the actual working scenario corresponding to the communication mold component based on the scene working indicators.

[0101] Among them, the factor range refers to the interval or range of possible values ​​of the effect-dominant factor, including the minimum and maximum values, and all possible values ​​in between; the working scenario refers to a comprehensive description of the specific environment and conditions in which the communication mold component is located in actual application, including but not limited to the combination of multiple factors such as working temperature, pressure, frequency of use, electromagnetic environment, etc.; the scenario parameter group refers to a set of specific parameters that describe a specific working scenario, such as specific temperature values, pressure values, frequency of use values, etc.; the scenario working index refers to the specific parameters or standards used to measure the working performance and status of the communication mold component in a specific working scenario, such as the deformation of the component, the stability of signal transmission, the energy consumption level, etc.

[0102] Furthermore, the identification of the factor range corresponding to the effect-dominant factor can be achieved through a range identification tool, such as Excel, Matlab and other tools; the analysis of the working scenario corresponding to the communication mold component can be achieved through a scenario analysis method, such as FMEA, FTA and other methods; the query of the scenario parameter group in the working scenario can be achieved through a data query tool, such as SQL, Python's Pandas library and other tools; the monitoring of the scenario work indicators in the scenario parameter group can be achieved through an indicator monitoring tool, such as NI LabVIEW, Siemens SIMATIC and other tools; the simulation of the actual working scenario corresponding to the communication mold component can be achieved through a scenario simulation method, such as numerical simulation, physical simulation and other methods.

[0103] By acquiring the performance data in the actual working scenario and analyzing the data change trend corresponding to the performance data, the present invention can understand the performance dynamics of the communication mold component under actual use conditions, discover potential problems or anomalies in time, and thus improve the process, material or structure in a targeted manner to improve the overall performance and reliability of the component.

[0104] Among them, the performance data refers to specific values ​​or information that can reflect the various performance characteristics of the communication mold component in the actual working scenario, such as quantitative indicators of the component's compressive strength, wear resistance, signal transmission stability and accuracy, working temperature, energy consumption level, etc.; the data change trend refers to the overall trend and change pattern of the performance data in a certain time series or under different conditions, which may be rising, falling, fluctuating, stable and other trends. For example, with the increase of usage time, the compressive strength of the component gradually decreases; or at different working temperatures, the stability of signal transmission shows a trend of first increasing and then decreasing. Optionally, the acquisition of the performance data in the actual working scenario can be achieved through high-precision sensors, such as pressure sensors, temperature sensors and other tools; the analysis of the data change trend corresponding to the performance data can be achieved through Python's data analysis library, such as Pandas, NumPy and other tools.

[0105] S4. Based on the data change trend, determine the mold performance area corresponding to the communication mold assembly, locate the defects in the mold performance area, obtain performance defect points, and construct a performance defect curve corresponding to the communication mold assembly based on the performance defect points.

[0106] The present invention determines the mold performance area corresponding to the communication mold assembly based on the data change trend, and can accurately locate the advantages and disadvantages of the mold in performance, which helps to carry out targeted optimization and improvement, concentrate resources on key areas, and improve the efficiency and effectiveness of subsequent mold improvements.

[0107] The mold performance area refers to a region with similar performance characteristics obtained by dividing the mold performance data according to the performance boundary.

[0108] As an embodiment of the present invention, determining the mold performance area corresponding to the communication mold component based on the data change trend includes: identifying the performance change law corresponding to the data change trend; querying the performance change sequence corresponding to the performance change law; determining the performance boundary corresponding to the communication mold component based on the performance change sequence; and dividing the mold performance data in the communication mold component into mold performance areas based on the performance boundary.

[0109] Among them, the performance change law refers to the general and repetitive change pattern and characteristics of the performance indicators of the communication mold components under different conditions or over time, such as the rise, fall, periodic fluctuation or stability of the performance indicators; the performance change sequence refers to the record or numerical set of performance changes arranged in a certain order, reflecting the specific changes in performance at different time points or under different conditions; the performance boundary distinguishes the boundaries of different performance levels or states, which defines the threshold for the transition of mold performance from one level to another; the mold performance data refers to the specific values ​​and information that describe the performance of the communication mold components at work, such as measurement data on strength, precision, durability, etc.

[0110] Furthermore, the identification of the performance change law corresponding to the data change trend can be achieved through a clustering algorithm, such as the K-Means algorithm, which can cluster similar data change trends into one category to discover the law; the query of the performance change sequence corresponding to the performance change law can be achieved through a binary search algorithm, such as the binary search algorithm can quickly find the performance change sequence that meets the conditions in the ordered data; the determination of the performance boundary corresponding to the communication mold component can be achieved through a box plot algorithm, such as the box plot algorithm can determine the approximate range of the performance boundary based on the quartiles of the data; the division of the mold performance data in the communication mold component into mold performance areas can be achieved through a decision tree algorithm, such as the decision tree algorithm can classify the mold performance data according to the set conditions and rules and divide it into different areas.

[0111] The present invention locates defects in the mold performance area to obtain performance defect points, and can accurately find the specific location where the mold performance problem occurs, avoiding blind overall investigation and repair, and greatly improving the efficiency and pertinence of problem solving.

[0112] Among them, the performance defect points refer to those specific positions or links in the mold performance area of ​​the communication mold assembly that show performance levels significantly lower than normal or do not meet the expected performance standards. Optionally, the defect location of the mold performance area can be achieved through non-destructive testing methods, such as ultrasonic testing, X-ray testing, etc.

[0113] Furthermore, based on the performance defect points, the present invention constructs a performance defect curve corresponding to the communication mold assembly, which can present in an intuitive manner the changes in performance defects over time, number of uses or other related variables, so as to better predict possible problems in the future and take preventive measures in advance.

[0114] The performance defect curve refers to a curve constructed based on the defect density distribution diagram, which describes the changing trend of defects over time or other influencing factors.

[0115] As an embodiment of the present invention, constructing a performance defect curve corresponding to the communication mold component based on the performance defect point includes: identifying defect point data corresponding to the performance defect point; analyzing data frequency values ​​corresponding to the defect point data; querying the defect mode corresponding to the data frequency value; based on the defect mode, determining the defect distribution area corresponding to the communication mold component; performing defect density analysis on the defect distribution area to obtain a defect density distribution map; and constructing a performance defect curve corresponding to the communication mold component based on the defect density distribution map.

[0116] Among them, the defect point data refers to the specific information of performance defects recorded in actual production or testing, including defect type, occurrence location, impact degree, etc.; the data frequency value refers to the frequency of occurrence of various defects in the defect point data; the defect pattern refers to the common defect types and distribution patterns identified based on the data frequency value; the defect distribution area refers to the area where defects are more concentrated in the communication mold assembly determined based on the defect pattern; the defect density distribution diagram refers to a diagram obtained through defect density analysis, which shows the spatial distribution of defects in the communication mold assembly.

[0117] Furthermore, the identification of the defect point data corresponding to the performance defect point can be achieved through database management tools, such as MySQL, PostgreSQL and other tools; the analysis of the data frequency value corresponding to the defect point data can be achieved through statistical tools, such as SPSS, R language and other tools; the query of the defect mode corresponding to the data frequency value can be achieved through data mining tools, such as WEKA, RapidMiner and other tools; the determination of the defect distribution area corresponding to the communication mold component can be achieved through data visualization tools, such as Tableau, Power BI and other tools; the defect density analysis of the defect distribution area can be achieved through Python's scipy library; the construction of the performance defect curve corresponding to the communication mold component can be achieved through functions in R language, such as lm function, nls function, etc.

[0118] S5. Perform an impact analysis on the performance defect curve to obtain defect influencing factors, formulate mold improvement measures corresponding to the communication mold assembly based on the defect influencing factors, simulate and implement the mold improvement measures to obtain simulation effect parameters, and generate an effect analysis report corresponding to the communication mold assembly based on the simulation effect parameters.

[0119] The present invention obtains defect influencing factors by performing an impact analysis on the performance defect curve, and can fundamentally understand the causes of performance defects in communication mold components, and can specifically improve material selection, processing accuracy, assembly methods and other aspects, thereby improving the overall quality and reliability of the mold.

[0120] Among them, the defect influencing factors refer to various specific reasons and conditions that directly or indirectly lead to performance defects of communication mold components, such as material quality, processing accuracy, frequency of use, etc.

[0121] As an embodiment of the present invention, the impact analysis of the performance defect curve to obtain defect influencing factors includes: identifying key defect features in the performance defect curve; analyzing historical data and environmental conditions corresponding to the key defect features; querying common cause patterns corresponding to the historical data and environmental conditions; based on the common cause patterns, determining the cause-related area of ​​the communication mold component; and performing impact analysis on the cause-related area to obtain defect influencing factors.

[0122] Among them, the key defect characteristics refer to the defect characteristics that are particularly significant, representative or have a significant impact on the mold performance in the performance defect curve, such as the sharp decline section and frequent fluctuation area in the curve; the historical data refers to various relevant data recorded about the communication mold components in the past production, use, testing, etc., including manufacturing process parameters, performance test results, maintenance records, etc.; the environmental conditions refer to the external factors and conditions of the communication mold components during operation or storage, such as temperature, humidity, pressure, electromagnetic interference, etc.; the common cause mode refers to the typical cause combination and law that often appears in similar communication mold components or related fields that lead to defects; the cause association area refers to the specific part, working link or parameter range of the mold component that has a high correlation with the cause of the defect.

[0123] Furthermore, the identification of key defect features in the performance defect curve can be achieved through machine learning algorithms, such as cluster analysis, principal component analysis and other algorithms; the analysis of historical data and environmental conditions corresponding to the key defect features can be achieved through correlation analysis algorithms, such as Pearson correlation coefficient, etc.; the query of common cause patterns corresponding to the historical data and environmental conditions can be achieved through string matching algorithms, such as KMP algorithm, etc.; the determination of the cause-related area of ​​the communication mold component can be achieved through area determination tools, such as JMP, Stata and other tools; the impact analysis method of the cause-related area can be achieved through methods such as fault tree analysis, root cause analysis and other methods.

[0124] Based on the defect influencing factors, the present invention formulates mold improvement measures corresponding to the communication mold assembly, which can solve the problems existing in the mold in a targeted manner, avoid the waste of resources and time costs caused by blind improvements, and thus efficiently improve the performance and quality of the mold.

[0125] Among them, the mold improvement measures refer to a series of specific and targeted actions and methods taken based on the analysis and research of factors affecting the defects in order to eliminate or alleviate the performance defects in the communication mold components. These measures may include adjusting the design structure of the mold, replacing better quality materials, improving the manufacturing process, optimizing the use and maintenance methods, introducing new detection and monitoring methods, etc. Optionally, the mold improvement measures corresponding to the communication mold components can be implemented through CAD tools, such as AutoCAD, SolidWorks and other tools.

[0126] Furthermore, the present invention simulates the implementation of the mold improvement measures to obtain simulation effect parameters, and pre-evaluates the feasibility and effectiveness of the improvement measures, which can accurately predict the effects of the improvement measures and help improve the efficiency and quality of the entire production process.

[0127] Among them, the simulation effect parameters refer to a series of quantitative indicators and data obtained during the simulation implementation of mold improvement measures, which are used to measure and evaluate the improvement effect. These parameters may include but are not limited to mold production speed, product accuracy and quality indicators, material utilization, energy consumption, equipment operation stability, cost reduction, shortening of production cycle time, etc. Optionally, the simulation implementation of the mold improvement measures can be achieved through a simulation algorithm, such as: particle swarm optimization algorithm, ant colony algorithm and other algorithms.

[0128] Furthermore, the present invention generates an effect analysis report corresponding to the communication mold component based on the simulation effect parameters, which can present the actual effects of the mold improvement measures in a clear, systematic and quantitative manner, so that the specific performance of the improvement measures in various key aspects can be quickly and comprehensively understood.

[0129] Among them, the effect analysis report refers to a document that comprehensively, systematically, and in detail explains and evaluates the simulated effects produced by the improvement measures of the communication mold components. It integrates the simulation effect parameters and demonstrates the specific effects and effectiveness of the improvement measures in improving mold performance, optimizing production processes, reducing costs, and improving quality through data charts, text descriptions, and analytical reasoning. At the same time, it will also analyze and explain the parts that did not meet expectations, and put forward suggestions and directions for further improvement. Optionally, the generation of the effect analysis report corresponding to the communication mold component can be achieved through visualization tools, such as Microsoft Power BI, Tableau and other tools.

[0130] First of all, by acquiring the production and manufacturing data of the communication mold assembly, the present invention can fully understand the details of the mold assembly in the production process, which helps to accurately grasp the key factors such as material properties, processing parameters and component size specifications, and provide a solid foundation for subsequent effect analysis and performance evaluation. At the same time, by analyzing the component effect corresponding to the communication mold assembly based on the key characteristic parameters, the present invention can help us deeply understand the direct relationship between the performance of the component and the key characteristic parameters, and can accurately evaluate the degree of functional realization of the component under specific working conditions, thereby improving production efficiency. The present invention screens the effect dominant factors corresponding to the communication mold assembly based on the correlation degree value, which can help accurately identify the effects on the performance of the communication mold assembly. The factors that have the most critical impact on the effect can be adjusted and optimized by taking corresponding measures in the early stage of production, thereby reducing the defective rate and the subsequent maintenance costs. The present invention determines the mold performance area corresponding to the communication mold assembly based on the data change trend, and can accurately locate the advantages and disadvantages of the mold in terms of performance, which helps to optimize and improve in a targeted manner, concentrate resources on key areas, and improve the efficiency and effect of subsequent mold improvements. Furthermore, the present invention analyzes the performance defect curve to obtain the defect influencing factors, and can fundamentally understand the causes of performance defects of communication mold assemblies. It can improve material selection, processing accuracy, assembly methods and other links in a targeted manner, thereby improving the overall quality and reliability of the mold. Therefore, the method and system for implementing the effect analysis of communication mold assemblies based on deep learning proposed in the present invention can improve the accuracy of the effect analysis of communication mold assemblies.

[0131] like Figure 2 As shown, it is a functional module diagram of a system for analyzing the effect of a communication mold assembly based on deep learning provided by an embodiment of the present invention.

[0132] The effect analysis system 200 for realizing communication mold components based on deep learning of the present invention can be installed in an electronic device. According to the functions implemented, the effect analysis system 200 for realizing communication mold components based on deep learning can include a parameter identification module 201, a correlation degree value calculation module 202, a trend analysis module 203, a curve construction module 204 and a report generation module 205. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0133] In this embodiment, the functions of each module / unit are as follows:

[0134] The parameter identification module 201 is used to obtain the manufacturing data of the communication mold component, the manufacturing data includes: material properties, processing parameters and component size specifications, and extract key characteristic parameters in the manufacturing data;

[0135] The correlation degree value calculation module 202 is used to analyze the component effect corresponding to the communication mold component based on the key characteristic parameters, identify the core factors corresponding to the component effect, detect the mutual connection between the core factors, and calculate the correlation degree value corresponding to the mutual connection;

[0136] A trend analysis module 203 is used to screen the effect dominant factors corresponding to the communication mold component based on the correlation degree value, simulate the actual working scene corresponding to the communication mold component based on the effect dominant factors, obtain the performance data in the actual working scene, and analyze the data change trend corresponding to the performance data;

[0137] A curve construction module 204 is used to determine the mold performance area corresponding to the communication mold component based on the data change trend, locate the defects in the mold performance area, obtain performance defect points, and construct a performance defect curve corresponding to the communication mold component based on the performance defect points;

[0138] The report generation module 205 is used to perform an impact analysis on the performance defect curve to obtain defect influencing factors, formulate mold improvement measures corresponding to the communication mold assembly based on the defect influencing factors, simulate the implementation of the mold improvement measures to obtain simulation effect parameters, and generate an effect analysis report corresponding to the communication mold assembly based on the simulation effect parameters.

[0139] In detail, each module described in the effect analysis system 200 for implementing communication mold components based on deep learning described in the embodiment of the present invention adopts the same technical means as the effect analysis method for implementing communication mold components based on deep learning described in the accompanying drawings when used, and can produce the same technical effects, which will not be repeated here.

[0140] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for analyzing the effect of communication mold components based on deep learning, characterized in that: The method comprises: Acquire manufacturing data of the communication mold component, the manufacturing data including: material properties, processing parameters and component size specifications, and extract key characteristic parameters from the manufacturing data, wherein the extracting of the key characteristic parameters from the manufacturing data includes: Performing data cleaning on the production and manufacturing data to obtain cleaned manufacturing data; Extracting detailed features from the cleaning manufacturing data; performing preliminary feature screening on the detailed features to obtain screening features; Calculating the feature similarity corresponding to the screening feature, wherein calculating the feature similarity corresponding to the screening feature includes: The feature similarity corresponding to the screening feature is calculated using the following formula: in, represents the feature similarity corresponding to the screening feature, represents the feature index corresponding to the screening feature, Indicates the number of features corresponding to the screening feature, Indicates The weight coefficient of the feature, Represents the feature vector in the screening feature No. elements, Represents the feature vector The mean of Represents the feature vector in the screening feature No. elements, Represents the feature vector The mean of Based on the feature similarity, determining key features corresponding to the production and manufacturing data; Extracting key feature parameters corresponding to the key features; Based on the key characteristic parameters, analyzing the component effects corresponding to the communication mold components, identifying the core factors corresponding to the component effects, detecting the mutual connections between the core factors, and calculating the correlation degree values ​​corresponding to the mutual connections; Based on the correlation degree value, the effect dominant factor corresponding to the communication mold component is screened; based on the effect dominant factor, the actual working scene corresponding to the communication mold component is simulated, the performance data in the actual working scene is obtained, and the data change trend corresponding to the performance data is analyzed; Based on the data change trend, determine the mold performance area corresponding to the communication mold component, locate the defects in the mold performance area, obtain performance defect points, and construct a performance defect curve corresponding to the communication mold component based on the performance defect points; An impact analysis is performed on the performance defect curve to obtain defect influencing factors, and based on the defect influencing factors, mold improvement measures corresponding to the communication mold assembly are formulated, and the mold improvement measures are simulated and implemented to obtain simulation effect parameters, and based on the simulation effect parameters, an effect analysis report corresponding to the communication mold assembly is generated.

2. The method for analyzing the effect of communication mold components based on deep learning according to claim 1, characterized in that: The analyzing the component effect corresponding to the communication mold component based on the key characteristic parameters includes: Performing data integration on the key characteristic parameters to obtain integrated data; Performing data visualization on the integrated data to obtain a visual scatter plot; Performing correlation analysis on the visual scatter plot to obtain a correlation coefficient matrix; Based on the correlation coefficient matrix, determining the component factors corresponding to the communication mold; Based on the component factors, component effects corresponding to the communication mold components are analyzed.

3. The method for analyzing the effect of communication mold components based on deep learning according to claim 1, characterized in that: The calculating the association degree value corresponding to the mutual connection degree includes: The correlation degree value corresponding to the mutual connection degree is calculated using the following formula: in, represents the correlation degree value corresponding to the mutual connection degree, represents the factor index of the key core factor pair in the mutual connection degree, represents the number of key core factor pairs in the mutual connection degree, Indicates The interrelationship values ​​between core factors are Indicates The importance weight of the core factors, represents the factor index of the key influencing factor in the mutual connection degree, represents the number of key influencing factors in the mutual connection degree, Indicates The actual weight of the influencing factors.

4. The method for analyzing the effect of communication mold components based on deep learning according to claim 1, characterized in that: The screening of the effect dominant factors corresponding to the communication mold components based on the correlation degree values ​​includes: Standardizing the correlation degree value to obtain a standard correlation value; Based on the standard association value, setting an association threshold value corresponding to the communication mold component; Based on the association threshold, factor verification is performed on the dominant factor in the communication mold component to obtain a verified dominant factor; Analyze the effect factors corresponding to the verification leading factors; Based on the effect factors, the effect dominant factors corresponding to the communication mold components are screened.

5. The method for analyzing the effect of communication mold components based on deep learning according to claim 1, characterized in that: The simulating the actual working scene corresponding to the communication mold assembly based on the effect dominant factor includes: Identify the factor range corresponding to the dominant factor of the effect; Based on the range of factors, analyzing the working scenario corresponding to the communication mold assembly; Querying a scene parameter group in the working scene, and monitoring scene working indicators in the scene parameter group; Based on the scenario work indicators, the actual work scenario corresponding to the communication mold assembly is simulated.

6. The method for analyzing the effect of communication mold components based on deep learning according to claim 1, characterized in that: Determining the mold performance area corresponding to the communication mold component based on the data change trend includes: Identify the performance change law corresponding to the data change trend; Querying a performance change sequence corresponding to the performance change rule; Based on the performance change sequence, determining a performance boundary corresponding to the communication mold component; Based on the performance boundaries, mold performance data in the communication mold assembly is divided into mold performance areas.

7. The method for analyzing the effect of communication mold components based on deep learning according to claim 1, characterized in that: The step of constructing a performance defect curve corresponding to the communication mold component based on the performance defect point includes: Identify defect point data corresponding to the performance defect point, and analyze data frequency values ​​corresponding to the defect point data; Querying the defect mode corresponding to the data frequency value; Based on the defect mode, determining a defect distribution area corresponding to the communication mold component; Performing defect density analysis on the defect distribution area to obtain a defect density distribution map; Based on the defect density distribution diagram, a performance defect curve corresponding to the communication mold assembly is constructed.

8. The effect analysis system of communication mold components based on deep learning is characterized by: The system is used to execute the method for analyzing the effect of a communication mold component based on deep learning as described in any one of claims 1 to 7, and the system comprises: The parameter identification module is used to obtain the manufacturing data of the communication mold component, the manufacturing data includes: material properties, processing parameters and component size specifications, and extract key characteristic parameters from the manufacturing data, wherein the key characteristic parameters from the manufacturing data are extracted, including: Performing data cleaning on the production and manufacturing data to obtain cleaned manufacturing data; Extracting detailed features from the cleaning manufacturing data; performing preliminary feature screening on the detailed features to obtain screening features; Calculating the feature similarity corresponding to the screening feature, wherein calculating the feature similarity corresponding to the screening feature includes: The feature similarity corresponding to the screening feature is calculated using the following formula: in, represents the feature similarity corresponding to the screening feature, represents the feature index corresponding to the screening feature, Indicates the number of features corresponding to the screening feature, Indicates The weight coefficient of the feature, Represents the feature vector in the screening feature No. elements, Represents the feature vector The mean of Represents the feature vector in the screening feature No. elements, Represents the feature vector The mean of Based on the feature similarity, determining key features corresponding to the production and manufacturing data; Extracting key feature parameters corresponding to the key features; A correlation degree value calculation module is used to analyze the component effect corresponding to the communication mold component based on the key characteristic parameters, identify the core factors corresponding to the component effect, detect the mutual connection between the core factors, and calculate the correlation degree value corresponding to the mutual connection; A trend analysis module, for screening the effect dominant factors corresponding to the communication mold assembly based on the correlation degree value, simulating the actual working scene corresponding to the communication mold assembly based on the effect dominant factors, obtaining the performance data in the actual working scene, and analyzing the data change trend corresponding to the performance data; A curve construction module, used to determine the mold performance area corresponding to the communication mold component based on the data change trend, locate the defects in the mold performance area, obtain performance defect points, and construct a performance defect curve corresponding to the communication mold component based on the performance defect points; A report generation module is used to perform an impact analysis on the performance defect curve to obtain defect influencing factors, formulate mold improvement measures corresponding to the communication mold assembly based on the defect influencing factors, simulate the implementation of the mold improvement measures to obtain simulation effect parameters, and generate an effect analysis report corresponding to the communication mold assembly based on the simulation effect parameters.

Images