Manufacturing optimization method and system for electric iron accessories

Abstract

The invention relates to the technical field of electric iron accessories, and discloses a manufacturing optimization method and system for an electric iron accessory, and the system comprises a manufacturing process path determination module which is configured to determine a manufacturing process needing to be optimized based on the type of the electric iron accessory and a manufacturing route; the original process parameter determination module is configured to analyze the first image data and the original image data and determine original process parameters of the manufacturing process needing to be optimized based on an analysis result; the execution module is configured to obtain the manufacturing success rate of the sample finished product; the optimization judgment module is configured to judge whether the original process parameters are optimized or not according to the manufacturing success rate; and the parameter optimization module is configured to optimize the original process parameters based on the real-time process parameters and the second image data to obtain optimized process parameters. According to the manufacturing optimization system for the electric power iron accessory, resource waste caused by indifference optimization of all process procedures is avoided, and optimization pertinence and efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of electric railway accessories technology, and more specifically, to a manufacturing optimization method and system for electric railway accessories. Background Technology

[0002] Power line accessories are indispensable metal components in power systems, primarily used in power facilities such as transmission lines and substations. They play a vital role in supporting, fixing, connecting, insulating, and protecting conductors and equipment. Common power line accessories include crossarms, clamps, insulator string hardware, connecting plates, hanging plates, wire clamps, guy rods, protective pipes, and cable supports. They are typically made from materials such as cast iron and steel through processes such as casting, forging, welding, and machining. Their quality and performance directly affect the safe and stable operation of the power system.

[0003] When casting electrical ferroelectric accessories, traditional processes often rely on experienced technicians manually adjusting key process parameters such as melting temperature, casting speed, and molding sand ratio based on factors like casting structure and material properties. This not only demands extremely high skill levels from operators but also makes it difficult to establish a standardized and refined parameter control system. Human error can easily lead to defects such as porosity, shrinkage cavities, and cracks in the castings, affecting product quality stability. Furthermore, monitoring of raw material consumption, equipment operating status, and production progress during production relies heavily on manual recording and periodic inspections. Data collection is often untimely and inaccurate, failing to provide real-time and effective data support for production optimization, resulting in low production efficiency and significant resource waste. In addition, electrical ferroelectric accessories are diverse, with different models and specifications having varying manufacturing processes. Traditional production management models struggle to quickly and flexibly switch and adapt process schemes, failing to meet the market demands for small-batch, multi-variety production.

[0004] Therefore, it is necessary to design a manufacturing optimization method and system for electric railway accessories to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes a manufacturing optimization method and system for electric railway accessories, aiming to achieve precise optimization of process parameters, real-time monitoring of the production process, and efficient allocation of resources in the manufacturing process of electric railway accessories through intelligent and digital means.

[0006] In one aspect, the present invention proposes a manufacturing optimization system for electric railway accessories, comprising: The manufacturing process path determination module is configured to determine the type of electric iron accessory to be manufactured and the manufacturing route, and to determine the manufacturing process to be optimized based on the type of electric iron accessory and the manufacturing route. The original process parameter determination module is configured to collect first image data and corresponding original image data of the electric iron accessory to be manufactured before the manufacturing process to be optimized, and to parse the first image data and the original image data to determine the original process parameters of the manufacturing process to be optimized based on the parsing results. The execution module is configured to manufacture a sample of the electric iron accessory to be manufactured using the original process parameters, so as to obtain a finished sample and obtain the manufacturing success rate of the finished sample. The optimization judgment module is configured to determine whether to optimize the original process parameters based on the manufacturing success rate. The parameter optimization module is configured to collect real-time process parameters during the sample manufacturing process and second image data of the finished sample, and optimize the original process parameters based on the real-time process parameters and the second image data to obtain optimized process parameters.

[0007] Furthermore, when determining the need to optimize the manufacturing process based on the type and manufacturing route of the electric railway accessories, the following includes: Based on the type of electric railway accessory, determine the structural feature category corresponding to the type of electric railway accessory; Based on the manufacturing route, determine the set of process steps corresponding to the manufacturing process of the electric iron accessory to be manufactured; Based on the structural feature category and the set of process steps, the degree of influence of each process step on the manufacturing quality of the electric railway accessory to be manufactured is determined from the preset process association rules. The target process step is determined based on the degree of impact on manufacturing quality. The process corresponding to the target process step is identified as the manufacturing process that needs to be optimized.

[0008] Furthermore, when determining the target process step based on the degree of impact on manufacturing quality, the following steps are included: The degree of quality impact is compared with the corresponding threshold for the degree of quality impact, and the target process step is determined based on the comparison result. If the degree of quality impact is greater than or equal to the threshold of the degree of quality impact, then the process step is determined as the target process step; If the degree of quality impact is less than the threshold value of the degree of quality impact, then the process step will not be determined as the target process step.

[0009] Further, when analyzing the first image data and the original image data, and determining the original process parameters for the manufacturing process to be optimized based on the analysis results, the process includes: The original image data is parsed to obtain the original image features of the electric iron accessory to be manufactured; The first image data is parsed to obtain the first image features of the electric iron accessory to be manufactured; An image feature vector is constructed based on the original image features and the first image features; Collect historical image data that is the same type of electric iron accessory to be manufactured and has the same manufacturing process to be optimized, and construct a historical image dataset; The image feature vector is compared with the historical image feature set, and the original process parameters are determined based on the comparison results. If a historical image feature vector exists in the historical image feature set that is identical to the image feature vector, the historical process parameter corresponding to the historical image feature vector shall be used as the original process parameter. If there is no historical image feature vector in the historical image feature set that is the same as the image feature vector, then the original process parameters are determined based on the image feature vector.

[0010] Further, determining the original process parameters based on the image feature vector includes: Calculate the correlation degree between the original image features and each historical original image feature in the historical image feature set, and extract the historical original image feature corresponding to the highest correlation degree. If the historical original image feature corresponding to the maximum original image correlation degree is unique, then the historical first image feature corresponding to the historical original image feature is determined; Calculate the degree of correlation between the first image feature and the first image with the historical first image feature; The original process parameters are determined based on the maximum original image correlation degree and the first image correlation degree; If the historical original image features corresponding to the maximum original image correlation degree are not unique, then the historical first image feature corresponding to each historical original image feature is determined. Calculate the correlation degree between the first image feature and each of the historical first image features, and extract the maximum first image correlation degree; The original process parameters are determined based on the maximum original image correlation degree and the maximum first image correlation degree.

[0011] Further, when determining the original process parameters based on the maximum original image correlation degree and the first image correlation degree, the following steps are included: The basic process parameters for the electric iron accessory to be manufactured are determined based on the correlation degree of the maximum original image. The basic process parameters are corrected based on the correlation degree of the first image to determine the original process parameters.

[0012] Further, when determining the original process parameters based on the maximum original image correlation degree and the maximum first image correlation degree, the following steps are included: Obtain the average value of the correlation degree of all the original images, and calculate the difference between the maximum correlation degree of the original image and the average value, which is denoted as the original image correlation degree difference; Obtain the average value of the correlation degree of all the first images, and calculate the difference between the maximum correlation degree of the first image and the average value, which is denoted as the first image correlation degree difference; Based on the preset weight allocation rules, corresponding weight coefficients are assigned to the correlation difference between the original image and the correlation difference between the first image. The correlation difference of the original images is multiplied by the corresponding weight coefficient to obtain the correlation contribution value of the original images; Multiply the correlation difference of the first image by the corresponding weight coefficient to obtain the correlation contribution value of the first image; The original image association contribution value and the first image association contribution value are added together to obtain the comprehensive association contribution value; The comprehensive correlation contribution value is compared with a preset process parameter mapping table, and the original process parameters are determined based on the comparison results.

[0013] Furthermore, when determining whether to optimize the original process parameters based on the manufacturing success rate, the following steps are included: The manufacturing success rate is compared with the manufacturing success rate threshold, and the original process parameters are optimized based on the comparison results. If the manufacturing success rate is greater than or equal to the manufacturing success rate threshold, then it is determined that the original process parameters should be optimized. If the manufacturing success rate is less than the manufacturing success rate threshold, it is determined that the original process parameters will not be optimized.

[0014] Further, when optimizing the original process parameters based on the real-time process parameters and the second image data to obtain optimized process parameters, the process includes: The real-time process parameters are compared with the original process parameters to obtain the process parameter deviation value; The second image data is parsed, and the parsing result is compared with the standard second image data to obtain the second image deviation value; The process parameter deviation value is compared with the process parameter deviation threshold, and the second image deviation value is compared with the second image deviation threshold. Based on the comparison results, the optimized parameters of the original process parameters are determined. When the deviation value of the process parameter is greater than or equal to the deviation threshold of the process parameter, and the deviation value of the second image is greater than or equal to the deviation threshold of the second image, the optimized parameter is determined to be the first optimized parameter; When the deviation value of the process parameter is greater than or equal to the deviation threshold of the process parameter, and the deviation value of the second image is less than the deviation threshold of the second image, the optimized parameter is determined to be the second optimized parameter. When the deviation value of the process parameter is less than the deviation threshold of the process parameter, and the deviation value of the second image is greater than or equal to the deviation threshold of the second image, the optimized parameter is determined to be the third optimized parameter. When the deviation value of the process parameter is less than the deviation threshold of the process parameter, and the deviation value of the second image is less than the deviation threshold of the second image, the optimized parameter is determined to be the fourth optimized parameter.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: The manufacturing optimization system for electric railway accessories provided by this invention can accurately locate the manufacturing process to be optimized through the manufacturing process path determination module, avoiding resource waste caused by indiscriminate optimization of all process steps, and improving the targeting and efficiency of optimization. The original process parameter determination module analyzes the first image data and original image data of the electric railway accessory to be manufactured, so that the determination of the original process parameters no longer relies on empiricism, but is based on quantitative analysis of image features, improving the scientificity and accuracy of the initial parameter setting. The execution module provides an objective basis for judging whether subsequent optimization is needed by obtaining sample manufacturing and manufacturing success rate, ensuring the necessity of optimization work. The optimization judgment module intelligently decides whether to perform parameter optimization based on the comparison result of manufacturing success rate and threshold, avoiding blind optimization or ineffective operation when optimization is not needed. The parameter optimization module integrates real-time process parameters during sample manufacturing and second image data of the finished sample to optimize the original process parameters from both dynamic process and static result dimensions. This allows for a more comprehensive and accurate identification of the optimal combination of process parameters, thereby effectively improving the manufacturing success rate and product quality consistency of electric railway accessories. Simultaneously, it enables intelligent monitoring of the production process and efficient resource allocation, reducing production costs and improving overall production efficiency.

[0016] In another aspect, the present invention also proposes a manufacturing optimization method for electric railway accessories, comprising the following steps: Determine the types of electric railway accessories to be manufactured and the manufacturing route, and determine the manufacturing process to be optimized based on the types of electric railway accessories and the manufacturing route; First image data and corresponding original image data of the electric railway accessory to be manufactured before the manufacturing process needs to be optimized are collected, and the first image data and original image data are analyzed. Based on the analysis results, the original process parameters of the manufacturing process to be optimized are determined. The electric iron accessory to be manufactured is sample-manufactured using the original process parameters to obtain a finished sample and the manufacturing success rate of the finished sample is obtained. Whether to optimize the original process parameters is determined based on the manufacturing success rate; Real-time process parameters and second image data of the finished sample are collected during the sample manufacturing process. Based on the real-time process parameters and the second image data, the original process parameters are optimized to obtain optimized process parameters.

[0017] It is understandable that the above-mentioned manufacturing optimization methods and systems for electric railway accessories have the same beneficial effects, and will not be elaborated further here. 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 structural block diagram of a manufacturing optimization system for electric railway accessories provided in an embodiment of the present invention; Figure 2 A flowchart illustrating a manufacturing optimization method for electric railway accessories provided in an embodiment of the present 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] 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.

[0022] See Figure 1 As shown in some embodiments of this application, this embodiment provides a manufacturing optimization system for electric railway accessories, including: The manufacturing process path determination module is configured to determine the type of electric iron accessory to be manufactured and the manufacturing route, and to determine the manufacturing process to be optimized based on the type of electric iron accessory and the manufacturing route. The original process parameter determination module is configured to collect first image data and corresponding original image data of the electric iron accessory to be manufactured before the manufacturing process to be optimized, and to parse the first image data and the original image data to determine the original process parameters of the manufacturing process to be optimized based on the parsing results. The execution module is configured to manufacture a sample of the electric iron accessory to be manufactured using the original process parameters, so as to obtain a finished sample and obtain the manufacturing success rate of the finished sample. The optimization judgment module is configured to determine whether to optimize the original process parameters based on the manufacturing success rate. The parameter optimization module is configured to collect real-time process parameters during the sample manufacturing process and second image data of the finished sample, and optimize the original process parameters based on the real-time process parameters and the second image data to obtain optimized process parameters.

[0023] It is understood that the manufacturing optimization system for electric railway accessories provided in this embodiment can accurately locate the manufacturing process to be optimized through the manufacturing process path determination module, avoiding the waste of resources caused by indiscriminate optimization of all process steps, and improving the targeting and efficiency of optimization. The original process parameter determination module analyzes the first image data and the original image data of the electric railway accessory to be manufactured, so that the determination of the original process parameters no longer relies on empiricism, but is based on quantitative analysis of image features, improving the scientificity and accuracy of the initial parameter setting. The execution module provides an objective basis for judging whether subsequent optimization is needed by obtaining sample manufacturing and manufacturing success rate, ensuring the necessity of optimization work. The optimization judgment module intelligently decides whether to perform parameter optimization based on the comparison result of manufacturing success rate and threshold, avoiding blind optimization or ineffective operation when optimization is not needed. The parameter optimization module integrates real-time process parameters during sample manufacturing and second image data of the finished sample to optimize the original process parameters from both dynamic process and static result dimensions. This allows for a more comprehensive and accurate identification of the optimal combination of process parameters, thereby effectively improving the manufacturing success rate and product quality consistency of electric railway accessories. Simultaneously, it enables intelligent monitoring of the production process and efficient resource allocation, reducing production costs and improving overall production efficiency.

[0024] Specifically, when it is determined that the manufacturing process needs to be optimized based on the type and manufacturing route of the electric railway accessories, the following includes: Based on the type of electric railway accessory, determine the structural feature category corresponding to the type of electric railway accessory; Based on the manufacturing route, determine the set of process steps corresponding to the manufacturing process of the electric iron accessory to be manufactured; Based on the structural feature category and the set of process steps, the degree of influence of each process step on the manufacturing quality of the electric railway accessory to be manufactured is determined from the preset process association rules. The target process step is determined based on the degree of impact on manufacturing quality. The process corresponding to the target process step is identified as the manufacturing process that needs to be optimized.

[0025] In this embodiment, the types of power iron accessories include various categories such as angle steel crossarms, channel steel supports, bolt-type tension clamps, suspension clamps, clamps, and connecting plates. Different types of power iron accessories have significantly different structural characteristics due to different functional requirements.

[0026] In this embodiment, structural features include plate thickness, bending angle, hole size, number and distribution of welding points, and surface roughness requirements. These features directly determine the importance weight of different processes during manufacturing. For example, for bolt-type tension clamps, the thread accuracy of the U-bolts and the flatness of the clamping area are key structural features, and the influence of the corresponding thread processing and stamping processes is higher than that of general cutting processes. For clamp-type products, the tolerance control of the arc and opening size is crucial, so the influence of the rolling process and welding process (if it is a welded clamp) is more prominent. According to the manufacturing route (such as forging → cutting → welding → heat treatment → surface treatment, or stamping → bending → drilling → painting, etc.), the set of process steps that the electric railway accessories to be manufactured can be clearly identified. For example, the manufacturing route of a certain angle steel crossarm may include angle steel cutting, punching, galvanizing, etc.

[0027] In this embodiment, the preset process association rules are mapping relationships constructed based on historical production data, process knowledge, and expert experience. These rules record the influence coefficients or levels of each process step under different structural feature categories on manufacturing quality (such as dimensional accuracy, mechanical properties, surface defects, and assembly compatibility). For example, when the structural feature category is "high-stress welding area," the influence of welding process parameters such as current, voltage, and welding speed on welding strength (a key indicator of manufacturing quality) is set to "extremely high." For "non-stress decorative coatings," the influence of coating process thickness uniformity might be set to "medium." Based on the structural feature category, process steps with influence levels higher than a preset threshold (such as "high" or "extremely high") are selected from the process step set; these are the target process steps, and their corresponding processes are identified as the manufacturing processes requiring optimization. Through this process, the system can focus on key processes that play a decisive role in product quality, ensuring precise allocation of optimization resources.

[0028] Specifically, when determining the target process step based on the degree of impact on manufacturing quality, the following steps are included: The degree of quality impact is compared with the corresponding threshold for the degree of quality impact, and the target process step is determined based on the comparison result. If the degree of quality impact is greater than or equal to the threshold of the degree of quality impact, then the process step is determined as the target process step; If the degree of quality impact is less than the threshold value of the degree of quality impact, then the process step will not be determined as the target process step.

[0029] Understandably, by setting clear thresholds for the degree of quality impact, the system can quantitatively screen target processes, avoiding the subjective biases that may exist in traditional experience-based judgments. For example, when the preset threshold is "impact level ≥ high," if the impact of a certain process is assessed as "extremely high" or "high," it will be included in the target process scope, ensuring that all key processes that have a significant impact on the core quality indicators of the product are covered. Conversely, processes with an impact level of "medium" or "low" (such as some auxiliary cleaning processes, marking processes for non-critical parts, etc.) will not be included as optimization targets for the time being. This ensures the optimization effect while further improving the focus and resource utilization of the optimization work, avoiding the increase in the complexity of the optimization model or the dispersion of the optimization direction due to including too many low-impact processes.

[0030] Specifically, when parsing the first image data and the original image data, and determining the original process parameters for the manufacturing process to be optimized based on the parsing results, the process includes: The original image data is parsed to obtain the original image features of the electric iron accessory to be manufactured; The first image data is parsed to obtain the first image features of the electric iron accessory to be manufactured; An image feature vector is constructed based on the original image features and the first image features; Collect historical image data that is the same type of electric iron accessory to be manufactured and has the same manufacturing process to be optimized, and construct a historical image dataset; The image feature vector is compared with the historical image feature set, and the original process parameters are determined based on the comparison results. If a historical image feature vector exists in the historical image feature set that is identical to the image feature vector, the historical process parameter corresponding to the historical image feature vector shall be used as the original process parameter. If there is no historical image feature vector in the historical image feature set that is the same as the image feature vector, then the original process parameters are determined based on the image feature vector.

[0031] Understandably, the original image features include the overall outline and size features of the electric ferrule, such as the image pixel mapping values ​​of key geometric parameters like length, width, and height; surface texture features, including the image grayscale value distribution, texture direction, and texture density corresponding to surface roughness; structural detail features, covering the edge contours, shape parameters, and position coordinates of local structures such as bolt holes, weld bevels, and bending angles in the image; and potential initial defect features, such as abnormal grayscale areas or morphological features of scratches, dents, and cracks on the raw material surface as seen in the image. The first image features include the real-time structural state features of the electric ferrule to be manufactured before entering the manufacturing process that needs optimization, such as the image representation of the deviation between the actual size and design size of the semi-finished product after previous processing (e.g., edge offset, aperture error, etc. obtained through image comparison), the cleanliness image features after surface treatment (e.g., grayscale differences in oil residue areas, oxide scale coverage area, etc.), and the continuity features of potential process defects that may be introduced by previous processes in the image (e.g., the expansion trend of microcracks left by the stamping process in subsequent processing images).

[0032] Understandably, this image feature comparison method based on historical data allows the determination of original process parameters to fully draw on successful experiences in historical production. When encountering a workpiece to be manufactured that highly matches the features of historical images, the verified historical process parameters can be directly reused, significantly shortening the parameter debugging cycle.

[0033] Specifically, determining the original process parameters based on the image feature vector includes: Calculate the correlation degree between the original image features and each historical original image feature in the historical image feature set, and extract the historical original image feature corresponding to the highest correlation degree. If the historical original image feature corresponding to the maximum original image correlation degree is unique, then the historical first image feature corresponding to the historical original image feature is determined; Calculate the degree of correlation between the first image feature and the first image with the historical first image feature; The original process parameters are determined based on the maximum original image correlation degree and the first image correlation degree; If the historical original image features corresponding to the maximum original image correlation degree are not unique, then the historical first image feature corresponding to each historical original image feature is determined. Calculate the correlation degree between the first image feature and each of the historical first image features, and extract the maximum first image correlation degree; The original process parameters are determined based on the maximum original image correlation degree and the maximum first image correlation degree.

[0034] Understandably, the correlation degree is quantified using the cosine similarity algorithm. This algorithm represents the similarity by calculating the cosine of the angle between image feature vectors, with a value ranging from -1 to 1. A value closer to 1 indicates a higher correlation. For example, in original image feature comparison, if the cosine similarity between the original image feature vector of the power railway accessory to be manufactured and a certain historical original image feature vector in the historical image dataset is 0.92 (maximum original image correlation), and this historical original image feature is unique, then the corresponding historical first image feature vector is further extracted, and the cosine similarity (first image correlation) between the first image feature vector of the part to be manufactured and this historical first image feature vector is calculated, assumed to be 0.88. When the maximum original image correlation corresponds to multiple historical original image features, for example, if there are two historical original image feature vectors and the feature vector of the original image of the workpiece to be manufactured, both with a cosine similarity of 0.92 (maximum original image correlation), then the cosine similarity between the feature vector of the first image of the workpiece to be manufactured and the historical first image feature vectors corresponding to these two historical original image features is calculated respectively. Assuming they are 0.88 and 0.75 respectively, 0.88 is extracted as the maximum first image correlation. This is then combined with the determination of the final original process parameters, thereby ensuring that when there are multiple similar cases in the historical data, the process parameters that best match the current state of the workpiece to be manufactured can be accurately located, further improving the reliability of the original parameter setting.

[0035] Specifically, determining the original process parameters based on the maximum original image correlation degree and the first image correlation degree includes: The basic process parameters for the electric iron accessory to be manufactured are determined based on the correlation degree of the maximum original image. The basic process parameters are corrected based on the correlation degree of the first image to determine the original process parameters.

[0036] In this embodiment, determining the basic process parameters of the electric ferroelectric accessory to be manufactured based on the maximum correlation degree of the original image includes: extracting the historical original image features corresponding to the maximum correlation degree of the original image from the historical image feature set, and obtaining the process parameters actually applied in the historical production process of the historical image features containing these historical original image features, and using them as the basic process parameters of the electric ferroelectric accessory to be manufactured. For example, if a certain historical original image feature has the highest cosine similarity to the original image features of the part to be manufactured (e.g., 0.92), and the process parameters corresponding to the historical image features containing these historical original image features are "welding current 180A, welding voltage 24V, welding speed 300mm / min", then this set of parameters is initially set as the basic process parameters of the current part to be manufactured, ensuring that the parameter starting point is based on historical successful experience, providing a reliable benchmark for subsequent corrections.

[0037] In this embodiment, the basic process parameters are corrected based on the correlation degree of the first image. When determining the original process parameters, the correlation degree of the first image is compared with a preset correction coefficient mapping table to determine the correction coefficient of the basic process parameters; and each item in the basic process parameters is multiplied by its corresponding correction coefficient to obtain the original process parameters.

[0038] It is understandable that the correction coefficient mapping table is constructed based on the relationship between the correlation degree of the first image (i.e., the similarity between the first image features of the electric iron accessory to be manufactured and the historical first image features) and the adjustment range of process parameters.

[0039] Specifically, determining the original process parameters based on the maximum original image correlation degree and the maximum first image correlation degree includes: Obtain the average value of the correlation degree of all the original images, and calculate the difference between the maximum correlation degree of the original image and the average value, which is denoted as the original image correlation degree difference; Obtain the average value of the correlation degree of all the first images, and calculate the difference between the maximum correlation degree of the first image and the average value, which is denoted as the first image correlation degree difference; Based on the preset weight allocation rules, corresponding weight coefficients are assigned to the correlation difference between the original image and the correlation difference between the first image. The correlation difference of the original images is multiplied by the corresponding weight coefficient to obtain the correlation contribution value of the original images; Multiply the correlation difference of the first image by the corresponding weight coefficient to obtain the correlation contribution value of the first image; The original image association contribution value and the first image association contribution value are added together to obtain the comprehensive association contribution value; The comprehensive correlation contribution value is compared with a preset process parameter mapping table, and the original process parameters are determined based on the comparison results.

[0040] In this embodiment, the preset weight allocation rule refers to pre-setting the weight ratio of the original image correlation contribution value and the first image correlation contribution value based on the type of electric iron accessory to be manufactured and the characteristics of the manufacturing process to be optimized. For example, for machining optimization processes (such as boring and milling) where structural dimensional accuracy is the key indicator, the weight coefficient of the original image correlation degree difference (reflecting the similarity between the design dimensions and historical successful cases) can be set to 0.6, and the weight coefficient of the first image correlation degree difference (reflecting the deviation between the actual state of the previous processing and the historical state) can be set to 0.4, because the matching of the basic dimensions has a more significant impact on the process parameters at this time; while for processing processes where surface quality is the core requirement (such as spraying and electroplating), the weight coefficient of the first image correlation degree difference (reflecting the cleanliness, flatness, etc. of the previous surface treatment) can be increased to 0.7, and the weight coefficient of the original image correlation degree difference can be reduced to 0.3, so as to highlight the dominant role of the real-time surface state in adjusting the process parameters. The specific values ​​of the weighting coefficients can be dynamically adjusted through statistical analysis of historical process optimization data (such as determining the significance of the impact of each factor on the process results based on analysis of variance) or expert experience knowledge base, to ensure that the weighting allocation matches the core control objectives of different process types.

[0041] Specifically, when determining whether to optimize the original process parameters based on the manufacturing success rate, the following steps are included: The manufacturing success rate is compared with the manufacturing success rate threshold, and the original process parameters are optimized based on the comparison results. If the manufacturing success rate is greater than or equal to the manufacturing success rate threshold, then it is determined that the original process parameters should be optimized. If the manufacturing success rate is less than the manufacturing success rate threshold, it is determined that the original process parameters will not be optimized.

[0042] In this embodiment, the manufacturing success rate refers to the ratio of the number of qualified products (such as dimensional tolerances, mechanical properties, surface quality, structural integrity, etc.) manufactured using the current original process parameters within a certain production cycle to the total number of products produced in that cycle, usually expressed as a percentage. For example, in the continuous production of 100 pieces of a certain type of power iron accessory, if 92 of them meet the design requirements in all indicators after testing, the manufacturing success rate under the original process parameters is 92%. The manufacturing success rate threshold is the minimum acceptable success rate standard pre-set based on the application scenario, quality level requirements, and industry standards of the power iron accessory to be manufactured. Different types of power iron accessories (such as load-bearing pole iron accessories, connecting hardware iron accessories, etc.) or different quality level requirements (such as iron accessories used in UHV transmission projects and ordinary distribution network iron accessories) correspond to different thresholds.

[0043] Specifically, optimizing the original process parameters based on the real-time process parameters and the second image data to obtain optimized process parameters includes: The real-time process parameters are compared with the original process parameters to obtain the process parameter deviation value; The second image data is parsed, and the parsing result is compared with the standard second image data to obtain the second image deviation value; The process parameter deviation value is compared with the process parameter deviation threshold, and the second image deviation value is compared with the second image deviation threshold. Based on the comparison results, the optimized parameters of the original process parameters are determined. When the deviation value of the process parameter is greater than or equal to the deviation threshold of the process parameter, and the deviation value of the second image is greater than or equal to the deviation threshold of the second image, the optimized parameter is determined to be the first optimized parameter; When the deviation value of the process parameter is greater than or equal to the deviation threshold of the process parameter, and the deviation value of the second image is less than the deviation threshold of the second image, the optimized parameter is determined to be the second optimized parameter. When the deviation value of the process parameter is less than the deviation threshold of the process parameter, and the deviation value of the second image is greater than or equal to the deviation threshold of the second image, the optimized parameter is determined to be the third optimized parameter. When the deviation value of the process parameter is less than the deviation threshold of the process parameter, and the deviation value of the second image is less than the deviation threshold of the second image, the optimized parameter is determined to be the fourth optimized parameter.

[0044] In this embodiment, the real-time process parameters correspond to the original process parameters. The original process parameters include welding current, voltage, etc., while the real-time process parameters are the corresponding data collected by sensors during production. The deviation value between the two is the difference between the real-time value and the set value or the relative deviation percentage. For example, if the original welding current is set to 180A and the real-time acquisition is 185A, the deviation value is 5A (absolute) or 2.78% (relative). The second image data is an image of the finished sample after the optimized manufacturing process, such as images of the weld after welding and the metallographic structure after heat treatment, containing key quality information such as the surface morphology of the product. When parsing the second image data, features are extracted using algorithms such as image segmentation, such as quantitative indicators like weld width. The standard second image data is an image of a historically qualified product after processing in the same process, and the parsing result corresponds to a preset qualified feature range. For example, in a certain welding process, the standard weld width is 5-7mm and the porosity is 0-2 per square millimeter. If the current parsing shows a weld width of 5.8mm (qualified) and porosity of 4 per square millimeter (exceeding the standard), the second image deviation value is obtained by weighted summation of the feature deviation degrees. The process parameter deviation threshold and the second image deviation threshold are preset critical values ​​based on process stability and quality control accuracy, such as a welding current deviation threshold of ±10A and a second image deviation threshold of 0.2.

[0045] In this embodiment, the preferred manufacturing process to be optimized is welding, and the original process parameters include welding current, welding voltage, welding speed, and welding gas flow rate.

[0046] In this embodiment, the optimization parameters are: (current optimization parameters, voltage optimization parameters, speed optimization parameters, and flow rate optimization parameters).

[0047] In this embodiment, the preferred values ​​for the first optimized parameters are: current -8% to -5% of the original welding current (reduced by 5% to 8%), voltage -6% to -3% of the original welding voltage (reduced by 3% to 6%), speed +10% to +15% of the original welding speed (increased by 10% to 15%), and flow rate +5% to +8% of the original welding gas flow rate (increased by 5% to 8%). The preferred values ​​for the second optimized parameters are: current -5% to -2% of the original welding current (reduced by 2% to 5%), voltage -3% to 0% of the original welding voltage (reduced by 0% to 3% or unchanged), speed +5% to +10% of the original welding speed (increased by 5% to 10%), and flow rate +2% to +5% of the original welding gas flow rate (increased by 2% to 5%). The third optimized parameter values ​​are: current 0% to +3% of the original welding current (unchanged or increased by 0% to 3%), voltage 0% to +3% of the original welding voltage (unchanged or increased by 0% to 3%), speed -8% to -5% of the original welding speed (decreased by 5% to 8%), and flow rate +8% to +12% of the original welding gas flow rate (increased by 8% to 12%). The fourth optimized parameter values ​​are: current, voltage, speed, and flow rate are finely adjusted within ±2%, ±2%, ±5%, and ±3% of their original values, respectively.

[0048] It is understandable that the preferred values ​​for the above optimization parameters are derived from statistical analysis of core quality influencing factors in the welding process and a large amount of historical optimization data. For example, in the first optimization parameter scenario (both process parameter deviation and second image deviation exceed the threshold), it means that the welding heat input is too high and the weld formation quality has serious defects. In this case, reducing the welding current (5%-8%) and voltage (3%-6%) can reduce the arc heat input; increasing the welding speed (10%-15%) can reduce the risk of porosity; and increasing the welding gas flow rate (5%-8%) can enhance the protection effect. For the second optimization parameter scenario (process parameter deviation exceeds the threshold but the second image deviation does not), it indicates that the quality impact is not fully manifested or is in a critical state. A relatively mild adjustment range is adopted, such as reducing the current by 2%-5% and the voltage by 0%-3%, to avoid over-adjustment. The third optimization parameter scenario (process parameter deviation is within the threshold but the second image deviation exceeds the threshold) is often due to insufficient protection or poor molten pool fluidity. The focus is on increasing gas flow (8%-12%) to enhance protection, reducing welding speed (5%-8%) to improve molten pool formation, and fine-tuning current and voltage (±3%) to maintain stable heat input. In the fourth optimization parameter scenario, both process parameters and image data deviations are small, indicating process stability. Only ±2%-5% fine-tuning of each parameter is needed to maintain process stability. These optimization parameter ranges will be adjusted according to different welding materials (e.g., low-carbon steel), welding methods (e.g., manual arc welding), and the specific structure of the electric arc welding accessories (e.g., plate thickness). For example, the adjustment range can be appropriately widened for thick-plate load-bearing accessories, while fine-tuning is required for thin-plate connections to ensure quality issues are resolved and mechanical performance and structural safety requirements are met.

[0049] See Figure 2 As shown in some embodiments of this application, this embodiment provides a manufacturing optimization method for electric railway accessories, including the following steps: S100: Determine the type of electric railway accessories to be manufactured and the manufacturing route, and determine the manufacturing process to be optimized based on the type of electric railway accessories and the manufacturing route; S200: Collect the first image data and the corresponding original image data of the electric iron accessory to be manufactured before the manufacturing process to be optimized, and analyze the first image data and the original image data to determine the original process parameters of the manufacturing process to be optimized based on the analysis results; S300: The electric iron accessory to be manufactured is sample-manufactured using the original process parameters to obtain a finished sample and to obtain the manufacturing success rate of the finished sample. S400: Determine whether to optimize the original process parameters based on the manufacturing success rate; S500: Collect real-time process parameters and second image data of the finished sample during the sample manufacturing process, and optimize the original process parameters based on the real-time process parameters and the second image data to obtain optimized process parameters.

[0050] In the description of this invention, it should be understood that the terms "longitudinal", "lateral", "up", "down", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this invention, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention.

[0051] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A manufacturing optimization system for electrical iron accessories, characterized by, The method comprises the following steps: a manufacturing process path determination module is configured to determine the type of power iron accessory to be manufactured and the manufacturing route, and determine the manufacturing process to be optimized based on the type of power iron accessory and the manufacturing route; a raw process parameter determination module is configured to collect first image data of the power iron accessory to be manufactured before the manufacturing process to be optimized and corresponding raw image data, analyze the first image data and the raw image data, and determine the raw process parameters of the manufacturing process to be optimized based on the analysis result; an execution module is configured to perform trial manufacturing on the power iron accessory to be manufactured with the raw process parameters to obtain a trial product, and obtain the manufacturing success rate of the trial product; an optimization judgment module is configured to determine whether to optimize the raw process parameters according to the manufacturing success rate; a parameter optimization module is configured to collect real-time process parameters during trial manufacturing and second image data of the trial product, optimize the raw process parameters based on the real-time process parameters and the second image data, and obtain optimized process parameters.

2. The manufacturing optimization system for electrical iron accessories as claimed in claim 1 wherein, When determining the manufacturing process to be optimized based on the type of power iron accessory and the manufacturing route, the method comprises the following steps: determine the structure feature category corresponding to the type of power iron accessory according to the type of power iron accessory; determine the process procedure set corresponding to the power iron accessory to be manufactured in the manufacturing process according to the manufacturing route; determine the manufacturing quality influence degree of each process procedure on the power iron accessory to be manufactured based on the structure feature category and the process procedure set from the preset process association rule; determine the target process procedure according to the manufacturing quality influence degree; determine the process corresponding to the target process procedure as the manufacturing process to be optimized.

3. The manufacturing optimization system for electrical iron accessories of claim 2, wherein, When determining the target process procedure according to the manufacturing quality influence degree, the method comprises the following steps: compare the quality influence degree with the corresponding quality influence degree threshold value, and determine the target process procedure according to the comparison result; if the quality influence degree is greater than or equal to the quality influence degree threshold value, the process procedure is determined as the target process procedure; if the quality influence degree is less than the quality influence degree threshold value, the process procedure is not determined as the target process procedure.

4. The manufacturing optimization system for electrical iron accessories of claim 3, wherein, When analyzing the first image data and the raw image data and determining the raw process parameters of the manufacturing process to be optimized based on the analysis result, the method comprises the following steps: analyze the raw image data to obtain the raw image features of the power iron accessory to be manufactured; analyze the first image data to obtain the first image features of the power iron accessory to be manufactured; construct an image feature vector based on the raw image features and the first image features; collect historical image data of the same type of power iron accessory and the same manufacturing process to be optimized as the power iron accessory to be manufactured, and construct a historical image data set; compare the image feature vector with the historical image feature set, and determine the raw process parameters according to the comparison result; if there is a same historical image feature vector as the image feature vector in the historical image feature set, taking a historical process parameter corresponding to the historical image feature vector as the original process parameter; if there is no same historical image feature vector as the image feature vector in the historical image feature set, determining the original process parameter according to the image feature vector.

5. The manufacturing optimization system for electrical iron accessories of claim 4, wherein, When determining the original process parameter according to the image feature vector, comprising: calculating the original image correlation degree of the original image feature and each historical original image feature in the historical image feature set, and extracting the historical original image feature corresponding to the maximum original image correlation degree; if the historical original image feature corresponding to the maximum original image correlation degree is unique, determining the historical first image feature corresponding to the historical original image feature; calculating the first image correlation degree of the first image feature and the historical first image feature; determining the original process parameter according to the maximum original image correlation degree and the first image correlation degree; if the historical original image feature corresponding to the maximum original image correlation degree is not unique, determining the historical first image feature corresponding to each historical original image feature; calculating the first image correlation degree of the first image feature and each historical first image feature, and extracting the maximum first image correlation degree; determining the original process parameter according to the maximum original image correlation degree and the maximum first image correlation degree.

6. The manufacturing optimization system for electrical iron accessories of claim 5, wherein, When determining the original process parameter according to the maximum original image correlation degree and the first image correlation degree, comprising: determining the basic process parameter of the to-be-manufactured power iron accessory according to the maximum original image correlation degree; correcting the basic process parameter according to the first image correlation degree to determine the original process parameter.

7. The manufacturing optimization system for electrical iron accessories of claim 6, wherein, When determining the original process parameter according to the maximum original image correlation degree and the maximum first image correlation degree, comprising: obtaining the average value of all the original image correlation degrees, and calculating the difference between the maximum original image correlation degree and the average value, denoted as the original image correlation degree difference value; obtaining the average value of all the first image correlation degrees, and calculating the difference between the maximum first image correlation degree and the average value, denoted as the first image correlation degree difference value; allocating corresponding weight coefficients to the original image correlation degree difference value and the first image correlation degree difference value based on a preset weight allocation rule; multiplying the original image correlation degree difference value by the corresponding weight coefficient to obtain an original image correlation contribution value; multiplying the first image correlation degree difference value by the corresponding weight coefficient to obtain a first image correlation contribution value; adding the original image correlation contribution value and the first image correlation contribution value to obtain a comprehensive correlation contribution value; comparing the comprehensive correlation contribution value with a preset process parameter mapping table to determine the original process parameter according to the comparison result.

8. The manufacturing optimization system for electrical iron accessories of claim 7, wherein, When determining whether to optimize the original process parameter according to the manufacturing success rate, comprising: comparing the manufacturing success rate with a manufacturing success rate threshold, and determining whether to optimize the original process parameter according to a comparison result; if the manufacturing success rate is greater than or equal to the manufacturing success rate threshold, determining to optimize the original process parameter; if the manufacturing success rate is less than the manufacturing success rate threshold, determining not to optimize the original process parameter.

9. The manufacturing optimization system for electrical iron accessories of claim 8, wherein, optimizing the original process parameter based on the real-time process parameter and the second image data, and obtaining an optimized process parameter, including: comparing the real-time process parameter with the original process parameter to obtain a process parameter deviation value; analyzing the second image data and comparing an analysis result with standard second image data to obtain a second image deviation value; comparing the process parameter deviation value with a process parameter deviation threshold and comparing the second image deviation value with a second image deviation threshold, and determining an optimization parameter of the original process parameter according to a comparison result; when the process parameter deviation value is greater than or equal to the process parameter deviation threshold and the second image deviation value is greater than or equal to the second image deviation threshold, determining that the optimization parameter is a first optimization parameter; when the process parameter deviation value is greater than or equal to the process parameter deviation threshold and the second image deviation value is less than the second image deviation threshold, determining that the optimization parameter is a second optimization parameter; when the process parameter deviation value is less than the process parameter deviation threshold and the second image deviation value is greater than or equal to the second image deviation threshold, determining that the optimization parameter is a third optimization parameter; when the process parameter deviation value is less than the process parameter deviation threshold and the second image deviation value is less than the second image deviation threshold, determining that the optimization parameter is a fourth optimization parameter.

10. A method for manufacturing optimization of electrical iron accessories, applied in the system for manufacturing optimization of electrical iron accessories according to any one of claims 1-9, characterized in that, including: determining a type of power iron accessory to be manufactured and a manufacturing route, and determining an original process parameter of a manufacturing process to be optimized based on the type of power iron accessory and the manufacturing route; collecting first image data of the power iron accessory to be manufactured and corresponding original image data before the manufacturing process to be optimized, and analyzing the first image data and the original image data based on an analysis result to determine the original process parameter of the manufacturing process to be optimized; manufacturing a sample of the power iron accessory to be manufactured based on the original process parameter to obtain a sample finished product, and obtaining a manufacturing success rate of the sample finished product; determining whether to optimize the original process parameter according to the manufacturing success rate; collecting a real-time process parameter during the sample manufacturing process and second image data of the sample finished product, and optimizing the original process parameter based on the real-time process parameter and the second image data to obtain an optimized process parameter.

Images