Intelligent control method and device for automatic cutting of aluminum profile

Abstract

The invention provides an intelligent control method and device for automatic cutting of aluminum profiles, and relates to the technical field of cutting control, and the intelligent control method comprises the steps that cutting related characteristic parameters are extracted, the cutting yield is extracted, and an initial cutting model is constructed; performing clustering division on the initial cutting model according to cutting requirements and aluminum profile differences, performing control parameter homogenization simulation on a plurality of clustering models according to the cutting yield, performing integration to obtain a recombined clustering model, and caching the recombined clustering model; the target aluminum profile is scanned based on the cutting requirements and the aluminum profile difference, and the recombination clustering model is searched for to obtain a matched recombination clustering model; and according to the target aluminum profile input matching recombination clustering model, cutting automatic control is carried out, and a target cutting result is obtained. According to the method and the device, the technical problem of cutting quality fluctuation caused by inaccurate cutting parameter setting due to condition changes such as aluminum profile difference in the prior art can be solved, and the technical effect of improving the cutting quality and the production efficiency is achieved.

Description

Technical Field

[0001] The present application relates to the field of cutting control technology, and in particular to an intelligent control method and device for automated cutting of aluminum profiles. Background Art

[0002] With the continuous development of industrial automation, aluminum profile cutting technology is also improving. However, the existing technology still has some limitations, especially in how to deal with the characteristic differences of different aluminum profiles, optimize cutting parameters and real-time feedback control during the aluminum profile cutting process.

[0003] At present, in the existing technology, the cutting quality of aluminum profiles is affected by many factors, including the material, thickness, hardness and surface condition of the aluminum profiles. These factors will directly affect the selection of tools, cutting speed, feed rate and other parameters during the cutting process. Many aluminum profile cutting systems rely on preset cutting parameters and static models, but in actual production, due to the differences in aluminum profiles and changes in environmental conditions, the cutting effect often fails to meet expectations. Traditional systems lack dynamic adaptability and cannot monitor and adjust parameters during the cutting process in real time. For example, when the tool is worn or the surface of the aluminum profile is uneven, the cutting quality may fluctuate, and the existing system is difficult to detect problems and make adjustments in time, resulting in more waste and inefficient production during the cutting process.

[0004] In summary, the prior art has technical problems in that the cutting parameters are not accurately set due to changes in conditions such as differences in aluminum profiles, resulting in fluctuations in cutting quality, which further affects the stability of the cutting process and production efficiency. Summary of the invention

[0005] The purpose of this application is to provide an intelligent control method and device for automated cutting of aluminum profiles, so as to solve the technical problem in the prior art that the cutting parameters are not accurately set due to changes in conditions such as differences in aluminum profiles, resulting in fluctuations in cutting quality, which further affects the stability of the cutting process and production efficiency.

[0006] In view of the above problems, the present application provides an intelligent control method and device for automated cutting of aluminum profiles.

[0007] In a first aspect, the present application provides an intelligent control method for automated cutting of aluminum profiles, which is implemented by an intelligent control device for automated cutting of aluminum profiles, including: extracting cutting-related feature parameters with cutting-related features as a control based on historical aluminum profile cutting data, and extracting the cutting yield, and constructing an initial cutting model based on the cutting-related feature parameters and the cutting yield; clustering the initial cutting model according to cutting requirements and aluminum profile differences, performing control parameter homogenization simulation on multiple clustering models according to the cutting yield, integrating them to obtain a recombinant clustering model, and caching the recombinant clustering model; scanning the target aluminum profile based on the cutting requirements and aluminum profile differences, and searching the recombinant clustering model to obtain a matching recombinant clustering model; inputting the matching recombinant clustering model according to the target aluminum profile for automatic cutting control to obtain a target cutting result.

[0008] In a second aspect, the present application further provides an intelligent control device for automated cutting of aluminum profiles, which is used to execute an intelligent control method for automated cutting of aluminum profiles as described in the first aspect, including: an initial cutting model construction module, the initial cutting model construction module is used to extract cutting-related feature parameters with cutting-related features as a control based on historical aluminum profile cutting data, and extract the cutting yield, and construct an initial cutting model based on the cutting-related feature parameters and the cutting yield; a recombinant clustering model acquisition module, the recombinant clustering model acquisition module is used to cluster the initial cutting model based on cutting requirements and aluminum profile differences, perform control parameter homogeneity simulation on multiple clustering models based on the cutting yield, integrate to obtain a recombinant clustering model, and cache the recombinant clustering model; a matching recombinant clustering model acquisition module, the matching recombinant clustering model acquisition module is used to scan the target aluminum profile based on the cutting requirements and aluminum profile differences, and search for the recombinant clustering model to obtain a matching recombinant clustering model; a target cutting result acquisition module, the target cutting result acquisition module is used to input the matching recombinant clustering model according to the target aluminum profile for automatic cutting control to obtain a target cutting result.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The cutting-related feature parameters with cutting-related features as the control are extracted according to the historical aluminum profile cutting data, and the cutting yield is extracted, and an initial cutting model is constructed based on the cutting-related feature parameters and the cutting yield; the initial cutting model is clustered and divided according to the cutting requirements and aluminum profile differences, and a plurality of clustering models are simulated for homogenization of control parameters according to the cutting yield, and a recombinant clustering model is obtained by integration, and the recombinant clustering model is cached; the target aluminum profile is scanned based on the cutting requirements and aluminum profile differences, and the recombinant clustering model is searched to obtain a matching recombinant clustering model; the matching recombinant clustering model is input according to the target aluminum profile for automatic cutting control to obtain a target cutting result. That is to say, by realizing the technical goal of dynamic adaptive control in the cutting process, the technical effect of improving cutting quality and production efficiency is achieved.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0013] Figure 1 A schematic diagram of a process for intelligent control of automatic cutting of aluminum profiles for this application;

[0014] Figure 2 This is a schematic diagram of the structure of an intelligent control device for automatic cutting of aluminum profiles in this application.

[0015] Description of reference numerals:

[0016] An initial cutting model building module 11, a recombination clustering model obtaining module 12, a matching recombination clustering model obtaining module 13, and a target cutting result obtaining module 14. DETAILED DESCRIPTION

[0017] The present application provides an intelligent control method and device for automated cutting of aluminum profiles, thereby solving the technical problem in the prior art that the cutting parameter settings are not accurate due to changes in conditions such as differences in aluminum profiles, resulting in fluctuations in cutting quality, and further affecting the stability of the cutting process and production efficiency. The application achieves the technical goal of dynamic adaptive control in the cutting process, and achieves the technical effect of improving cutting quality and production efficiency.

[0018] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0019] For example, please refer to the attached Figure 1 The present application provides an intelligent control method for automatic cutting of aluminum profiles, which is applied to an intelligent control device for automatic cutting of aluminum profiles, and specifically includes the following steps:

[0020] Step 1: extract cutting-related feature parameters with cutting-related features as reference according to historical aluminum profile cutting data, extract cutting yield, and construct an initial cutting model based on the cutting-related feature parameters and the cutting yield.

[0021] Specifically, based on the historical aluminum profile cutting data, the features related to cutting quality and cutting accuracy in the cutting process are extracted as cutting-related features. Cutting-related features can include cutting speed, tool pressure, material hardness, feed rate, etc., which can affect the quality of the cutting results. For example, when the cutting speed is too high, too many burrs may appear on the surface of the aluminum profile, resulting in a decrease in cutting quality. Next, the cutting-related features and indicators for judging the cutting quality during the cutting process are used as the cutting yield. The cutting yield refers to the proportion of aluminum profiles that meet the quality standards among all cut pieces. The neural network is trained through cutting-related features and cutting yield, and then the initial cutting model is constructed, which can predict the yield under given cutting conditions. As the basis for subsequent optimization and adjustment, it helps to more accurately control the cutting process in actual production and improve cutting quality and production efficiency.

[0022] Step 2: Cluster the initial cutting model according to cutting requirements and aluminum profile differences, perform control parameter homogenization simulation on multiple clustering models according to the cutting yield, integrate to obtain a recombinant clustering model, and cache the recombinant clustering model.

[0023] Specifically, the initial cutting models are clustered according to the cutting requirements and the differences in aluminum profiles, which means that all cutting models are divided into several groups according to different cutting requirements and the characteristics of aluminum profiles, and each group of models performs similarly under specific conditions. For example, if soft aluminum profiles are cut, a lower cutting speed may be used, while hard aluminum profiles may require a higher cutting speed, so the models under different conditions are divided into different clusters. Then, according to the cutting yield of each cluster model, its control parameters are homogenized, that is, by adjusting the control parameters in the cutting process (such as cutting speed, feed rate, etc.) to make the parameters in each cluster tend to be consistent, so as to ensure that different models can achieve the best cutting effect under similar conditions. For example, if the cutting yield of a cluster model is lower than expected, its cutting speed and tool pressure parameters are adjusted to make its yield reach the predetermined standard. Then, these adjusted cluster models are integrated to obtain a reorganized cluster model to ensure that all cutting schemes under different conditions can be uniformly optimized to improve cutting efficiency and accuracy. Finally, the reorganized clustering model is cached to facilitate quick calling and application in subsequent production, reducing the time and computational burden of real-time adjustment, thereby improving the level of production automation and cutting quality.

[0024] Step three: Scan the target aluminum profile based on the cutting requirements and the aluminum profile differences, and search for the reorganization clustering model to obtain a matching reorganization clustering model.

[0025] Specifically, scanning the target aluminum profile according to the cutting requirements and the differences in aluminum profiles means that before cutting, the specific properties of the target aluminum profile, such as its material hardness, thickness, surface quality, etc., are used to find a recombinant clustering model that matches it. The recombinant clustering model is generated by optimizing different cutting conditions and contains a variety of cutting schemes that adapt to different aluminum profiles and cutting requirements. Therefore, the matching process is to find a clustering model that best matches the characteristics and cutting requirements of the target aluminum profile, so as to ensure the best cutting effect in the actual cutting process. For example, assuming that the thickness of the target aluminum profile is 5 mm and the hardness is high, then by selecting the recombinant clustering model that optimizes the high-hardness aluminum profile, and adjusting the cutting parameters according to this model, the cutting quality and yield rate can be guaranteed. Matching can accurately control the cutting process, improve efficiency and reduce waste.

[0026] Step 4: Input the matching reorganization clustering model according to the target aluminum profile to perform automatic cutting control to obtain the target cutting result.

[0027] Specifically, according to the characteristics of the target aluminum profile, it is input into the matching recombinant clustering model for automatic cutting control, which means that according to the actual properties of the target aluminum profile, such as thickness, hardness and material type, the clustering model that best matches it is selected, which contains the best combination of cutting parameters. Then, the working state of the cutting machine is automatically adjusted according to the cutting parameters in the model, and various factors in the cutting process, such as tool pressure, cutting depth, feed speed, etc., are controlled to ensure the stability and high quality of the cutting. During the cutting process, the cutting results are monitored in real time to ensure that each cutting meets the quality standards. Finally, the target cutting results are obtained through this automatic control to ensure that this standard is met or exceeded, and finally a precise cutting effect is achieved.

[0028] The method for intelligent control of automatic cutting of aluminum profiles is applied to an intelligent control device for automatic cutting of aluminum profiles, which can achieve the technical goal of dynamic adaptive control during the cutting process and achieve the technical effect of improving cutting quality and production efficiency.

[0029] Furthermore, the present application also includes:

[0030] The cutting-related features are extracted according to the historical aluminum profile cutting data; the historical aluminum profile cutting data is divided into batches based on the cutting requirements and aluminum profile differences to obtain multiple batches of aluminum profile cutting data; the first batch of aluminum profile cutting data is extracted from the multiple batches of aluminum profile cutting data; the first batch of aluminum profile cutting data is interacted with the cutting-related features to obtain first cutting-related feature parameters; the first batch of cutting yield is extracted from the first batch of aluminum profile cutting data; a first cutting model is constructed through the first cutting-related feature parameters and the first batch of cutting yield, and is added to the initial cutting model.

[0031] Specifically, by analyzing the historical aluminum profile cutting data, the features related to the cutting process are extracted, including cutting speed, tool pressure, material hardness, feed rate, etc. By analyzing these features, the key factors affecting the cutting quality and accuracy can be found, and these factors form the associated features of cutting.

[0032] Next, the historical data is divided into batches according to the different types of aluminum profiles and cutting requirements. Different batches of aluminum profiles may behave differently during the cutting process due to differences in production processes, material sources, environmental conditions, etc. Therefore, dividing the historical data into batches helps to accurately analyze the cutting effects under different conditions, so as to better adjust the control parameters in a targeted manner.

[0033] After data division, cutting data are randomly extracted from multiple batches of data as the first batch of aluminum profile cutting data, which provides a basis for subsequent analysis and model building.

[0034] Then, the cutting data of the first batch of aluminum profiles are subjected to parameter extraction according to the cutting-related features, and the cutting-related feature parameters of the first batch are obtained, thereby obtaining key parameters that can describe the cutting process.

[0035] Next, the corresponding cutting yields of multiple cutting production lines or multiple production rounds in the first batch of aluminum profile cutting data are extracted as the first batch cutting yield. The first batch cutting yield is an indicator used to measure the cutting quality of aluminum profiles, which generally refers to the proportion of all cut parts that meet the quality standards in multiple cutting production lines or multiple production rounds. The yield directly reflects the effectiveness of the cutting process.

[0036] The first cutting model is constructed by combining the first cutting-related characteristic parameters and the first batch yield rate. According to the method for generating the first cutting model, multiple batches of aluminum profile cutting data are traversed to generate multiple corresponding cutting models, which are combined to obtain the initial cutting model.

[0037] By obtaining the initial cutting model, not only can the efficiency of optimizing the cutting process be improved, but also accurate predictions and optimization suggestions can be provided for the subsequent cutting process.

[0038] Furthermore, the present application also includes:

[0039] A preset clustering benchmark is configured for the cutting requirements and aluminum profile differences; the initial cutting model is clustered and divided according to the preset clustering benchmark to obtain the multiple clustering models; a first clustering model is extracted according to the multiple clustering models; the first clustering cutting yield of the first clustering model is screened according to the cutting yield threshold as the screening benchmark to obtain a screened first clustering model and a filtered first clustering model; the filtered first clustering model is subjected to a control parameter homogeneity simulation tending to the screened first clustering model until the first clustering cutting yield of the filtered first clustering model meets the screening benchmark and is converted into the screened first clustering model; the screened first clustering model is reorganized to obtain a reorganized first clustering model, and is added to the reorganized clustering model.

[0040] Specifically, in the aluminum profile cutting process, different cutting requirements and differences in aluminum profiles will affect the cutting effect, so it is necessary to preset clustering benchmarks for these differences. The preset clustering benchmark is determined based on the cutting requirements (such as accuracy, cutting speed, etc.) and the differences in aluminum profiles (such as material hardness, thickness, etc.), and aluminum profiles and cutting conditions with similar characteristics are classified into a group. For example, for softer aluminum profiles, a lower cutting speed may be used, while for harder aluminum profiles, the cutting speed needs to be appropriately increased.

[0041] Next, based on these preset clustering benchmarks, the initial cutting model is clustered and divided to obtain multiple clustering models. The clustering process is to group the cutting data according to different characteristics. Each clustering model represents a specific cutting condition and result. For example, there is a clustering model suitable for low cutting speed for cutting soft aluminum profiles, and another clustering model is suitable for high cutting speed for hard aluminum profiles. Each clustering model contains a set of parameters that can optimize the cutting quality of a specific aluminum profile.

[0042] Then, from the multiple clustering models obtained, a clustering model is randomly extracted as the research object. The first clustering model represents the cutting process characteristics under a certain cutting requirement and aluminum profile difference. This model will have corresponding cutting accuracy and yield rate. By extracting the first clustering model, the cutting effect under this specific condition is analyzed.

[0043] Next, the first cluster cutting yield in the first cluster model is screened using the cutting yield threshold as the screening benchmark. The yield threshold is a set standard, indicating that the yield reaches a certain minimum value. For example, if the yield threshold is 95%, only cluster models with a cutting yield higher than 95% meet the requirements. Through this screening process, the first cluster model that meets the standard can be screened out, and the model with a yield lower than the threshold can be filtered out.

[0044] Afterwards, the first filtered cluster model is adjusted by means of control parameter homogenization simulation until its cutting yield meets the screening benchmark. Control parameter homogenization simulation means that under the same cutting conditions, the control parameters (such as cutting speed, feed rate, etc.) are adjusted to make the output result of the model close to the screening benchmark. For example, assuming that the yield of a cluster model is 80%, by adjusting parameters such as cutting speed and tool pressure, the yield is increased to 95%, so that it meets the requirements.

[0045] Finally, after all clustering models are optimized, the first clustering model that has been screened out and optimized is reorganized to obtain the reorganized first clustering model. Then, according to the method for obtaining the reorganized first clustering model, multiple clustering models are traversed to optimize parameters and combined to obtain the reorganized clustering model. The reorganized clustering model can better adapt to different cutting requirements and aluminum profile characteristics, and provide more accurate cutting predictions.

[0046] By continuously adjusting and optimizing different clustering models, we can eventually screen out the most suitable combination of control parameters based on specific cutting requirements and aluminum profile characteristics, and improve the yield rate in the cutting process, and then obtain a reorganized clustering model, which can obtain a more efficient and accurate cutting model, thereby improving the cutting quality and efficiency of production.

[0047] Furthermore, the present application also includes:

[0048] According to the cutting-related features, a yield correlation analysis is performed to obtain a feature yield correlation coefficient; according to the first cluster cutting yield of the screened first clustering model, a first control model is extracted from the screened first clustering model; according to the filtered first clustering model, a first adjustment model is extracted; the absolute value of the difference between the cutting-related feature parameters of the first adjustment model and the first control model is used as a first adjustment threshold; based on the cutting-related features, the cutting-related feature parameters of the first adjustment model are extracted to obtain the first feature parameters; with the feature yield correlation coefficient as an adjustment sequence, the first feature parameters are adjusted toward the first control model in sequence, and when the adjustment amplitude of the previous sequence in the adjustment sequence meets the adjustment threshold, the next adjustment sequence is executed until the first adjustment cutting yield of the first adjustment model meets the screening benchmark after any sequence is executed; according to the first adjustment cutting yield that meets the screening benchmark, the first adjustment model is converted, the filtered first clustering model is traversed, and added to the screened first clustering model.

[0049] Specifically, the yield rate correlation analysis is performed based on the cutting correlation characteristics. The relationship between different cutting parameters and cutting yield rate is analyzed, such as the impact of cutting speed, feed rate, tool pressure, etc. on yield rate, and then the correlation coefficient between each cutting parameter and yield rate is calculated to measure the degree of dependence.

[0050] Next, based on the cutting yield in the first clustering model, the cutting model in any batch or any round is randomly extracted to obtain the first control model. The first control model represents the standard cutting effect and yield under specific cutting conditions, and is used for the subsequent control of the parameter adjustment of the first clustering model.

[0051] Then, a first adjustment model is randomly extracted from the first cluster model that has been screened. Next, the absolute value of the difference between the cutting-related characteristic parameters of the first adjustment model and the first control model is used as the first adjustment threshold. The setting of the adjustment threshold is to control the variation range of the parameters during the model optimization process. For example, if the cutting speed of the first control model is fifty meters per minute, and the cutting speed of the first adjustment model is forty meters per minute, then the absolute value of the difference between the two, ten meters per minute, constitutes the adjustment threshold.

[0052] Based on the cutting-related features, the first adjustment model is subjected to cutting-related feature parameter extraction to obtain the first feature parameter. The first feature parameter refers to the cutting parameters that need to be paid attention to when adjusting the cutting control parameters of the first adjustment model, such as cutting speed, feed rate, etc., which are the basis for ensuring the effectiveness of the optimization process.

[0053] Next, the first characteristic parameters are adjusted in sequence with the characteristic yield correlation coefficient as the adjustment sequence, with the goal of moving these parameters toward the standard values ​​of the first control model. In the adjustment process of analog control, when the adjustment amplitude of the previous sequence meets the set adjustment threshold, by analyzing the current cutting yield, if the cutting yield still does not meet the set threshold, the next adjustment sequence will be executed to finally achieve the required cutting yield.

[0054] Finally, when the cutting yield of the first adjustment model meets the screening criteria, it is converted, and the first clustering model is traversed and filtered according to the converted cutting model, the control parameters are optimized, and the first clustering model is obtained and screened, ensuring that the obtained adjustment model can improve the final cutting effect through careful adjustment and optimization, so that the cutting quality meets the expected standards.

[0055] Through the updating of multiple clustering models, an efficient cutting model is finally formed which can be widely applied to different aluminum profiles and cutting requirements, ensuring that the cutting yield rate reaches the expected level, thereby improving production efficiency and cutting quality.

[0056] Furthermore, the present application also includes:

[0057] Based on the cutting requirements and the differences in aluminum profiles, the matching and recombination clustering model is used to perform similarity matching of the target aluminum profile to obtain a similarity result; based on the similarity result, a first matching model is determined; the target aluminum profile is input into the first matching model, and a first matching yield is output; it is determined whether the first matching yield meets a cutting yield threshold; if the first matching yield does not meet the cutting yield threshold, the target aluminum profile is input into the matching and recombination clustering model with the similarity results as a sequence until the matching yield meets the cutting yield threshold; if the first matching yield meets the cutting yield threshold, the target aluminum profile is automatically cut and controlled according to a first matching parameter of the first matching model to obtain a target cutting result.

[0058] Specifically, based on the cutting requirements and aluminum profile differences, the matched recombinant clustering models are matched for the target aluminum profile similarity, with the goal of calculating the similarity between the characteristics of the target aluminum profile and the cutting conditions of each clustering model. For example, if the target aluminum profile is thinner and has lower hardness, the most similar clustering model will be calculated based on these characteristics to ensure the rationality of the cutting parameters and avoid poor cutting. Then, based on the calculated similarity results, the first matching model is determined. The first matching model is the clustering model that is most similar to the target aluminum profile and can provide the most appropriate combination of cutting parameters, thereby maximizing the cutting quality.

[0059] Next, the target aluminum profile is input into the first matching model to obtain the first matching yield rate. The first matching yield rate refers to the proportion of aluminum profiles that meet the quality requirements after cutting under the model conditions. Then, it is determined whether the first matching yield rate meets the set cutting yield rate threshold. If the first matching yield rate does not reach the threshold, it means that the cutting effect is not ideal. There may be an ideal match, but there is actually a mismatch due to environmental factors. Therefore, according to the similarity results, the target aluminum profile will be input into the matching recombination clustering model, and the model will be matched step by step until a matching model that meets the cutting yield rate threshold is found.

[0060] If the first matching yield rate meets the cutting yield rate threshold, the cutting of the target aluminum profile can be automatically controlled according to the cutting parameters in the first matching model. At this time, the cutting operation is performed according to the optimized parameters to ensure that the quality of the target aluminum profile after cutting meets the standard.

[0061] Through iterative matching and optimization, the cutting yield rate can be continuously improved, and ultimately efficient and precise automated cutting control can be achieved.

[0062] Furthermore, the present application also includes:

[0063] The actual yield rate of the target cutting result is obtained by real-time monitoring by the visual monitoring unit; if the actual yield rate does not meet the actual yield threshold, a feedback relearning mechanism is triggered; an actual cutting model is obtained based on the feedback relearning mechanism, and automatic cutting control is performed according to the actual cutting model.

[0064] Specifically, the actual yield rate of the target cutting result is monitored in real time through the visual monitoring unit. The visual monitoring unit detects the cut aluminum profiles through cameras or other sensors, analyzes their surface quality, and calculates the proportion of good products that meet the standards in the cutting results.

[0065] Next, if the actual yield rate detected does not meet the preset yield rate threshold, the feedback re-learning mechanism will be triggered. The yield rate threshold refers to the preset minimum qualified standard in the cutting process. If the cutting result is lower than this standard, the re-learning mechanism will be triggered to improve the yield rate by adjusting certain parameters in the cutting process.

[0066] Once the feedback relearning mechanism is triggered, adjustments will be made based on the actual cutting model. The actual cutting model is a model generated based on the current cutting data and results, reflecting the performance of the cutting process under specific conditions. If the actual cutting result is lower than expected, the control parameters in the current model will be analyzed to identify factors that may cause substandard quality, such as excessive cutting speed and excessive tool wear, and adjustments will be made based on these issues. After adjustment, the new cutting model will be used for the next cutting operation.

[0067] Finally, the cutting will be automatically controlled according to the actual cutting model after adjustment, ensuring that a higher yield rate can be achieved in the subsequent cutting process and that the cutting quality is continuously optimized.

[0068] The visual monitoring system provides real-time feedback on the quality of the cutting results and can dynamically adjust the cutting parameters to ensure that the yield rate meets the set standards.

[0069] Furthermore, the present application also includes:

[0070] If the feedback re-learning mechanism is triggered, incremental learning is performed according to real-time failure samples, and the actual cutting model is obtained by adjustment.

[0071] Specifically, when the feedback re-learning mechanism is triggered, incremental learning is performed based on real-time failure samples. Real-time failure samples refer to those aluminum profile samples that fail to achieve the expected yield rate during the cutting process. For example, some cut aluminum profiles may have cracks, burrs, or do not meet the size requirements. These aluminum profiles are marked as failed samples. These failed samples are analyzed to identify the causes of the problems, such as excessive cutting speed, tool wear, uneven feed, etc. By performing incremental learning on failed samples, it is possible to continuously learn from new cutting data, thereby fine-tuning the original model so that the model can better adapt to different situations that occur during the actual cutting process.

[0072] In the incremental learning process, the actual cutting model obtained is adjusted. By analyzing the common features in the failed samples, some key parameters in the cutting process are adjusted, such as reducing the cutting speed, increasing the tool preheating time, adjusting the tool pressure, etc., to reduce the probability of cutting failure. The adjusted model can reflect the new cutting experience, thereby providing higher accuracy and better quality assurance in the next cutting operation.

[0073] Through the feedback re-learning mechanism, the actual cutting model is continuously optimized to improve the cutting quality and achieve a more precise and stable cutting effect.

[0074] In summary, the intelligent control method for automatic cutting of aluminum profiles provided in this application has the following technical effects:

[0075] The cutting-related feature parameters with cutting-related features as the control are extracted according to the historical aluminum profile cutting data, and the cutting yield is extracted, and an initial cutting model is constructed based on the cutting-related feature parameters and the cutting yield; the initial cutting model is clustered and divided according to the cutting requirements and aluminum profile differences, and a plurality of clustering models are simulated for homogenization of control parameters according to the cutting yield, and a recombinant clustering model is obtained by integration, and the recombinant clustering model is cached; the target aluminum profile is scanned based on the cutting requirements and aluminum profile differences, and the recombinant clustering model is searched to obtain a matching recombinant clustering model; the matching recombinant clustering model is input according to the target aluminum profile for automatic cutting control to obtain a target cutting result. That is to say, by realizing the technical goal of dynamic adaptive control in the cutting process, the technical effect of improving cutting quality and production efficiency is achieved.

[0076] Embodiment 2: Based on the same invention concept as the aluminum profile automatic cutting intelligent control method in the previous embodiment, this application also provides an aluminum profile automatic cutting intelligent control device, please refer to the attached Figure 2 ,include:

[0077] An initial cutting model construction module 11, the initial cutting model construction module 11 is used to extract cutting-related feature parameters with cutting-related features as a control according to historical aluminum profile cutting data, and extract the cutting yield, and construct an initial cutting model based on the cutting-related feature parameters and the cutting yield; a recombinant clustering model acquisition module 12, the recombinant clustering model acquisition module 12 is used to cluster the initial cutting model according to cutting requirements and aluminum profile differences, perform control parameter homogeneity simulation on multiple clustering models according to the cutting yield, integrate to obtain a recombinant clustering model, and cache the recombinant clustering model; a matching recombinant clustering model acquisition module 13, the matching recombinant clustering model acquisition module 13 is used to scan the target aluminum profile based on the cutting requirements and aluminum profile differences, and search for the recombinant clustering model to obtain a matching recombinant clustering model; a target cutting result acquisition module 14, the target cutting result acquisition module 14 is used to input the matching recombinant clustering model according to the target aluminum profile for automatic cutting control to obtain a target cutting result.

[0078] Furthermore, the intelligent control device for automatic cutting of aluminum profiles is also used for:

[0079] The cutting-related features are extracted according to the historical aluminum profile cutting data; the historical aluminum profile cutting data is divided into batches based on the cutting requirements and aluminum profile differences to obtain multiple batches of aluminum profile cutting data; the first batch of aluminum profile cutting data is extracted from the multiple batches of aluminum profile cutting data; the first batch of aluminum profile cutting data is interacted with the cutting-related features to obtain first cutting-related feature parameters; the first batch of cutting yield is extracted from the first batch of aluminum profile cutting data; a first cutting model is constructed through the first cutting-related feature parameters and the first batch of cutting yield, and is added to the initial cutting model.

[0080] Furthermore, the intelligent control device for automatic cutting of aluminum profiles is also used for:

[0081] A preset clustering benchmark is configured for the cutting requirements and aluminum profile differences; the initial cutting model is clustered and divided according to the preset clustering benchmark to obtain the multiple clustering models; a first clustering model is extracted according to the multiple clustering models; the first clustering cutting yield of the first clustering model is screened according to the cutting yield threshold as the screening benchmark to obtain a screened first clustering model and a filtered first clustering model; the filtered first clustering model is subjected to a control parameter homogeneity simulation tending to the screened first clustering model until the first clustering cutting yield of the filtered first clustering model meets the screening benchmark and is converted into the screened first clustering model; the screened first clustering model is reorganized to obtain a reorganized first clustering model, and is added to the reorganized clustering model.

[0082] Furthermore, the intelligent control device for automatic cutting of aluminum profiles is also used for:

[0083] According to the cutting-related features, a yield correlation analysis is performed to obtain a feature yield correlation coefficient; according to the first cluster cutting yield of the screened first clustering model, a first control model is extracted from the screened first clustering model; according to the filtered first clustering model, a first adjustment model is extracted; the absolute value of the difference between the cutting-related feature parameters of the first adjustment model and the first control model is used as a first adjustment threshold; based on the cutting-related features, the cutting-related feature parameters of the first adjustment model are extracted to obtain first feature parameters; with the feature yield correlation coefficient as an adjustment sequence, the first feature parameters are adjusted toward the first control model in sequence, and when the adjustment amplitude of the previous sequence in the adjustment sequence meets the adjustment threshold, a subsequent adjustment sequence is executed until the first adjustment cutting yield of the first adjustment model meets the screening benchmark after any sequence is executed; according to the first adjustment cutting yield that meets the screening benchmark, the first adjustment model is converted, the filtered first clustering model is traversed, and added to the screened first clustering model.

[0084] Furthermore, the intelligent control device for automatic cutting of aluminum profiles is also used for:

[0085] Based on the cutting requirements and the differences in aluminum profiles, the matching and recombination clustering model is used to perform similarity matching of the target aluminum profile to obtain a similarity result; based on the similarity result, a first matching model is determined; the target aluminum profile is input into the first matching model, and a first matching yield is output; it is determined whether the first matching yield meets a cutting yield threshold; if the first matching yield does not meet the cutting yield threshold, the target aluminum profile is input into the matching and recombination clustering model with the similarity results as a sequence until the matching yield meets the cutting yield threshold; if the first matching yield meets the cutting yield threshold, the target aluminum profile is automatically cut and controlled according to a first matching parameter of the first matching model to obtain a target cutting result.

[0086] Furthermore, the intelligent control device for automatic cutting of aluminum profiles is also used for:

[0087] The actual yield rate of the target cutting result is obtained by real-time monitoring by the visual monitoring unit; if the actual yield rate does not meet the actual yield threshold, a feedback relearning mechanism is triggered; an actual cutting model is obtained based on the feedback relearning mechanism, and automatic cutting control is performed according to the actual cutting model.

[0088] Furthermore, the intelligent control device for automatic cutting of aluminum profiles is also used for:

[0089] If the feedback re-learning mechanism is triggered, incremental learning is performed according to real-time failure samples, and the actual cutting model is obtained by adjustment.

[0090] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The intelligent control method for automated cutting of aluminum profiles and the specific examples in the aforementioned embodiment one are also applicable to an intelligent control device for automated cutting of aluminum profiles in this embodiment. Through the aforementioned detailed description of the intelligent control method for automated cutting of aluminum profiles, those skilled in the art can clearly understand the intelligent control device for automated cutting of aluminum profiles in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0091] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0092] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. An intelligent control method for automatic cutting of aluminum profiles, characterized in that: include: Extracting cutting-related feature parameters with cutting-related features as reference according to historical aluminum profile cutting data, extracting cutting yield, and constructing an initial cutting model based on the cutting-related feature parameters and the cutting yield; The initial cutting model is clustered and divided according to the cutting requirements and the difference of aluminum profiles, a control parameter homogenization simulation is performed on multiple clustering models according to the cutting yield, and a recombined clustering model is obtained by integration, and the recombined clustering model is cached; Scanning the target aluminum profile based on the cutting requirements and the aluminum profile differences, and searching for the recombinant clustering model to obtain a matching recombinant clustering model; The matching reorganization clustering model is input according to the target aluminum profile to perform automatic cutting control to obtain the target cutting result.

2. The intelligent control method for automatic cutting of aluminum profiles according to claim 1, characterized in that: The method extracts cutting-related feature parameters with cutting-related features as reference according to historical aluminum profile cutting data, extracts cutting yield, and constructs an initial cutting model based on the cutting-related feature parameters and the cutting yield, including: Extracting the cutting-related features according to the historical aluminum profile cutting data; Dividing the historical aluminum profile cutting data into batches based on the cutting requirements and aluminum profile differences to obtain multiple batches of aluminum profile cutting data; Extracting the first batch of aluminum profile cutting data from the multiple batches of aluminum profile cutting data; Interacting the first batch of aluminum profile cutting data with the cutting-related feature to obtain first cutting-related feature parameters; Extracting the first batch cutting yield rate from the first batch of aluminum profile cutting data; A first cutting model is constructed by using the first cutting-associated characteristic parameters and the first batch cutting yield, and is added to the initial cutting model.

3. The intelligent control method for automatic cutting of aluminum profiles according to claim 1, characterized in that: The initial cutting model is clustered and divided according to the cutting requirements and the difference of aluminum profiles, and the control parameter homogenization simulation of multiple clustering models is performed according to the cutting yield, and the recombinant clustering model is obtained by integration, including: Configuring a preset clustering benchmark for the cutting requirements and aluminum profile differences; Clustering the initial cutting model based on the preset clustering benchmark to obtain the multiple clustering models; Extracting a first clustering model according to the multiple clustering models; Taking the cutting yield rate threshold as a screening benchmark, screening the first cluster cutting yield rate of the first cluster model to obtain a screened first cluster model and a filtered first cluster model; The first filtering cluster model is subjected to a control parameter homogenization simulation tending to the first screening cluster model, until the first cluster cutting yield of the first filtering cluster model meets the screening benchmark, and then converted into the first screening cluster model; The screened first clustering model is recombined to obtain a recombined first clustering model, and added to the recombined clustering model.

4. The intelligent control method for automatic cutting of aluminum profiles according to claim 3, characterized in that: The step of subjecting the first filtering cluster model to a control parameter homogenization simulation tending to the first screening cluster model until the first cluster cutting yield of the first filtering cluster model meets the screening benchmark and converting the model to the first screening cluster model includes: Performing a yield rate correlation analysis according to the cutting correlation feature to obtain a feature yield rate correlation coefficient; Extracting a first control model from the screened first cluster model according to the first cluster cutting yield of the screened first cluster model; extracting a first adjustment model according to the filtering of the first clustering model; The absolute value of the difference between the cutting-related characteristic parameters of the first adjustment model and the first control model is used as a first adjustment threshold; Extracting the cutting-related feature parameters from the first adjustment model based on the cutting-related features to obtain first feature parameters; Taking the characteristic yield rate correlation coefficient as the adjustment sequence, the first characteristic parameter is adjusted toward the parameter of the first control model in sequence, and when the adjustment amplitude of the previous sequence in the adjustment sequence meets the adjustment threshold, the next adjustment sequence is executed, until the first adjustment cutting yield of the first adjustment model meets the screening benchmark after any sequence is executed; The first adjustment model that satisfies the screening benchmark according to the first adjustment cutting yield is converted, traversed through the filtered first clustering model, and added to the screened first clustering model.

5. The intelligent control method for automatic cutting of aluminum profiles according to claim 1, characterized in that: The step of inputting the matching reorganization clustering model according to the target aluminum profile to automatically control the cutting to obtain the target cutting result includes: Based on the cutting requirements and the aluminum profile differences, the matching reorganization clustering model is used to perform similarity matching of the target aluminum profile to obtain a similarity result; Determine a first matching model based on the similarity result; Inputting the target aluminum profile into the first matching model, and outputting a first matching yield rate; Determining whether the first matching yield rate meets a cutting yield rate threshold; If the first matching yield rate does not meet the cutting yield rate threshold, the target aluminum profile is input into the matching recombination clustering model with the similarity results as a sequence until the matching yield rate meets the cutting yield rate threshold; If the first matching yield rate meets the cutting yield rate threshold, the cutting of the target aluminum profile is automatically controlled according to the first matching parameter of the first matching model to obtain a target cutting result.

6. The intelligent control method for automatic cutting of aluminum profiles according to claim 1, characterized in that: Also includes: Obtaining the actual yield rate of the target cutting result through real-time monitoring by a visual monitoring unit; If the actual yield rate does not meet the actual yield threshold, a feedback re-learning mechanism is triggered; An actual cutting model is obtained based on the feedback relearning mechanism, and automatic cutting control is performed according to the actual cutting model.

7. The intelligent control method for automatic cutting of aluminum profiles according to claim 6, characterized in that: If the feedback re-learning mechanism is triggered, incremental learning is performed according to real-time failure samples, and the actual cutting model is obtained by adjustment.

8. An intelligent control device for automatic cutting of aluminum profiles, characterized in that: The steps for implementing the intelligent control method for automatic cutting of aluminum profiles as described in any one of claims 1 to 7 include: An initial cutting model building module, the initial cutting model building module is used to extract cutting-related feature parameters with cutting-related features as a control according to historical aluminum profile cutting data, extract cutting yield, and build an initial cutting model based on the cutting-related feature parameters and the cutting yield; A recombinant clustering model acquisition module, the recombinant clustering model acquisition module is used to cluster the initial cutting model according to the cutting requirements and the aluminum profile differences, perform control parameter homogenization simulation on multiple clustering models according to the cutting yield, integrate and obtain a recombinant clustering model, and cache the recombinant clustering model; A matching recombinant clustering model acquisition module, wherein the matching recombinant clustering model acquisition module is used to scan the target aluminum profile based on the cutting requirements and the aluminum profile differences, and to search for the recombinant clustering model to obtain a matching recombinant clustering model; The target cutting result acquisition module is used to input the matching reorganization clustering model according to the target aluminum profile to perform automatic cutting control to obtain the target cutting result.

Images