A system and method for comparative analysis of domestic and imported welding materials

Abstract

This invention provides a comparative analysis system and method for domestic and imported welding materials, characterized by scientific accuracy and strong adaptability. By establishing a complete performance evaluation index system and combining a multi-criteria decision-making algorithm and an adaptive learning mechanism, it can provide accurate material comparison analysis results for different welding requirements. The system uses the entropy weight method to objectively determine index weights and combines it with the TOPSIS algorithm for comprehensive scoring, effectively avoiding interference from subjective factors. The introduced adaptive learning function can continuously optimize the weight allocation strategy based on user feedback, enabling the system to continuously improve. Multi-dimensional visualization output makes the comparison results more intuitive and clear. Compared with traditional methods, this invention not only improves the scientificity and accuracy of material selection but also significantly enhances work efficiency, making it particularly suitable for manufacturing fields with high welding quality requirements.

Description

Technical Field

[0001] This invention relates to the field of comparative analysis of domestic and imported welding materials, and in particular to a comparative analysis system and method for domestic and imported welding materials. Background Technology

[0002] The choice of welding materials directly affects the quality and performance of welded joints, and is particularly crucial in important fields such as aerospace, shipbuilding, and pressure vessels. Currently, domestic manufacturers face the dilemma of choosing between domestic and imported welding materials. While imported welding materials offer stable performance, they are expensive and have long delivery cycles; domestic welding materials are cheaper, but their performance varies, and a systematic comparative evaluation system is lacking.

[0003] Existing methods for selecting welding materials suffer from the following problems: First, they rely on subjective judgment based on individual engineer experience, lacking scientifically unified evaluation standards, which may lead to completely different conclusions from different engineers. Second, existing comparison methods often employ simple listing of indicators or weighted averages, failing to address the complex relationships between multiple interdependent performance indicators. Furthermore, traditional methods are ill-suited to the demands of different welding processes; for example, high-strength steel welding emphasizes mechanical properties, while stainless steel welding prioritizes corrosion resistance, and existing systems lack the capability for such targeted adjustments.

[0004] Most existing material selection systems remain at the data management level, lacking intelligent analysis capabilities. Some systems attempting to use evaluation models suffer from problems such as simplistic algorithms and unreasonable weight settings. Especially for complex processes like welding, which involve multiple coupled parameters, simple linear weighting methods cannot accurately reflect the nonlinear relationships between various factors during actual welding. Furthermore, existing systems generally lack learning capabilities, failing to accumulate experience from historical selection data, resulting in the need to reset parameters each time, leading to inefficiency.

[0005] Therefore, there is an urgent need in this field to develop a scientific, rational, and intelligent welding material comparison and analysis system and method that can comprehensively consider multiple factors, adapt to different welding needs, and provide objective and accurate comparison results, thus providing reliable technical support for the selection of welding materials. Summary of the Invention

[0006] The purpose of this invention is to provide a comparative analysis system and method for domestic and imported welding materials to solve the problems existing in the prior art.

[0007] To achieve the above objectives, the present invention provides the following solution: This invention provides a comparative analysis system for domestic and imported welding materials, comprising: The data input module is used to receive performance index data of welding materials and user requirement data; A data preprocessing module, connected to the data input module, is used to standardize the input raw data; The weight allocation module, connected to the data preprocessing module, is used to calculate the weight coefficients of each performance index. The model analysis module, which connects the weight allocation module and the data preprocessing module, is used to calculate the comprehensive score of welding materials based on a multi-criteria decision algorithm. The results output module is connected to the model analysis module and is used to output comparative analysis results and visualization reports.

[0008] Preferably, the data preprocessing module employs a range standardization method to standardize both benefit-type and cost-type indicators. The standardization formula for the benefit-type indicators is as follows: ; The standardized formula for the cost-type indicator is: ; in, Let i be the standardized value of the i-th solder on the j-th index. The original measurement value of the i-th solder on the j-th index. Let j be the minimum value of all solders on the j-th index. The maximum value of all solders on the j-th index.

[0009] Preferably, the weight allocation module uses the entropy weight method to calculate the weight coefficients, specifically including: The entropy value of the j-th index is calculated using the following formula: ; in, Let the entropy value be the j-th index. Calculate the coefficients for the entropy value. , The total number of solders participating in the evaluation. Let i be the proportion of the i-th solder in the j-th index. ; The weighting coefficients are calculated using the following formula: ; in, Let j be the weight coefficient of the j-th indicator. Let be the difference coefficient of the j-th indicator. This represents the total number of evaluation indicators.

[0010] Preferably, the weight allocation module further includes an adaptive learning unit, used to dynamically adjust the weight coefficients based on the user's historical selection data. The formula for calculating the weight coefficients is as follows: ; in, The adjusted weight of the j-th indicator. For learning rate, The average of the historical weights for each user.

[0011] Preferably, the model analysis module employs the TOPSIS algorithm, including: Construct the weighted decision matrix using the following formula: ; in, These are the weighted and standardized index values; The formula for determining the positive and negative ideal solutions is: ; ; in, Let be the positive ideal solution vector. Let j be the value of the j-th indicator in the positive ideal solution. For benefit-type indicators, For cost-related indicators, ; The negative ideal solution vector. Let j be the value of the j-th indicator in the negative ideal solution. For benefit-type indicators, For cost-related indicators, The formula for calculating relative proximity is: ; in, Let be the relative proximity of the i-th solder. Let be the Euclidean distance from the i-th solder to the positive ideal solution. , Let be the Euclidean distance from the i-th solder to the negative ideal solution. .

[0012] This invention also provides a comparative analysis method for domestic and imported welding materials, comprising the following steps: S1. Determine the performance evaluation index system for welding materials; S2. Collect performance data of domestic and imported welding materials; S3. Standardize and preprocess the collected data; S4. Calculate the weighting coefficients for each performance index; S5. Calculate the comprehensive score of each welding material based on a multi-criteria decision-making model; S6. Output the comparative analysis results and the optimal material recommendation.

[0013] Preferably, in step S4, a combined weight determination method is used, including a combination of entropy weighting and analytic hierarchy process (AHP). The AHP calculates subjective weights by constructing a judgment matrix, and the final weights are: ; in, This is the weighting adjustment coefficient. The subjective weights are based on the analytic hierarchy process. The objective weights are obtained based on the entropy weight method.

[0014] Preferably, step S5 further includes establishing a material performance prediction model and using a neural network algorithm to predict and complete the missing performance data. The neural network structure includes an input layer, a hidden layer, and an output layer, wherein the activation function of the hidden layer is the ReLU function.

[0015] Preferably, step S6 includes generating a multi-dimensional radar chart comparison analysis report, wherein the radar chart includes a comparison of at least four dimensions: mechanical properties, process performance, economic performance, and environmental adaptability.

[0016] Preferably, it further includes: S7. Establish a material selection knowledge base, analyze historical selection data through machine learning algorithms, and optimize the weight allocation strategy, including using the random forest algorithm to filter important features.

[0017] The present invention achieves the following beneficial technical effects compared to the prior art: This invention provides a comparative analysis system and method for domestic and imported welding materials, characterized by scientific accuracy and strong adaptability. By establishing a complete performance evaluation index system and combining a multi-criteria decision-making algorithm and an adaptive learning mechanism, it can provide accurate material comparison analysis results for different welding requirements. The system uses the entropy weight method to objectively determine index weights and combines it with the TOPSIS algorithm for comprehensive scoring, effectively avoiding interference from subjective factors. The introduced adaptive learning function can continuously optimize the weight allocation strategy based on user feedback, enabling the system to continuously improve. Multi-dimensional visualization output makes the comparison results more intuitive and clear. Compared with traditional methods, this invention not only improves the scientificity and accuracy of material selection but also significantly enhances work efficiency, making it particularly suitable for manufacturing fields with high welding quality requirements. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating the comparative analysis method for domestic and imported welding materials provided by this invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] The purpose of this invention is to provide a comparative analysis system and method for domestic and imported welding materials. Its core lies in establishing a scientific multi-level evaluation index system, combined with advanced algorithm models and adaptive learning mechanisms, to achieve a comprehensive and objective comparative analysis of domestic and imported welding materials.

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Example 1: The implementation of the comparative analysis system for domestic and imported welding materials of the present invention requires the support of a hardware platform and a software environment. The hardware platform includes a computer processor, storage device and display device, and the software environment includes a database management system and an algorithm running environment.

[0024] The system's implementation begins with the data input module. This module receives various performance data of the welding materials through a human-computer interface, including but not limited to mechanical properties such as tensile strength, yield strength, elongation, and impact toughness; technological performance indicators such as welding speed and spatter rate; and economic indicators such as cost and delivery time. Simultaneously, this module also receives specific welding requirements from the user, such as key parameters like the type of welded structure, service environment conditions, and quality grade requirements. The data input module employs a structured data storage method, classifying and storing different types of data to lay the foundation for subsequent processing.

[0025] After receiving the raw data from the data input module, the data preprocessing module first cleans the data, removing outliers and missing values, and then normalizes the data using the range standardization method. For benefit-type indicators (i.e., indicators where higher values ​​are better), the following formula is used: ; For cost-related indicators (i.e., indicators where smaller values ​​are better), the following formula is used: ; in, Let i be the standardized value of the i-th solder on the j-th index. The original measurement value of the i-th solder on the j-th index. Let j be the minimum value of all solders on the j-th index. The maximum value of all solders on the j-th index. This processing method effectively eliminates the influence of different index units, making the data of each index comparable.

[0026] After receiving the preprocessed data, the weight allocation module objectively calculates the weight coefficients of each indicator using the entropy weight method. The specific implementation process includes: first, calculating the entropy value of the j-th indicator: ; in, Let the entropy value be the j-th index. Calculate the coefficients for the entropy value. , The total number of solders participating in the evaluation. Let i be the proportion of the i-th solder in the j-th index. Then, the weighting coefficients are calculated based on the entropy values: ; in, Let j be the weight coefficient of the j-th indicator. Let be the difference coefficient of the j-th indicator. This represents the total number of evaluation metrics. This module also includes an adaptive learning unit that dynamically adjusts the weight coefficients based on the user's historical data selections. The adjustment formula is: ; in, The adjusted weight of the j-th indicator. This is the learning rate (usually set to 0.7). The weights are averaged based on the user's historical data. This design allows the system to continuously optimize the weight allocation strategy to adapt to different application scenarios.

[0027] The model analysis module uses the TOPSIS (Top-Side Distance Method) algorithm for comprehensive scoring. The specific implementation steps include: first, constructing a weighted decision matrix, where each element... This represents the weighted standardized index value; then, the positive ideal solution is determined. and negative ideal solution ,in, Let be the positive ideal solution vector. Let j be the value of the j-th indicator in the positive ideal solution. For benefit-type indicators, For cost-related indicators, ; The negative ideal solution vector. Let j be the value of the j-th indicator in the negative ideal solution. For benefit-type indicators, For cost-related indicators, Next, calculate the Euclidean distance from each solder to the positive and negative ideal solutions. and Finally, calculate the relative proximity. This value is between 0 and 1, and the larger the value, the better the overall performance of the solder.

[0028] In practical applications, when data is missing, the system also establishes a material performance prediction model, employing a three-layer neural network structure to predict and complete the missing data. The number of nodes in the neural network input layer corresponds to the number of known performance indicators, and the hidden layer uses the ReLU activation function. The output layer corresponds to the metric value that needs to be predicted. The neural network is trained using historical data to accurately predict missing performance data, ensuring the completeness of the analysis process.

[0029] The results output module presents the model analysis results in multiple formats, including a comprehensive score ranking list, optimal material recommendations, and a multi-dimensional radar chart comparative analysis report. The radar chart includes at least four dimensions: mechanical properties, process performance, economics, and environmental adaptability, visually demonstrating the differences in performance of various solders across these dimensions. Simultaneously, the system generates a detailed analysis report, including the weight distribution of each indicator, the scoring calculation process, and selection recommendations.

[0030] During system operation, a material selection knowledge base was also established. Historical selection data was analyzed using the random forest algorithm to filter important features and optimize weight allocation strategies. By constructing multiple decision trees and performing ensemble learning, the random forest algorithm can accurately identify key factors influencing material selection, thereby continuously improving the system's intelligence level.

[0031] Example 2: The comparative analysis method of domestic and imported welding materials using the system in Example 1 is as follows: Figure 1 As shown, it includes the following steps: S1. Determine the performance evaluation index system for welding materials; S2. Collect performance data of domestic and imported welding materials; S3. Standardize and preprocess the collected data; S4. Calculate the weighting coefficients for each performance index; S5. Calculate the comprehensive score of each welding material based on a multi-criteria decision-making model; S6. Output comparative analysis results and optimal material recommendations; S7. Establish a knowledge base for material selection.

[0032] The effectiveness of this invention is verified through a specific application example: In the selection of welding materials for a pressure vessel manufacturing enterprise, the system compared and analyzed five domestic welding materials and three imported welding materials. First, eight key evaluation indicators were determined based on the usage requirements of the pressure vessel, including tensile strength, yield strength, elongation, impact toughness, welding speed, spatter rate, cost, and delivery cycle. Through data preprocessing, weight calculation, and model analysis, the system accurately recommended the domestic welding material with the best overall performance. This welding material, while ensuring mechanical properties, reduced costs by 35% and shortened the delivery cycle by 60% compared to imported welding materials. Actual welding tests verified that the weld joint quality of the recommended welding material fully met the design requirements, proving the effectiveness and practicality of the system.

[0033] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0034] It should be noted that the components mentioned in the above embodiments are all general standard parts or components known to those skilled in the art. Their structures and principles can be learned by those skilled in the art through technical manuals or conventional experimental methods.

[0035] This invention has illustrated its principles and implementation methods using specific examples. The descriptions of these embodiments are merely illustrative of the method and its core ideas; furthermore, those skilled in the art will recognize that modifications may be made to the specific implementation methods and application scope based on the principles of this invention. Therefore, the content of this specification should not be construed as limiting the invention.

Claims

1. A comparative analysis system for domestic and imported welding materials, characterized in that, include: The data input module is used to receive performance index data of welding materials and user requirement data; A data preprocessing module, connected to the data input module, is used to standardize the input raw data; The weight allocation module, connected to the data preprocessing module, is used to calculate the weight coefficients of each performance index. The model analysis module, which connects the weight allocation module and the data preprocessing module, is used to calculate the comprehensive score of welding materials based on a multi-criteria decision algorithm. The results output module is connected to the model analysis module and is used to output comparative analysis results and visualization reports.

2. The comparative analysis system for domestic and imported welding materials according to claim 1, characterized in that, The data preprocessing module employs the range standardization method to standardize both benefit-type and cost-type indicators. The standardization formula for the benefit-type indicators is as follows: ； The standardized formula for the cost-type indicator is: ； in, Let i be the standardized value of the i-th solder on the j-th index. The original measurement value of the i-th solder on the j-th index. Let j be the minimum value of all solders on the j-th index. The maximum value of all solders on the j-th index.

3. The comparative analysis system for domestic and imported welding materials according to claim 1, characterized in that, The weight allocation module uses the entropy weight method to calculate the weight coefficients, specifically including: The entropy value of the j-th index is calculated using the following formula: ； in, Let the entropy value be the j-th index. Calculate the coefficients for the entropy value. , The total number of solders participating in the evaluation. Let i be the proportion of the i-th solder in the j-th index. ; The weighting coefficients are calculated using the following formula: ； in, Let j be the weight coefficient of the j-th indicator. Let be the difference coefficient of the j-th indicator. This represents the total number of evaluation indicators.

4. The comparative analysis system for domestic and imported welding materials according to claim 3, characterized in that, The weight allocation module also includes an adaptive learning unit, used to dynamically adjust the weight coefficients based on the user's historical selection data. The formula for calculating the weight coefficients is as follows: ； in, The adjusted weight of the j-th indicator. For learning rate, The average of the historical weights for each user.

5. The comparative analysis system for domestic and imported welding materials according to claim 1, characterized in that, The model analysis module employs the TOPSIS algorithm and includes: Construct the weighted decision matrix using the following formula: ； in, These are the weighted and standardized index values; The formula for determining the positive and negative ideal solutions is: ； ； in, The vector is the positive ideal solution. Let j be the value of the j-th indicator in the positive ideal solution. For benefit-type indicators, For cost-related indicators, ; The negative ideal solution vector. Let j be the value of the j-th indicator in the negative ideal solution. For benefit-type indicators, For cost-related indicators, The formula for calculating relative proximity is: ； in, Let be the relative proximity of the i-th solder. Let be the Euclidean distance from the i-th solder to the positive ideal solution. , Let be the Euclidean distance from the i-th solder to the negative ideal solution. .

6. A comparative analysis method for domestic and imported welding materials, characterized in that, Includes the following steps: S1. Determine the performance evaluation index system for welding materials; S2. Collect performance data of domestic and imported welding materials; S3. Standardize and preprocess the collected data; S4. Calculate the weighting coefficients for each performance index; S5. Calculate the comprehensive score of each welding material based on a multi-criteria decision-making model; S6. Output the comparative analysis results and the optimal material recommendation.

7. The comparative analysis method for domestic and imported welding materials according to claim 6, characterized in that, In step S4, a combined weight determination method is adopted, including the combined application of the entropy weight method and the analytic hierarchy process (AHP). The AHP calculates the subjective weights by constructing a judgment matrix, and the final weights are: ； in, This is the weighting adjustment coefficient. The subjective weights are based on the analytic hierarchy process. The objective weights are obtained based on the entropy weight method.

8. The comparative analysis method for domestic and imported welding materials according to claim 6, characterized in that, Step S5 also includes establishing a material performance prediction model and using a neural network algorithm to predict and complete the missing performance data. The neural network structure includes an input layer, a hidden layer, and an output layer, wherein the activation function of the hidden layer is the ReLU function.

9. The comparative analysis method for domestic and imported welding materials according to claim 6, characterized in that, Step S6 includes generating a multi-dimensional radar chart comparison analysis report, wherein the radar chart includes a comparison of at least four dimensions: mechanical properties, process performance, economic performance, and environmental adaptability.

10. The comparative analysis method for domestic and imported welding materials according to claim 6, characterized in that, Also includes: S7. Establish a material selection knowledge base, analyze historical selection data through machine learning algorithms, and optimize the weight allocation strategy, including using the random forest algorithm to filter important features.

Images